U.S.–China Tension and Corporate Capital Structure in ASEAN Countries
Lộ trình 10 bước thực hiện bài nghiên cứu: từ phương pháp và kiểm định đến diễn giải kết quả, hoàn thiện bản thảo và nộp bài.
BƯỚC 1. XÂY DỰNG PHƯƠNG PHÁP NGHIÊN CỨU
Bài mẫu: U.S.–China Tension and Corporate Capital Structure in ASEAN Countries
Mục tiêu đầu ra: Hoàn thành bản nháp có căn cứ tài liệu tham khảo cho phần Data and Methodology của paper.
Quy ước của bài mẫu
- Biến độc lập chính: U.S.–China Tension Index (UCT) của Rogers, Sun, and Sun (2024).
- UCT là một news-based aggregate time-series index, không phải country-specific index.
- Dữ liệu UCT được lấy từ Economic Policy Uncertainty / PolicyUncertainty.com.
- Research universe: các quốc gia thành viên ASEAN.
- Tính đến năm 2026, ASEAN có 11 Member States:
- Brunei Darussalam;
- Cambodia;
- Indonesia;
- Lao PDR;
- Malaysia;
- Myanmar;
- Philippines;
- Singapore;
- Thailand;
- Timor-Leste;
- Vietnam.
- Timor-Leste trở thành thành viên ASEAN thứ 11 ngày 26 October 2025.
- Dữ liệu tài chính doanh nghiệp: Compustat Global và các nguồn firm-level bổ sung nếu cần.
- Đơn vị quan sát: firm-year.
- Final empirical sample phải dựa trên actual data availability, không ép đủ 11 countries nếu Compustat không có đủ firm-year observations.
- Nếu final sample chỉ gồm Indonesia, Malaysia, Philippines, Singapore, Thailand và Vietnam, phải gọi đúng là:
ASEAN-6 sample
chứ không viết “all ASEAN countries”.
- Financial firms và utilities bị loại.
- Require:
và:
- Continuous dependent variables và continuous firm-level controls được winsorize ở 1st–99th percentiles.
- UCT và macro variables không winsorize trong baseline.
- Main Capital Structure measure:
Book Leverage
- Alternative outcomes:
- Short-Term Leverage;
- Long-Term Leverage;
- Debt Maturity;
- Market Leverage.
- Nếu code nghiên cứu hiện tại sử dụng các biến tên như
TDB,TDM,LDB,LDM, sinh viên phải map từng tên code vào exact economic definition trước khi viết paper.- Baseline timing ưu tiên:
- Critical identification issue: nếu UCT annual chỉ có một giá trị chung cho toàn bộ firms trong từng năm, Year Fixed Effects sẽ hấp thụ main UCT hoàn toàn.
- Vì vậy paper phải chọn rõ một trong ba designs:
Design A. Annual UCT direct-effect model without Year FE
Design B. Fiscal-year-aligned UCT từ monthly index
Design C. Exposure-based model: UCT × ASEAN firm/country exposure
- Trong một paper mạnh, Design C nên được ưu tiên nếu có credible exposure data.
PHẦN 0. BỘ PAPER BẮT BUỘC VÀ CÁCH DÙNG ĐỂ VIẾT NGAY
0.1. Bộ tài liệu đã được cung cấp
Nhóm không bắt đầu bằng việc tìm literature từ số 0. Trước hết phải đọc và khai thác trực tiếp bộ paper đã được cung cấp.
Trong thư mục hiện có các paper sau:
Peng, Deng, Xie, and Liu (2025)
US-China Tension and Stock Market Performance in US and China: New Insights from Time-varying Quantile Causality Method
Finance Research Letters, 85, 107888.
DOI: 10.1016/j.frl.2025.107888Chai and Wang (2026)
China–US Relations and Overseas Capital Allocation of Chinese Firms
Finance Research Letters, 93, 109621.
DOI: 10.1016/j.frl.2026.109621Wang, Wang, and Johnson
Geography and Capital Structure
Canadian Journal of Administrative Sciences, 35(1), 107–122.
DOI: 10.1002/cjas.1383
Lưu ý: paper được first published online năm 2016 nhưng nằm trong volume 35(1) xuất bản năm 2018. Khi làm reference list phải kiểm tra style của journal và dùng metadata của bản chính thức.Jimoh, Goh, and Tham (2026)
Between Giants: The US-China Trade War and ASEAN’s Evolving Role in Global Trade
International Economics and Economic Policy, 23(2), 1–24.
DOI: 10.1007/s10368-026-00730-xVega-Gutiérrez, López-Iturriaga, and Rodríguez-Sanz (2025)
Economic Policy Uncertainty and Capital Structure in Europe: An Agency Approach
The European Journal of Finance, 31(1), 53–75.
DOI: 10.1080/1351847X.2024.2364044Chowdhury, Aram, Javadi, and Nejadmalayeri (2025)
Geopolitical Risk and Corporate Capital Structure
Journal of Corporate Finance, 93, 102796.
DOI: 10.1016/j.jcorpfin.2025.102796
Lưu ý về file: hai PDF Chai–Wang trong thư mục có vẻ là hai bản của cùng một paper. Không tính hai lần trong literature review.
Ngoài sáu paper trên, Rogers, Sun, and Sun (2024) vẫn là paper bắt buộc để định nghĩa và mô tả UCT nếu nghiên cứu sử dụng U.S.–China Tension Index của PolicyUncertainty.com.
0.2. Thứ tự đọc bắt buộc
Sinh viên đọc theo thứ tự sau:
1. Chowdhury et al. (2025)
→ học cách một macro geopolitical-risk index được đưa vào capital-structure regression.
2. Jimoh et al. (2026)
→ học vì sao ASEAN không đồng nhất và cách dùng pre-tension trade dependence/exposure.
3. Vega-Gutiérrez et al. (2025)
→ học uncertainty–capital structure literature và competing theoretical channels.
4. Peng et al. (2025)
→ học UCT của Rogers et al. và bằng chứng UCT truyền sang financial markets theo cách time-varying/asymmetric.
5. Chai and Wang (2026)
→ học micro-level corporate response, technological blockade và supply-chain reallocation.
6. Wang et al.
→ học information asymmetry và vai trò của geography trong capital-structure decisions.
Không đọc sáu paper theo kiểu “đọc abstract rồi thôi”. Mỗi paper phải được dùng để tạo ra một phần cụ thể của manuscript.
PHẦN 0A. PAPER-BY-PAPER EXTRACTION
0.3. Paper 1 — Chowdhury et al. (2025): PAPER MẪU QUAN TRỌNG NHẤT
Paper này dùng để làm gì?
Đây là model paper chính cho phần:
- capital-structure measurement;
- timing;
- baseline regression;
- macro uncertainty → leverage logic;
- mechanism;
- heterogeneity;
- robustness;
- endogeneity.
Sinh viên phải đọc kỹ
- Introduction.
- Capital-structure measurement.
- GPR measurement.
- Baseline equation.
- Baseline controls.
- Endogeneity section.
- Demand and supply channels.
- Cross-sectional/subsample tests.
- Appendix variable definitions.
Phải trích ra ít nhất các thông tin sau
| Nội dung | Chowdhury et al. (2025) | Bài ASEAN của chúng ta |
|---|---|---|
| Main X | GPR | UCT |
| X variation | Time series | Time series nếu annual UCT |
| Main DV | TDM, LDM | Book/market leverage + maturity |
| Robustness DVs | TDB, LDB | TDB/LDB/ST debt/maturity |
| Timing | DV at t+1 | DV at t+1 |
| Firm FE | Yes | Yes |
| Year FE | Không dùng cho main GPR | Không dùng cho annual UCT direct effect |
| Main firm controls | industry leverage, tangibility, profitability, size, MTB | Verify/adapt |
| Macro control | inflation + richer macro set | GDP growth, inflation, global controls |
| Winsorization | 1%–99% | 1%–99% DV + firm controls |
| Mechanisms | debt demand + debt supply | financing demand + credit supply |
| Heterogeneity | existing leverage, irreversibility, GPR sensitivity | leverage, constraints, trade exposure |
Chowdhury et al. use total and long-term market leverage as main outcomes and book leverage as robustness measures. They also use one-year forward leverage, explicitly because the geopolitical-risk measure is known at time $t$ before the capital-structure outcome at $t+1$.
Điểm cực kỳ quan trọng phải học từ paper này
Chowdhury et al. recognize that their GPR variable has only time-series variation. Consequently, they do not include ordinary year fixed effects in the main GPR specification.
Đây chính là logic phải áp dụng cho annual UCT.
Sinh viên được phép viết gần theo cấu trúc này
Following recent research on geopolitical risk and corporate financing, we relate the aggregate geopolitical environment in year $t$ to firms' leverage choices in year $t+1$. Chowdhury et al. (2025) document a persistent negative relationship between geopolitical risk and corporate leverage and show that the effect operates through both debt-demand and credit-supply channels. Their design also highlights an important identification issue for aggregate geopolitical indices: when the main explanatory variable varies only over time, conventional year fixed effects absorb the variable of interest.
Không được copy
- coefficient;
- sample size;
- GPR-specific wording;
- U.S.-specific institutional setting.
Phải chuyển logic từ:
sang:
0.4. Paper 2 — Jimoh et al. (2026): PAPER ASEAN QUAN TRỌNG NHẤT
Paper này dùng để làm gì?
Dùng trực tiếp cho:
- lý do chọn ASEAN;
- ASEAN heterogeneity;
- trade-dependence channel;
- exposure-based design;
- country-level heterogeneity;
- contribution.
Key evidence cần lấy
Jimoh et al. examine ASEAN export responses to U.S.–China trade tensions from 1993–2024 and show that effects are not homogeneous across ASEAN countries.
Họ sử dụng:
- Trade Tension Index;
- pre-tension export dependence;
- interaction between tension and pre-tension dependence;
- country-specific analysis;
- Driscoll–Kraay inference.
Results show asymmetric responses between exports to:
- U.S.;
- China;
and substantial heterogeneity among ASEAN members.
Phần này phải đi thẳng vào Introduction
Sinh viên có thể viết:
ASEAN is not a homogeneous recipient of U.S.–China tensions. Jimoh et al. (2026) show that ASEAN members respond asymmetrically to bilateral trade tensions because their export structures and pre-existing dependence on the U.S. and China differ substantially. This heterogeneity provides a natural setting for examining whether the financing responses of ASEAN firms also depend on their economic exposure to the two major economies.
Phần này phải đi vào empirical design
Jimoh et al. provide direct justification for:
Sinh viên phải tạo ít nhất hai exposure measures
Nếu data cho phép:
and:
rather than only:
Tại sao?
Jimoh et al. show that dependence on the U.S. and dependence on China can create different responses.
Writing output
Sau khi đọc paper này, sinh viên phải viết được một đoạn 150–200 từ mang tiêu đề:
Why ASEAN?
và một đoạn 100–150 từ mang tiêu đề:
Why exposure heterogeneity?
0.5. Paper 3 — Vega-Gutiérrez et al. (2025): UNCERTAINTY–CAPITAL STRUCTURE
Paper này dùng để làm gì?
Dùng cho:
- literature review;
- agency explanation;
- institutional moderation;
- financial-development moderation;
- competing prediction về dấu của coefficient.
Important result
Vega-Gutiérrez et al. study 3,175 firms from 11 European countries and find a positive relationship between EPU and financial leverage.
They also show moderation by:
- uncertainty avoidance;
- institutional quality;
- financial development;
- corporate diversification.
Vì sao paper này rất quan trọng dù dấu khác Chowdhury?
Vì nó chứng minh:
uncertainty không bắt buộc luôn làm leverage giảm.
Do đó phần theory của bài ASEAN phải trình bày competing channels, không được viết theo kiểu “uncertainty chắc chắn làm debt giảm”.
Sinh viên phải viết một đoạn như sau
The theoretical effect of uncertainty on capital structure is not unambiguous. One strand of research argues that heightened uncertainty raises expected distress costs, weakens credit supply, and encourages firms to preserve financial flexibility, predicting lower leverage. Consistent with this view, Chowdhury et al. (2025) document a negative effect of geopolitical risk on leverage. In contrast, Vega-Gutiérrez et al. (2025) report a positive association between economic policy uncertainty and leverage in European firms and interpret debt as one mechanism through which firms respond to agency and informational frictions. Taken together, these findings suggest that the leverage response to uncertainty is likely to depend on the type of uncertainty and the institutional and financial environment.
Heterogeneity lấy từ paper này
Chuẩn bị:
and/or:
cho later analysis.
Không nhất thiết chạy ngay ở baseline.
0.6. Paper 4 — Peng et al. (2025): VALIDATION OF THE UCT CONCEPT
Paper này dùng để làm gì?
Không dùng làm capital-structure model paper.
Dùng cho:
- Introduction;
- relevance of UCT;
- financial-market transmission;
- time variation;
- asymmetry;
- robustness motivation.
Critical point
Peng et al. explicitly use the U.S.–China Tension Index constructed by:
Rogers et al. (2024).
They find that UCT has:
- significant financial-market effects;
- time-varying effects;
- asymmetric effects;
- stronger response in Chinese stock markets than U.S. markets.
Sinh viên phải dùng paper này để viết
Recent evidence indicates that U.S.–China tensions are not confined to trade flows or diplomatic relations but are transmitted to financial markets. Using the UCT measure of Rogers et al. (2024), Peng et al. (2025) document significant, time-varying, and asymmetric stock-market responses to changes in U.S.–China tensions. This evidence motivates examining whether the same bilateral tension also affects slower-moving corporate financial policies such as leverage and debt maturity.
Robustness ideas lấy từ Peng
Because UCT effects are time-varying/asymmetric, later analysis may consider:
- high-UCT vs low-UCT periods;
- post-2018 period;
- COVID vs non-COVID;
- positive changes vs negative changes in UCT;
- nonlinear specifications.
Không cần đưa quantile Granger causality vào firm-level capital-structure regression chỉ vì Peng dùng phương pháp đó.
0.7. Paper 5 — Chai and Wang (2026): MICRO CORPORATE RESPONSE
Paper này dùng để làm gì?
Dùng cho:
- corporate transmission channel;
- technology blockade;
- supply-chain security;
- strategic reallocation;
- heterogeneity;
- sample-treatment precedent.
Key evidence
Chai and Wang use Chinese A-share firms from 2007–2024.
They:
- exclude financial firms;
- winsorize continuous variables;
- examine corporate responses to deteriorating China–U.S. relations;
- find greater overseas capital allocation;
- identify technological-blockade pressure as a mechanism;
- find stronger effects for non-SOEs and firms with executives possessing overseas experience.
Extremely important warning
Chai and Wang do NOT use the Rogers et al. UCT index.
Their tension measure is based on a China–U.S. relations database from:
the Institute of International Relations at Tsinghua University.
Therefore:
Được viết
Chai and Wang (2026) provide firm-level evidence that deteriorating China–U.S. relations affect corporate strategic decisions through technological restrictions and supply-chain considerations.
Không được viết
“Following Chai and Wang (2026), we measure UCT using the Rogers index.”
Điều đó sai.
Cách dùng cho ASEAN
Convert the China-firm mechanism into ASEAN logic:
but also:
The second channel is particularly relevant in ASEAN.
0.8. Paper 6 — Wang, Wang, and Johnson: INFORMATION ASYMMETRY AND GEOGRAPHY
Paper này dùng để làm gì?
Không phải uncertainty paper.
Dùng để support:
- capital structure depends on location/information environment;
- information asymmetry affects financing;
- cross-country ASEAN setting should not be treated as homogeneous;
- geography/information environment may matter for debt choice.
Main result
Wang et al. find that firm geographic location affects capital structure and that the effect is stronger when information asymmetry is higher.
Cách dùng trong paper ASEAN
Không cần dành cả một subsection riêng.
Use it in one of these places:
Theory
Capital-structure choices depend not only on firm fundamentals but also on the information environment in which firms operate. Wang et al. show that geographic location affects leverage and that this relation becomes stronger under greater information asymmetry.
ASEAN motivation
Cross-country differences in information environments and financing access may therefore amplify heterogeneous financing responses across ASEAN firms.
Không được dùng paper này để claim
- UCT causes leverage;
- ASEAN firms behave the same as remote U.S./Canadian firms;
- geography itself is the main mechanism of the current paper.
PHẦN 0B. PAPER ROLE MATRIX
0.9. Bảng bắt buộc phải hoàn thành trước khi viết
| Paper | Dùng cho Section nào? | Main finding cần nhớ | Method/variable học được | Không được dùng sai |
|---|---|---|---|---|
| Peng et al. (2025) | Introduction; robustness motivation | UCT has time-varying/asymmetric financial effects | Rogers UCT usage | Không dùng làm leverage model |
| Chai & Wang (2026) | Introduction; mechanisms | Tension changes corporate strategic allocation | Exclusion/winsorization; mechanism logic | Tension index khác Rogers |
| Wang et al. | Theory; regional heterogeneity | Geography/information asymmetry affect leverage | Capital-structure context | Không phải uncertainty paper |
| Jimoh et al. (2026) | ASEAN motivation; exposure design | ASEAN responses are asymmetric | Pre-tension dependence interaction | Không giả định ASEAN homogeneous |
| Vega-Gutiérrez et al. (2025) | Theory; heterogeneity | EPU positively related to leverage | IQ/financial-development moderators | Không ép uncertainty sign = negative |
| Chowdhury et al. (2025) | Main model reference | GPR reduces leverage | Forward leverage; FE; controls; mechanisms | GPR ≠ UCT |
| Rogers et al. (2024) | Variable measurement | Constructs UCT | Exact UCT definition/data | Bắt buộc cho main X |
PHẦN 0C. TỪ 6 PAPER ĐẾN CÁC ĐOẠN VĂN CỤ THỂ
0.10. Đoạn 1 — Mở đầu về U.S.–China tension
Sources
- Rogers et al. (2024);
- Peng et al. (2025);
- Chai and Wang (2026).
Nội dung phải có
- U.S.–China tensions affect more than bilateral trade.
- They affect financial markets and corporate strategic decisions.
- This motivates examining financing policy.
Draft mẫu
U.S.–China relations have become an important source of geopolitical and economic uncertainty for firms and financial markets. Recent studies show that changes in bilateral tensions affect asset prices and corporate strategic behavior. Using the news-based U.S.–China Tension Index developed by Rogers et al. (2024), Peng et al. (2025) document significant and time-varying stock-market responses to bilateral tensions. At the corporate level, Chai and Wang (2026) show that deteriorating China–U.S. relations alter firms' overseas capital allocation through technological-blockade and supply-chain channels. These findings suggest that U.S.–China tensions can affect not only operating and investment decisions but also firms' financing policies.
0.11. Đoạn 2 — Vì sao ASEAN?
Main source
Jimoh et al. (2026)
Supporting sources
ASEAN/IMF literature nếu được cung cấp.
Draft mẫu
ASEAN offers a particularly informative setting for examining the financing consequences of U.S.–China tensions because its member economies are deeply connected to both major powers through trade, foreign direct investment, and regional production networks. Importantly, these connections are highly heterogeneous. Jimoh et al. (2026) show that the effects of U.S.–China trade tensions on ASEAN exports differ substantially across countries and depend on pre-existing export exposure to the United States and China. This heterogeneity implies that a common bilateral tension shock can generate different cash-flow, growth-opportunity, and financing effects across ASEAN firms.
0.12. Đoạn 3 — Uncertainty và Capital Structure
Sources
- Chowdhury et al. (2025);
- Vega-Gutiérrez et al. (2025);
- Wang et al.;
- foundational capital-structure literature.
Draft mẫu
The effect of external uncertainty on capital structure is theoretically ambiguous. Higher geopolitical risk can increase expected distress costs, reduce debt supply, and increase the value of financial flexibility, leading firms to reduce leverage. Consistent with this mechanism, Chowdhury et al. (2025) find that geopolitical risk has a persistent negative impact on corporate leverage and provide evidence of contractions in both debt demand and credit supply. However, uncertainty may also increase firms' reliance on debt under some institutional conditions. Vega-Gutiérrez et al. (2025), for example, document a positive relation between economic policy uncertainty and leverage in European firms and show that institutional quality and financial development moderate this relationship. These contrasting findings indicate that the financing response to uncertainty depends on the nature of the shock and the surrounding information and institutional environment.
0.13. Đoạn 4 — H1
Directional hypothesis recommended for this paper
Given:
- Chowdhury et al. (2025);
- closest UCT–leverage evidence if available;
- financial flexibility and credit-supply channel;
baseline hypothesis:
H1: Higher U.S.–China tension is associated with lower corporate leverage among ASEAN firms.
Paragraph mẫu
We expect the net effect of U.S.–China tensions on ASEAN firms' leverage to be negative. Heightened tensions increase uncertainty about trade conditions, input availability, export demand, and external financing, thereby raising expected distress costs and the value of preserving financial flexibility. They may also induce lenders to tighten credit supply. These mechanisms are consistent with the negative geopolitical-risk–leverage relation documented by Chowdhury et al. (2025). Accordingly, we hypothesize that higher U.S.–China tension reduces firms' reliance on debt financing.
Important
Because Vega-Gutiérrez et al. find the opposite sign under EPU:
the paper must acknowledge competing evidence before stating H1.
0.14. Đoạn 5 — Main X measurement
Sources
- Rogers et al. (2024);
- Peng et al. (2025) as usage precedent.
Draft mẫu
Our main explanatory variable is the U.S.–China Tension Index developed by Rogers et al. (2024). The index is constructed from newspaper coverage of bilateral U.S.–China tensions and is designed to capture the intensity of contentious relations between the two countries. Higher values indicate greater tension. Peng et al. (2025) provide a recent application of this index and document that its financial-market effects are both time-varying and asymmetric.
Do not cite Chai and Wang as the source of this variable.
0.15. Đoạn 6 — Capital Structure measurement
Main source
Chowdhury et al. (2025)
Supporting
- Frank and Goyal (2009);
- Rajan and Zingales (1995).
Draft mẫu
We examine multiple dimensions of corporate capital structure. Our primary measure is total book leverage, defined as interest-bearing short-term and long-term debt scaled by total assets. We separately examine short-term and long-term leverage to determine whether U.S.–China tensions affect refinancing-sensitive and longer-horizon debt differently. We also consider market leverage and debt maturity as alternative outcomes. This multi-measure approach follows the broader capital-structure literature and is particularly consistent with Chowdhury et al. (2025), who examine total and long-term leverage using both market- and book-based measures.
0.16. Đoạn 7 — Timing
Main source
Chowdhury et al. (2025)
Draft mẫu
We relate U.S.–China tension in year $t$ to capital structure in year $t+1$. The forward specification places the tension measure in the information set preceding the financing outcome and reduces concerns that contemporaneous leverage mechanically drives the measured relationship. Chowdhury et al. (2025) use a similar forward-leverage structure in their analysis of geopolitical risk and corporate capital structure.
0.17. Đoạn 8 — Fixed Effects
Main source
Chowdhury et al. (2025)
Draft mẫu
The baseline annual-index specification includes firm fixed effects to absorb time-invariant firm characteristics. However, annual UCT is a common time-series variable and therefore cannot be separately identified from a full set of year fixed effects. This issue is analogous to the treatment of aggregate geopolitical risk in Chowdhury et al. (2025), who note that their GPR measure has only time-series variation and consequently do not include year fixed effects in the main specification.
This paragraph is mandatory.
0.18. Đoạn 9 — ASEAN Exposure Design
Main source
Jimoh et al. (2026)
Draft mẫu
Because ASEAN members differ substantially in their pre-existing economic dependence on the United States and China, we complement the aggregate UCT specification with exposure-based tests. Following the logic of Jimoh et al. (2026), we construct predetermined measures of U.S. and China trade exposure and interact them with UCT. These specifications test whether the same bilateral tension shock produces different financing responses across firms located in economies with different structural exposure to the two major markets.
PHẦN 0D. SINH VIÊN PHẢI TẠO PAPER EVIDENCE MATRIX
0.19. Không được đọc paper mà không ghi lại bằng chứng
Tạo bảng:
| Paper | Exact page/section | Sample | X | Y | Main result | Controls | FE | VCE | Robustness | Mechanism | Heterogeneity | Sentence to cite |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Peng et al. | ||||||||||||
| Chai & Wang | ||||||||||||
| Wang et al. | ||||||||||||
| Jimoh et al. | ||||||||||||
| Vega-Gutiérrez et al. | ||||||||||||
| Chowdhury et al. |
Rule
Mỗi dòng phải có:
- page/section;
- exact evidence;
- manuscript use.
Không được ghi:
“Paper này liên quan.”
Phải ghi cụ thể:
“Use in Introduction paragraph 2 to support ASEAN heterogeneous exposure.”
PHẦN 0E. MINIMUM WRITING OUTPUT TỪ BỘ PAPER
0.20. Sau khi đọc sáu paper, sinh viên phải viết ngay
Output A — Research Motivation
400–600 từ
Dùng:
- Peng;
- Chai & Wang;
- Jimoh;
- Chowdhury.
Output B — Literature Bridge
400–600 từ
Dùng:
- Chowdhury;
- Vega-Gutiérrez;
- Wang;
- capital-structure foundational papers.
Output C — ASEAN Setting
250–400 từ
Dùng:
- Jimoh;
- ASEAN official/macroeconomic sources.
Output D — Variable Measurement
300–450 từ
Dùng:
- Rogers;
- Peng;
- Chowdhury.
Output E — Empirical Specification
300–500 từ
Dùng:
- Chowdhury;
- Jimoh exposure logic;
- methodological clustering references.
Output F — H1
150–250 từ
Dùng:
- Chowdhury;
- Vega-Gutiérrez;
- trade-off / pecking-order;
- ASEAN context.
Nếu chưa hoàn thành sáu outputs này thì chưa được coi là đã khai thác xong bộ papers.
PHẦN 0F. NHỮNG LỖI CITATION CẦN TRÁNH
0.21. Sai nguồn UCT
Sai:
Chai and Wang (2026) develop the UCT index used in this study.
Đúng:
Rogers et al. (2024) develop the UCT index used in this study.
Chai and Wang use a different bilateral-relations index.
0.22. Sai dấu literature
Sai:
Prior research uniformly shows uncertainty reduces leverage.
Vega-Gutiérrez et al. (2025) find a positive EPU–leverage relation.
Phải viết:
Prior evidence is mixed and depends on the type of uncertainty and institutional setting.
0.23. Sai khi mô tả ASEAN
Sai:
ASEAN responds uniformly to U.S.–China trade tensions.
Jimoh et al. (2026) show strong cross-country asymmetry.
0.24. Sai khi dùng Peng
Sai:
Peng et al. (2025) show UCT reduces corporate leverage.
Paper này nghiên cứu stock-market performance, không capital structure.
0.25. Sai khi dùng Wang et al.
Sai:
Wang et al. show geopolitical tension affects leverage.
Paper nghiên cứu geography/information asymmetry and capital structure.
0.26. Sai khi dùng Year Fixed Effects
Nếu annual UCT only varies over time:
không được viết:
“We include year fixed effects and estimate the main coefficient on UCT.”
Nếu software vẫn trả coefficient:
phải kiểm tra variable construction hoặc FE implementation ngay.
PHẦN A. MỤC TIÊU CỦA BƯỚC
1. Sau Bước 1 phải trả lời được những câu hỏi nào?
Sinh viên phải xác định đầy đủ:
- U.S.–China Tension Index lấy từ đâu?
- UCT được xây dựng như thế nào?
- UCT đo tension hay uncertainty?
- UCT variation ở cấp nào?
- UCT có monthly hay annual series?
- Annual UCT được merge với firm-year data thế nào?
- Fiscal-year-aligned UCT có thể được xây thế nào?
- Tại sao annual UCT bị Year FE hấp thụ?
- Có thể xử lý vấn đề đó bằng những empirical designs nào?
- Vì sao ASEAN là setting phù hợp?
- ASEAN hiện có bao nhiêu countries?
- Những countries nào thực sự có trong Compustat sample?
- Final sample là ASEAN-11, ASEAN-6 hay nhóm khác?
- Nếu không đủ 11 countries, manuscript phải gọi sample thế nào?
- Tại sao loại financial firms?
- Tại sao loại utilities?
- Tại sao require AT > 0?
- Tại sao require SALE ≥ 0?
- Missing observations được xử lý thế nào?
- Biến nào được winsorize?
- Macro variables có winsorize không?
- UCT có winsorize không?
- Capital Structure được đo bằng những variables nào?
- Book Leverage formula là gì?
- Short-Term Leverage formula là gì?
- Long-Term Leverage formula là gì?
- Debt Maturity formula là gì?
- Market Leverage formula là gì?
- Zero-debt firms được xử lý thế nào trong Debt Maturity?
- Vì sao cần phân biệt debt level và debt maturity?
- Controls được lựa chọn từ literature nào?
- Tangibility ảnh hưởng leverage thế nào?
- Profitability ảnh hưởng leverage thế nào?
- Size ảnh hưởng leverage thế nào?
- Market-to-Book ảnh hưởng leverage thế nào?
- GDP Growth và Inflation có vai trò gì?
- Trade-off theory dự đoán gì?
- Pecking-order theory dự đoán gì?
- Financial flexibility channel là gì?
- Credit-supply channel là gì?
- Supply-chain channel là gì?
- Trade-diversion channel là gì?
- Vì sao UCT có thể tác động khác nhau giữa ASEAN countries?
- UCT có thể tạo cả negative và positive financing effects không?
- Paper trực tiếp về Vietnam năm 2026 tìm thấy gì?
- Bài ASEAN phải khác paper Vietnam đó ở đâu?
- Baseline equation là gì?
- FE nào được phép dùng trong Design A?
- FE nào được phép dùng trong Design B?
- FE nào được phép dùng trong Design C?
- VCE phải xử lý thế nào khi X chỉ biến thiên theo time?
- Vì sao cluster country không tự động đúng cho annual direct UCT?
- Có few-year-cluster problem không?
- Nếu interaction dùng country exposure, có few-country-cluster problem không?
- Có thể dùng trade exposure nào?
- Có thể dùng U.S. exposure và China exposure riêng không?
- Có thể dùng firm-level exposure không?
- Heterogeneity nào nên chuẩn bị?
- ASEAN trade fragmentation literature nói gì?
- Mọi methodological decision đã có citation chưa?
Thành phẩm cuối Bước 1 phải đủ để viết Section 3. Data and Methodology của paper.
PHẦN B. DATA SOURCES
2. Firm-Level Financial Data
Bài mẫu sử dụng chủ yếu:
Compustat Global
cho publicly listed firms thuộc ASEAN.
Đơn vị quan sát
firm-year
Sinh viên phải xác định
| Nội dung | Kết quả cần điền |
|---|---|
| Financial database | |
| Sample period | |
| Unit of observation | |
| Firm identifier | |
| Fiscal-year variable | |
| Fiscal-year-end variable | |
| Country identifier | |
| Industry identifier | |
| Industry classification | |
| Number of firms | |
| Firm-year observations | |
| Number of ASEAN countries |
Yêu cầu
Không tự tạo:
- sample period;
- number of firms;
- number of observations;
- number of countries.
Phải lấy từ:
- actual dataset;
- Stata output;
- sample-selection output;
- final regression sample.
PHẦN C. ASEAN COUNTRY UNIVERSE
3. ASEAN Member States
Research universe gồm:
| Country | ISO alpha-3 | ASEAN Member |
|---|---|---|
| Brunei Darussalam | BRN | Yes |
| Cambodia | KHM | Yes |
| Indonesia | IDN | Yes |
| Lao PDR | LAO | Yes |
| Malaysia | MYS | Yes |
| Myanmar | MMR | Yes |
| Philippines | PHL | Yes |
| Singapore | SGP | Yes |
| Thailand | THA | Yes |
| Timor-Leste | TLS | Yes |
| Vietnam | VNM | Yes |
Important
Country codes trong bảng là reference codes.
Phải verify:
- Compustat
FIC; - exchange-country variable;
- incorporation country;
- headquarters country;
trước khi filter.
Không được giả định raw FIC luôn giống exact ISO code nếu chưa kiểm tra data dictionary.
4. ASEAN-11 Target vs Empirical Sample
Research universe
Empirical sample
Chỉ giữ countries có:
- sufficient firms;
- sufficient years;
- required accounting variables;
- enough variation for panel regression.
Possible practical core
Nếu coverage phù hợp:
- Indonesia;
- Malaysia;
- Philippines;
- Singapore;
- Thailand;
- Vietnam.
Đây có thể được gọi:
Manuscript rule
Nếu final sample là 6 countries:
viết:
“Our final sample covers six ASEAN economies: Indonesia, Malaysia, the Philippines, Singapore, Thailand, and Vietnam.”
Không viết:
“We examine firms across all ASEAN countries.”
5. Country Selection Table
Sinh viên bắt buộc hoàn thành:
| Country | Initial firms | Initial firm-years | First year | Last year | Final firms | Keep? | Reason |
|---|---|---|---|---|---|---|---|
| Brunei | |||||||
| Cambodia | |||||||
| Indonesia | |||||||
| Lao PDR | |||||||
| Malaysia | |||||||
| Myanmar | |||||||
| Philippines | |||||||
| Singapore | |||||||
| Thailand | |||||||
| Timor-Leste | |||||||
| Vietnam |
Final label
After completing this table, write:
Final geographic sample = ______
Examples:
- ASEAN-11;
- ASEAN-8;
- ASEAN-6.
PHẦN D. U.S.–CHINA TENSION INDEX
6. Source
Main independent variable:
U.S.–China Tension Index (UCT)
Developed by:
Rogers, John H.; Sun, Bo; and Sun, Tony (2024).
Paper:
U.S.–China Tension
SSRN Working Paper.
DOI:
10.2139/ssrn.4815838
Data source:
PolicyUncertainty.com – U.S.–China Tension Index
7. Construction of UCT
UCT is a newspaper-based index.
Rogers et al. measure the share of newspaper articles that jointly contain:
- references to:
- United States / U.S.;
- China / Chinese;
- contentious bilateral issues;
- tension-related phrases.
Search terms are informed by topic-modeling methods including:
- K-means;
- guided LDA;
- Newsmap.
The approach builds on the newspaper-based methodology of:
Baker, Bloom, and Davis (2016).
Interpretation
Important wording
Do not write:
“UCT is simply a trade-war dummy.”
UCT captures broader bilateral tensions including:
- trade;
- technology;
- sanctions;
- diplomacy;
- geopolitical issues;
- investment restrictions;
- supply chains.
8. Tension Actions and Uncertainty Channel
Rogers et al. show that transmission operates through:
- tension actions;
- uncertainty surrounding tension actions.
They report that uncertainty-related transmission is particularly important.
Relevance
For capital structure:
uncertainty may affect:
- debt demand;
- debt supply;
- maturity;
- refinancing;
- financial flexibility.
PHẦN E. WHY ASEAN?
9. ASEAN as an Economic Setting
ASEAN economies are closely connected to both:
- China;
- United States.
Relevant features include:
- global value-chain participation;
- manufacturing supply chains;
- exports to U.S.;
- imports/intermediate inputs from China;
- FDI relocation;
- “China + 1” strategies;
- trade diversion.
Therefore UCT can create:
- negative supply-chain shocks;
- positive trade-diversion opportunities;
- investment reallocation;
- financing uncertainty;
- exchange-rate and credit-market effects.
10. IMF Evidence on ASEAN Trade Reallocation
Donato and Kitsios (2026), IMF Working Paper 2026/118:
ASEAN's Trade and Investment in a Fragmented World
examines effects of 2018–2019 China–U.S. tariff increases.
The paper finds:
- several ASEAN countries experienced disproportionate export growth in tariff-targeted products;
- sectoral FDI shifted toward parts of ASEAN;
- Vietnam experienced especially strong FDI growth in targeted sectors;
- trade gains were not uniform across ASEAN;
- longer-run fragmentation can impose aggregate costs.
Implication
UCT may have heterogeneous effects across ASEAN.
Therefore a simple:
average coefficient may conceal:
- winners from trade diversion;
- losers from supply-chain disruption.
PHẦN F. UCT FREQUENCY AND IDENTIFICATION
11. Annual UCT
If using:
all firms in:
- Indonesia;
- Malaysia;
- Vietnam;
- etc.
receive the same UCT value in year $t$.
Critical implication
Annual UCT is a:
time-series regressor
not a country-year regressor.
12. Why Year FE Absorb Annual UCT
Suppose:
where:
Because UCT varies only by year:
is perfectly collinear with a full set of year dummies.
Therefore main UCT cannot be separately estimated.
Critical rule
If annual UCT is identical across all firms:
do not include Year FE in the direct-effect model.
PHẦN G. THREE EMPIRICAL DESIGNS
13. Design A: Annual UCT Direct Effect
Model:
Include
- Firm FE;
- firm controls;
- country macro controls;
- selected global controls;
- possibly linear/quadratic trend.
Do not include
- full Year FE;
- Industry × Year FE;
- Country × Year FE.
These would absorb a pure time-only UCT main effect.
Limitation
Without Year FE, model remains vulnerable to:
- global business cycle;
- COVID;
- global credit conditions;
- world interest rates;
- global risk shocks.
Therefore Design A should not be the only evidence in a strong paper.
14. Design B: Fiscal-Year-Aligned UCT
Use monthly UCT and exact firm fiscal-year end.
Construct:
Example:
December year-end:
June year-end:
Why useful?
Firms with different fiscal-year ends receive different UCT exposure windows.
Can we include Year FE?
Possibly.
But only if:
is economically meaningful.
Must report
| Diagnostic | Result |
|---|---|
| % December fiscal year-end | |
| Number of fiscal-year-end months | |
| Within-year SD of UCT-FY | |
| Minimum within-year UCT-FY | |
| Maximum within-year UCT-FY | |
| Collinearity with Year FE |
Warning
If almost all ASEAN firms have December year-end:
fiscal-year alignment may provide little useful within-year variation.
15. Design C: Exposure-Based Model
Preferred stronger design:
Main UCT effect
Absorbed by Year FE.
This is expected.
Parameter of interest
answers:
When U.S.–China tension increases, do firms with greater economic exposure adjust capital structure differently?
PHẦN H. COUNTRY-LEVEL EXPOSURE
16. ASEAN Trade Exposure
Possible measure:
Alternative:
Prefer predetermined exposure
For causal interpretation, use:
- lagged exposure;
- pre-2018 exposure;
- beginning-of-sample exposure.
Avoid contemporaneous exposure if UCT itself changes trade shares.
17. Country-Exposure Model
Can use Country × Year FE?
No.
The interaction varies at country-year level:
and Country × Year FE would absorb it.
VCE problem
If final sample has only six ASEAN countries:
country-clustered asymptotics are very weak.
Must use extensive sensitivity analysis.
PHẦN I. FIRM-LEVEL EXPOSURE
18. Firm-Level U.S./China Exposure
Stronger if data exist.
Possible proxies:
- firm sales to China;
- firm sales to U.S.;
- foreign sales;
- supply-chain links;
- customer/supplier exposure;
- segment-level geographic sales;
- tariff exposure;
- industry exposure;
- textual exposure.
Preferred model
Why powerful?
Country × Year FE absorb:
- country GDP;
- inflation;
- exchange rates;
- domestic credit conditions;
- domestic policy;
- common UCT main effect;
- country-specific UCT spillovers.
Identification comes from:
firms with different exposure within the same country-year.
This is the strongest extension recommended in this guide.
PHẦN J. SAMPLE CONSTRUCTION
19. Exclude Financial Firms
If GICS Industry Group:
If SIC:
Why?
Financial firms:
- use leverage as part of operations;
- face regulatory capital requirements;
- have different debt/equity interpretation;
- have distinct balance sheets.
Sample paragraph
We exclude financial firms because leverage has a structurally different economic meaning for regulated financial institutions and is strongly shaped by capital-adequacy requirements.
20. Exclude Utilities
If GICS Industry Group:
If SIC:
Why?
Utilities have:
- regulated rates;
- stable cash flows;
- regulated infrastructure investment;
- distinctive debt capacity.
Sample paragraph
We exclude regulated utilities because their financing choices and debt capacity are substantially influenced by sector-specific regulatory arrangements.
21. Total Assets
Require:
Why?
AT is denominator for:
- leverage;
- profitability;
- tangibility;
- cash;
- many firm controls.
22. Sales
Require:
according to the data workflow.
Rule
Do not replace with:
unless the actual code changes.
23. Missing Data
Baseline sample drops observations missing:
- capital-structure DV;
- UCT/exposure;
- firm controls;
- macro controls;
- identifiers;
- FE variables.
Different DVs may have different N
Report separately for:
- Book Leverage;
- ST Leverage;
- LT Leverage;
- Debt Maturity;
- Market Leverage.
24. Sample Selection Table
| Step | Observations removed | Observations remaining |
|---|---|---|
| Initial ASEAN Compustat sample | ||
| Apply ASEAN country filter | ||
| Exclude GIND 40 | ||
| Exclude GIND 55 | ||
| Require AT > 0 | ||
| Require SALE ≥ 0 | ||
| Construct debt variables | ||
| Merge UCT | ||
| Merge macro data | ||
| Remove missing firm controls | ||
| Remove missing macro controls | ||
| Require lead DV | ||
| Final baseline sample |
Also report:
- firms;
- countries;
- industries;
- first year;
- last year.
PHẦN K. WINSORIZATION
25. Main Rule
Winsorize:
- dependent variables;
- continuous firm-level controls;
at:
Do not winsorize
- UCT;
- GDP Growth;
- Inflation;
- other macro variables;
- country codes;
- firm IDs;
- dummies.
Sample paragraph
To mitigate the influence of extreme firm-level accounting observations, we winsorize the dependent variables and continuous firm-level controls at the 1st and 99th percentiles. We do not winsorize UCT or macroeconomic variables because extreme aggregate values may reflect economically meaningful shocks rather than measurement errors.
PHẦN L. CAPITAL STRUCTURE MEASURES
26. Main DV: Book Leverage
where:
- $DLC$ = Debt in Current Liabilities / short-term debt;
- $DLTT$ = Long-Term Debt;
- $AT$ = Total Assets.
Interpretation
Why not Total Liabilities/Assets?
Total liabilities include operating liabilities such as:
- accounts payable;
- accrued expenses;
- tax liabilities.
Capital structure is more directly about financing debt/equity choices.
27. Short-Term Leverage
Why important?
UCT can increase:
- rollover risk;
- refinancing uncertainty;
- bank risk aversion.
Short-term debt may adjust fastest.
Direct precedent
Bui Tien Thinh et al. (2026) find higher U.S.–China tension reduces:
- overall leverage;
- short-term leverage
among Vietnamese listed firms.
28. Long-Term Leverage
Interpretation
Captures long-horizon debt financing.
Possible results
If:
- ST leverage falls;
- LT leverage unchanged;
firms may be reducing refinancing exposure.
If:
- both fall;
firms may be broadly deleveraging.
29. Debt Maturity
require:
Interpretation
Zero-debt firms
Debt maturity is undefined when:
Do not automatically code as zero.
Report:
- zero-debt observations removed;
- resulting debt-maturity sample size.
30. Market Leverage
Common approximation:
where:
using exact code variables.
Rule
Exact Compustat construction must match code.
31. Capital Structure Definition Table
| Variable | Formula | Interpretation | Data source | Original/empirical reference |
|---|---|---|---|---|
| Book Leverage | (DLC+DLTT)/AT | Total debt intensity | Compustat | Capital-structure literature |
| ST Leverage | DLC/AT | Short-term debt intensity | Compustat | Bui et al. (2026) |
| LT Leverage | DLTT/AT | Long-term debt intensity | Compustat | |
| Debt Maturity | DLTT/(DLC+DLTT) | Share of long-term debt | Compustat | Debt-maturity literature |
| Market Leverage | Debt/(Debt+MVE) | Market-based leverage | Compustat |
PHẦN M. CAPITAL STRUCTURE THEORY
32. Modigliani–Miller Benchmark
Modigliani and Miller (1958) provide the frictionless-market benchmark.
Capital structure becomes economically relevant with:
- taxes;
- bankruptcy costs;
- agency conflicts;
- information asymmetry;
- financing frictions.
33. Trade-Off Theory
Firms balance:
- tax benefits of debt;
- expected distress costs.
UCT can increase:
- cash-flow volatility;
- business risk;
- credit spreads;
- probability of distress.
Thus:
Expected:
34. Pecking-Order Theory
Myers (1984) and Myers and Majluf (1984):
financing preference:
- internal funds;
- debt;
- equity.
UCT creates ambiguous effects.
If UCT lowers investment demand
If UCT lowers cash flow
firms may need more external finance:
unless lenders simultaneously tighten supply.
Therefore sign is empirical.
35. Financial Flexibility
Under uncertainty, firms value:
- cash;
- unused debt capacity;
- low refinancing exposure.
Possible response:
36. Credit-Supply Channel
Banks and debt investors may respond to UCT by:
- raising spreads;
- reducing loan amounts;
- shortening maturities;
- demanding collateral.
Observed leverage can fall even if firms would prefer more debt.
Critical interpretation
Leverage reflects:
- financing demand;
- financing supply.
Do not attribute all changes to managerial preference.
PHẦN N. ASEAN-SPECIFIC ECONOMIC CHANNELS
37. Supply-Chain Disruption
ASEAN firms may depend on:
- Chinese intermediate goods;
- U.S. final demand;
- regional supply chains.
UCT may:
38. Trade Diversion
ASEAN can also benefit.
U.S.–China tariffs may redirect:
- exports;
- production;
- FDI
toward ASEAN countries.
Then some firms may experience:
potentially increasing leverage.
Therefore
ASEAN setting naturally supports heterogeneous effects.
39. FDI Reallocation
The 2026 IMF evidence suggests selected ASEAN sectors received additional FDI following tariff escalation, particularly in Vietnam.
Possible consequences:
- stronger growth;
- new factories;
- financing needs;
- foreign-owned-firm changes in capital structure.
40. Exchange-Rate / Global-Finance Channel
UCT can affect:
- investor risk appetite;
- exchange rates;
- global financial conditions.
ASEAN firms with foreign-currency debt may respond differently.
Potential later extension:
if data exist.
PHẦN O. FIRM CONTROLS
41. Core Controls
Canonical firm-level determinants:
- Tangibility;
- Profitability;
- Size;
- Market-to-Book.
Core references:
- Rajan and Zingales (1995);
- Frank and Goyal (2009).
42. Tangibility
Rationale
Tangible assets:
- provide collateral;
- reduce creditor loss given default.
Expected:
with leverage.
43. Profitability
Example:
or exact code definition.
Pecking-order prediction
More profitable firms generate more internal cash:
Often expected:
44. Size
Large firms often have:
- better debt-market access;
- lower default risk;
- greater diversification.
Common expected sign:
45. Market-to-Book
Exact code formula required.
Proxy for:
- growth opportunities;
- market valuation.
High-growth firms may avoid debt because of:
- distress costs;
- underinvestment problems.
Often expected:
46. Optional Cash
Can capture:
- liquidity;
- internal financing;
- financial flexibility.
Caution
UCT may itself change cash holdings.
If Cash is a mechanism, do not automatically control for contemporaneous Cash.
Prefer lagged value where justified.
47. Control Evidence Table
| Variable | Formula | Why? | Expected sign | Reference 1 | Reference 2 |
|---|---|---|---|---|---|
| Tangibility | PPENT/AT | Collateral | + | Rajan & Zingales | Frank & Goyal |
| Profitability | Internal finance | − | Rajan & Zingales | Frank & Goyal | |
| Size | ln(AT) | Debt access | + | Rajan & Zingales | |
| Market-to-Book | Growth opportunities | −/? | Rajan & Zingales | ||
| Cash | CHE/AT | Financial flexibility | −/? |
PHẦN P. MACRO CONTROLS
48. GDP Growth
Captures:
- domestic business cycle;
- investment opportunities;
- credit demand.
49. Inflation
Affects:
- nominal rates;
- real debt values;
- financing conditions.
50. Additional Macro Controls
Potential:
- policy interest rate;
- exchange-rate change;
- domestic credit/GDP;
- financial development;
- stock-market development.
Rule
Do not add mechanically.
Each macro control needs:
- economic rationale;
- source;
- citation.
Workflow rule
Macro variables:
not winsorized.
PHẦN Q. GLOBAL CONTROLS FOR DESIGN A
51. Why Needed?
Annual direct-effect model cannot include Year FE.
Therefore UCT may proxy for:
- Global EPU;
- global geopolitical risk;
- VIX;
- U.S. interest rates;
- world growth;
- COVID.
Potential controls
- Global EPU;
- GPR;
- VIX;
- Fed Funds Rate / U.S. rate;
- world GDP growth.
Caution
UCT can be correlated with these variables.
Report:
- correlations;
- multicollinearity diagnostics;
- coefficient stability.
PHẦN R. TIMING
52. Baseline Timing
Preferred:
Why?
Debt adjustment takes time:
- debt contracts mature;
- refinancing occurs;
- banks revise terms;
- firms alter financing plans.
Sample paragraph
We relate U.S.–China tension in year $t$ to corporate capital structure in year $t+1$, allowing bilateral tension to precede firms' subsequent financing decisions. The lead specification provides a clearer temporal ordering and recognizes that debt adjustment may occur gradually through refinancing and new financing decisions.
Caution
Lead Y:
does not solve all endogeneity.
PHẦN S. FE RULES BY DESIGN
53. Design A: Annual Direct UCT
Allowed
- Firm FE;
- macro controls;
- global controls;
- trend.
Not allowed if estimating main UCT
- Year FE;
- Industry × Year FE;
- Country × Year FE.
All contain full time variation that absorbs pure UCT_t.
54. Design B: Fiscal-Year-Aligned UCT
Potential:
- Firm FE;
- Year FE
if within-year UCT-FY variation remains.
Must test first.
Do not assume.
55. Design C: UCT × Country Exposure
Can use:
- Firm FE;
- Year FE;
- Industry × Year FE.
Cannot use:
- Country × Year FE
if treatment interaction varies only country-year.
56. Design C: UCT × Firm Exposure
Can use:
- Firm FE;
- Year FE;
- Country × Year FE;
- Industry × Year FE.
This is the preferred saturated-FE design.
PHẦN T. VCE / STANDARD ERRORS
57. Design A: Time-Series UCT
UCT varies by:
not by:
- firm;
- country.
Therefore standard errors must reflect common-year dependence.
Recommended sensitivity
Two-way clustering:
or appropriate alternatives.
Critical issue
If sample period has about 20 years:
is small.
Conventional cluster asymptotics may be imperfect.
Report:
- number of year clusters;
- alternative inference;
- coefficient stability.
58. Why Country Clustering Alone is Not Enough in Design A
UCT is identical across ASEAN countries in each year.
Country clustering treats countries as independent treatment clusters, but the treatment itself is common across them.
Therefore:
country clustering alone does not solve the time-series common-shock problem.
This is a crucial difference from country-year variables such as:
- WGI Institutional Quality;
- country TUI.
59. Country-Exposure Design
If treatment:
varies country-year:
possible VCE:
- cluster country;
- cluster year;
- two-way country and year.
But
ASEAN has at most 11 countries.
If empirical sample is ASEAN-6:
Conventional country-clustered inference is weak.
Required robustness
- wild-cluster bootstrap with caution;
- leave-one-country-out;
- country-specific estimates;
- permutation/randomization-style tests if defensible;
- coefficient stability;
- alternative exposure measures.
60. Firm-Exposure Design
If exposure varies by firm:
potential VCE:
- cluster firm;
- two-way firm + year.
Country × Year FE remove shared country-year shocks.
This is statistically and economically preferable when credible firm exposure exists.
PHẦN U. HETEROGENEITY
61. Financial Constraints
The Vietnam direct study finds stronger effects for financially constrained firms.
Possible measures:
- SA Index;
- WW Index;
- Size;
- dividend status;
- credit rating.
Test:
62. Short-Term Debt Dependence
Firms with high historical ST debt may face stronger refinancing risk.
63. Cash Holdings
Cash-rich firms may have greater financing flexibility.
64. Trade Exposure
Core extension:
Why separate?
ASEAN firms may:
- benefit from U.S. trade diversion;
- suffer from China input dependence.
Signs can differ.
65. Industry Heterogeneity
Potential high-exposure industries:
- electronics;
- machinery;
- textiles;
- manufacturing;
- semiconductors;
- transport;
- export-intensive sectors.
Rule
Do not hard-code industries without evidence.
Use:
- trade data;
- tariff exposure;
- input-output exposure;
- literature.
66. Institutional Quality
UCT may affect leverage differently under:
- stronger creditor rights;
- better institutions;
- deeper capital markets.
Potential interaction:
Caution
If IQ is country-year:
interaction structure and FE identification must be checked carefully.
67. Financial Development
Possible moderators:
- Domestic Credit/GDP;
- stock-market capitalization;
- banking depth.
Countries with deeper financial systems may absorb external tension shocks better.
PHẦN V. CLOSEST DIRECT PAPER
68. Bui Tien Thinh et al. (2026)
Paper:
Corporate Leverage and U.S.-China Tensions: Evidence from Vietnam
Cogent Economics & Finance, 14(1), 2613610.
DOI:
10.1080/23322039.2026.2613610
Sample
- Vietnamese listed nonfinancial firms;
- 2017–2023;
- 1,301 firms;
- 8,315 observations.
Main findings
Higher U.S.–China tensions are associated with:
- lower overall leverage;
- lower short-term leverage.
Effects are stronger for:
- financially constrained firms;
- particular industries;
- COVID period.
Methods include:
- baseline regressions;
- 2SLS;
- dynamic GMM robustness.
69. Why the ASEAN Paper Must Go Beyond the Vietnam Paper
The ASEAN paper cannot claim novelty simply because:
“No one has studied UCT and leverage.”
A direct Vietnam paper already exists.
Potential contributions:
- multi-country ASEAN evidence;
- cross-country heterogeneity;
- long-term vs short-term debt;
- debt maturity;
- U.S. vs China exposure;
- trade-diversion effects;
- institutional heterogeneity;
- financial-development heterogeneity;
- firm-level exposure design;
- ASEAN fragmentation context.
Strong contribution statement
We extend the emerging literature from a single-country setting to ASEAN and examine whether firms' financing responses to U.S.–China tensions depend on their trade exposure and domestic financial environments.
PHẦN W. CORE CAPITAL STRUCTURE REFERENCES
70. Foundational Theory
Modigliani, F., & Miller, M. H. (1958).
The Cost of Capital, Corporation Finance and the Theory of Investment.
American Economic Review, 48(3), 261–297.
Jensen, M. C., & Meckling, W. H. (1976).
Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure.
Journal of Financial Economics, 3(4), 305–360.
DOI: 10.1016/0304-405X(76)90026-X
Myers, S. C. (1984).
The Capital Structure Puzzle.
The Journal of Finance, 39(3), 575–592.
DOI: 10.1111/j.1540-6261.1984.tb03646.x
Myers, S. C., & Majluf, N. S. (1984).
Corporate Financing and Investment Decisions When Firms Have Information That Investors Do Not Have.
Journal of Financial Economics, 13(2), 187–221.
71. Empirical Capital Structure
Rajan, R. G., & Zingales, L. (1995).
What Do We Know about Capital Structure? Some Evidence from International Data.
The Journal of Finance, 50(5), 1421–1460.
Lemmon, M. L., Roberts, M. R., & Zender, J. F. (2008).
Back to the Beginning: Persistence and the Cross-Section of Corporate Capital Structure.
The Journal of Finance, 63(4), 1575–1608.
Frank, M. Z., & Goyal, V. K. (2009).
Capital Structure Decisions: Which Factors Are Reliably Important?
Financial Management, 38(1), 1–37.
PHẦN X. UCT AND ASEAN REFERENCES
72. U.S.–China Tension
Rogers, J. H., Sun, B., & Sun, T. (2024).
U.S.–China Tension.
SSRN Working Paper.
DOI: 10.2139/ssrn.4815838
Data:
PolicyUncertainty.com – U.S.–China Tension Index
Rogers et al. document that elevated UCT is associated with:
- lower U.S. corporate investment;
- stronger effects among firms more exposed to China;
- supply-chain reconfiguration away from China.
73. ASEAN Fragmentation
Donato, G., & Kitsios, E. (2026).
ASEAN's Trade and Investment in a Fragmented World.
IMF Working Paper 2026/118.
DOI: 10.5089/9798229050128.001
Useful for:
- ASEAN relevance;
- 2018–2019 U.S.–China tariffs;
- trade diversion;
- FDI reallocation;
- heterogeneous country responses.
74. ASEAN Membership
Official source:
ASEAN Secretariat.
Timor-Leste was admitted as ASEAN's 11th Member State on:
26 October 2025.
Methodological implication
A paper written in 2026 should not describe ASEAN as a ten-member organization.
However, historical sample inclusion of Timor-Leste depends on:
- listed-firm data;
- sample years;
- research design.
PHẦN Y. METHOD DECISION MATRIX
75. Bảng bắt buộc
| Methodological decision | What the paper does | Why? | Supporting reference | Verified? |
|---|---|---|---|---|
| Firm database | Compustat Global | ASEAN firm data | ||
| ASEAN universe | 11 member states | Regional scope | ASEAN Secretariat | |
| Final countries | Data availability | Country Selection Table | ||
| Sample label | ASEAN-X | Accuracy | ||
| UCT source | Rogers et al. (2024) | Bilateral tension | Rogers et al. | |
| UCT frequency | ||||
| Fiscal-year alignment | Better timing | |||
| Exclude GIND 40 | Yes | Financial regulation | Capital-structure literature | |
| Exclude GIND 55 | Yes | Utility regulation | Capital-structure literature | |
| AT > 0 | Yes | Valid ratios | ||
| SALE ≥ 0 | Yes | Data validity | ||
| Winsor DV | 1st–99th | Outliers | Empirical literature | |
| Winsor firm controls | 1st–99th | Outliers | ||
| Winsor macro | No | Preserve aggregate variation | ||
| Winsor UCT | No | Preserve tension shocks | ||
| Main DV | Book Leverage | Debt intensity | Rajan & Zingales | |
| ST Leverage | Alternative | Refinancing risk | Bui et al. (2026) | |
| LT Leverage | Alternative | Long-term debt | ||
| Debt Maturity | Alternative | Maturity structure | Debt-maturity literature | |
| Market Leverage | Alternative | Market structure | Capital-structure literature | |
| Timing | UCT_t → Y_t+1 | Temporal ordering | ||
| Design A Year FE | No | UCT absorbed | Identification logic | |
| Design B Year FE | Conditional | FY variation | ||
| Design C Year FE | Yes | Common global shocks | Exposure design | |
| Firm FE | Yes | Persistent heterogeneity | ||
| Industry × Year FE | Design C / valid Design B | Sector shocks | ||
| Country × Year FE | Firm-exposure Design C only | Stronger identification | ||
| VCE Design A | Firm + Year sensitivity | Time-series treatment | Cameron et al. | |
| VCE country-exposure | Country + Year sensitivity | Country-year treatment | Cameron et al. | |
| Few-cluster diagnostics | Yes | ASEAN has few countries | Cluster literature |
PHẦN Z. IDENTIFICATION DIAGNOSTIC TABLE
76. Bảng bắt buộc
| Diagnostic | Result |
|---|---|
| UCT frequency | |
| UCT varies by country? | No |
| Unique UCT values per year | |
| Fiscal-year alignment used? | |
| Within-year UCT-FY variation | |
| Main design | A / B / C |
| Exposure level | None / Country / Industry / Firm |
| Year FE allowed? | |
| Industry × Year FE allowed? | |
| Country × Year FE allowed? | |
| Main source of identification | |
| Number of year clusters | |
| Number of country clusters | |
| Few-cluster concern | |
| Planned inference robustness |
Rule
Sinh viên chưa hoàn thành bảng này:
không được chạy baseline regression.
PHẦN AA. MODEL REFERENCE MATRIX
77. Bảng bắt buộc
| Study | Sample | Main X | Main Y | Controls | FE/Model | VCE | Bài hiện tại học gì? |
|---|---|---|---|---|---|---|---|
| Rogers et al. (2024) | U.S. firms / aggregate | UCT | Investment/supply chain | UCT construction/exposure | |||
| Bui et al. (2026) | Vietnam | UCT | Leverage/ST leverage | 2SLS/GMM robustness | Closest direct study | ||
| Donato & Kitsios (2026) | ASEAN | U.S.–China tariff shock | Trade/FDI | DiD | ASEAN economic context | ||
| Rajan & Zingales (1995) | International | Firm determinants | Leverage | Firm controls | |||
| Frank & Goyal (2009) | Corporate sample | Determinants | Leverage | Robust controls | |||
| ASEAN capital-structure paper 1 | ASEAN | Leverage | Regional design | ||||
| ASEAN capital-structure paper 2 | ASEAN | Debt/maturity | Regional heterogeneity | ||||
| Debt-maturity study | Maturity | DV construction | |||||
| Cameron et al. (2011) | Method | Multiway clustering | |||||
| Cameron & Miller (2015) | Method | Cluster inference |
Tối thiểu:
10 model/method references.
PHẦN AB. SAMPLE PARAGRAPHS FOR DATA AND METHODOLOGY
78. Sample paragraph: ASEAN sample
We construct a firm-year panel of publicly listed corporations from ASEAN using Compustat Global. Our geographic universe corresponds to the current ASEAN membership, while the final estimation sample retains only economies with sufficient firm-level accounting coverage and the variables required for the empirical analysis. We explicitly report the countries included in the final sample rather than assuming equal data availability across all ASEAN members. Financial firms and regulated utilities are excluded because their balance-sheet structures and financing decisions differ materially from those of other nonfinancial corporations. We further require positive total assets and valid nonnegative sales observations. Continuous firm-level dependent variables and controls are winsorized at the 1st and 99th percentiles, whereas UCT and macroeconomic variables are not winsorized.
79. Sample paragraph: UCT
Our main explanatory variable is the U.S.–China Tension Index developed by Rogers et al. (2024). The index is constructed from leading U.S. newspapers and captures the share of articles that simultaneously reference the United States and China, contentious bilateral issues, and language indicating heightened tension. Higher values therefore represent periods of more intense U.S.–China tension. The index captures a broader bilateral relationship than a simple trade-war indicator, including trade, technology, sanctions, investment restrictions, geopolitical disputes, and related uncertainty.
80. Sample paragraph: ASEAN rationale
ASEAN provides a particularly relevant setting because its economies are deeply integrated with both China and the United States through trade, foreign direct investment, and global value chains. U.S.–China tensions can therefore generate competing effects for ASEAN firms. Supply-chain disruption and heightened uncertainty may weaken cash flows and debt capacity, while trade diversion and production relocation may create new investment opportunities and external financing needs. Recent IMF evidence documents substantial reallocation of trade and investment across ASEAN following the 2018–2019 U.S.–China tariff escalation, with effects varying across countries and sectors (Donato & Kitsios, 2026).
81. Sample paragraph: Capital Structure
Our primary capital-structure measure is book leverage, defined as short-term plus long-term interest-bearing debt divided by total assets. We separately examine short-term and long-term leverage to distinguish refinancing-sensitive borrowing from longer-horizon financing. We also examine debt maturity, measured as long-term debt relative to total interest-bearing debt, and market leverage as alternative specifications. These outcomes allow us to determine whether U.S.–China tensions affect firms' overall use of debt or primarily alter the maturity composition of debt financing.
82. Sample paragraph: Identification issue
A key feature of the UCT index is that annual UCT is an aggregate time-series variable common to all ASEAN firms in a given year. Consequently, a full set of year fixed effects would absorb the main UCT effect. We therefore distinguish the direct annual-index specification from stronger exposure-based designs. The direct specification uses firm fixed effects and macroeconomic and global controls without full year fixed effects, whereas exposure-based specifications interact UCT with predetermined country- or firm-level exposure and include year fixed effects. Where firm-level exposure is available, we further include country-by-year fixed effects so that identification comes from differences in exposure across firms operating in the same country and year.
83. Sample paragraph: Closest literature
Our study is closely related to Bui Tien Thinh et al. (2026), who document that greater U.S.–China tension is associated with lower overall and short-term leverage among Vietnamese listed firms. We extend this setting to ASEAN in order to examine cross-country differences, debt-maturity responses, and heterogeneous effects arising from differences in firms' and countries' exposure to the United States and China.
PHẦN AC. CORE REFERENCES
84. U.S.–China Tension
Rogers, J. H., Sun, B., & Sun, T. (2024).
U.S.–China Tension.
SSRN Working Paper.
DOI: 10.2139/ssrn.4815838
Data:
PolicyUncertainty.com – U.S.–China Tension Index
85. Closest Direct Leverage Study
Bui Tien Thinh, Le Thi My Nuong, Ha Thi Thuy, & Dao Nhat Minh. (2026).
Corporate Leverage and U.S.-China Tensions: Evidence from Vietnam.
Cogent Economics & Finance, 14(1), 2613610.
DOI: 10.1080/23322039.2026.2613610
86. ASEAN Fragmentation
Donato, G., & Kitsios, E. (2026).
ASEAN's Trade and Investment in a Fragmented World.
IMF Working Paper 2026/118.
DOI: 10.5089/9798229050128.001
87. Capital Structure Foundations
Modigliani, F., & Miller, M. H. (1958).
The Cost of Capital, Corporation Finance and the Theory of Investment.
American Economic Review, 48(3), 261–297.
Jensen, M. C., & Meckling, W. H. (1976).
Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure.
Journal of Financial Economics, 3(4), 305–360.
Myers, S. C. (1984).
The Capital Structure Puzzle.
The Journal of Finance, 39(3), 575–592.
Myers, S. C., & Majluf, N. S. (1984).
Corporate Financing and Investment Decisions When Firms Have Information That Investors Do Not Have.
Journal of Financial Economics, 13(2), 187–221.
88. Empirical Capital Structure
Rajan, R. G., & Zingales, L. (1995).
What Do We Know about Capital Structure? Some Evidence from International Data.
The Journal of Finance, 50(5), 1421–1460.
Lemmon, M. L., Roberts, M. R., & Zender, J. F. (2008).
Back to the Beginning: Persistence and the Cross-Section of Corporate Capital Structure.
The Journal of Finance, 63(4), 1575–1608.
Frank, M. Z., & Goyal, V. K. (2009).
Capital Structure Decisions: Which Factors Are Reliably Important?
Financial Management, 38(1), 1–37.
89. Clustered Inference
Moulton, B. R. (1990).
An Illustration of a Pitfall in Estimating the Effects of Aggregate Variables on Micro Units.
The Review of Economics and Statistics, 72(2), 334–338.
Cameron, A. C., Gelbach, J. B., & Miller, D. L. (2011).
Robust Inference With Multiway Clustering.
Journal of Business & Economic Statistics, 29(2), 238–249.
Cameron, A. C., & Miller, D. L. (2015).
A Practitioner's Guide to Cluster-Robust Inference.
Journal of Human Resources, 50(2), 317–372.
PHẦN AD. STUDENT LITERATURE SEARCH
90. UCT Corporate Effects
Find at least:
3 papers
on:
- UCT and corporate investment;
- UCT and leverage;
- UCT and supply chains;
- UCT and stock-market effects.
91. Uncertainty and Capital Structure
Find at least:
4 papers
on:
- EPU and leverage;
- geopolitical risk and debt;
- trade-policy uncertainty and financing;
- financial flexibility under uncertainty.
92. ASEAN Capital Structure
Find at least:
3 empirical papers
covering multiple ASEAN economies.
Extract:
- countries;
- leverage definitions;
- firm controls;
- macro controls;
- FE;
- VCE.
93. Debt Maturity
Find at least:
2 papers
on:
- maturity structure;
- short-term debt;
- rollover risk.
94. Exposure-Based Designs
Find at least:
3 papers
using:
- trade exposure;
- tariff exposure;
- China exposure;
- U.S. exposure;
- supply-chain exposure.
95. Total Methodology References
Recommended:
20–25 papers/sources
before moving to Theory.
PHẦN AE. AI RULES
96. AI May Help With
- UCT explanation;
- capital-structure formulas;
- ASEAN country checks;
- trade exposure concepts;
- FE identification;
- clustering;
- literature keywords;
- consistency checks;
- methodology drafting.
97. AI Must Not Invent
- sample countries;
- firm counts;
- UCT values;
- Compustat variables;
- exposure variables;
- coefficients;
- p-values;
- DOI;
- paper methods not verified in original sources.
Every empirical statement must be checked against:
- original paper;
- PolicyUncertainty.com;
- Compustat;
- actual Stata code;
- official ASEAN/IMF/WDI/trade data;
- regression output.
PHẦN AF. CHECKLIST CUỐI BƯỚC 1
Đã xác định 11 ASEAN Member States hiện nay.
Đã xác định Timor-Leste là thành viên từ 26 October 2025.
Đã tạo Country Selection Table.
Đã xác định final empirical countries.
Đã gọi đúng sample là ASEAN-X.
Đã xác định Compustat source.
Đã đọc Rogers et al. (2024).
Đã xác định exact UCT file.
Đã xác định UCT frequency.
Đã hiểu UCT annual là time-series regressor.
Đã hiểu Year FE absorb annual UCT.
Đã chọn Design A, B hoặc C.
Nếu Design A: không dùng Year FE.
Nếu Design A: không dùng Industry × Year FE.
Nếu Design A: đã thêm appropriate global/time controls.
Nếu Design B: đã xây fiscal-year-aligned UCT.
Nếu Design B: đã kiểm tra within-year variation.
Nếu Design C: đã xác định exposure variable.
Nếu Design C country exposure: đã dùng predetermined exposure.
Nếu Design C firm exposure: đã xác định credible firm-level source.
Đã xác định lý do loại GIND 40.
Đã xác định lý do loại GIND 55.
Đã kiểm tra AT > 0.
Đã kiểm tra SALE ≥ 0.
Đã xác định missing-data treatment.
Đã winsor DV 1st–99th.
Đã winsor firm controls 1st–99th.
Đã xác định macro variables không winsorize.
Đã xác định UCT không winsorize.
Đã xác định Book Leverage formula.
Đã xác định ST Leverage formula.
Đã xác định LT Leverage formula.
Đã xác định Debt Maturity formula.
Đã xác định zero-debt treatment.
Đã xác định Market Leverage formula.
Đã xác định Tangibility.
Đã xác định Profitability.
Đã xác định Size.
Đã xác định Market-to-Book.
Đã xác định GDP Growth.
Đã xác định Inflation.
Đã đọc Bui et al. (2026).
Đã xác định contribution beyond Vietnam paper.
Đã đọc Donato & Kitsios (2026).
Đã viết ASEAN economic rationale.
Đã viết baseline equation.
Đã xác định FE đúng theo design.
Đã xác định VCE đúng theo design.
Đã báo number of year clusters.
Đã báo number of country clusters.
Đã kiểm tra few-cluster concern.
Đã lên kế hoạch leave-one-country-out.
Đã hoàn thành Identification Diagnostic Table.
Đã hoàn thành Sample Selection Table.
Đã hoàn thành Country Selection Table.
Đã hoàn thành Capital Structure Definition Table.
Đã hoàn thành Method Decision Matrix.
Đã hoàn thành Model Reference Matrix.
Đã có ít nhất 20–25 methodology references.
Đã viết Draft Data and Methodology 1.800–2.500 từ.
Mọi citation đã được kiểm chứng.
PHẦN AG. THÀNH PHẨM CUỐI CÙNG CỦA BƯỚC 1
Bước 1 chỉ được coi là hoàn thành khi nhóm có đủ:
- Country Selection Table
- Sample Selection Table
- UCT Data Note
- Capital Structure Definition Table
- Identification Diagnostic Table
- Method Decision Matrix
- Model Reference Matrix
- Reference List cho Methodology
- Draft Section 3. Data and Methodology, 1.800–2.500 từ
Khi các sản phẩm trên hoàn thành, nhóm chuyển sang:
BƯỚC 2. KIỂM TRA VÀ DIỄN GIẢI KẾT QUẢ MÔ HÌNH CƠ SỞ
PHẦN AH. REFERENCE LIST CỦA BỘ PAPER ĐƯỢC CUNG CẤP
Peng, C., Deng, H., Xie, J., & Liu, X. (2025).
US-China tension and stock market performance in US and China: New insights from time-varying quantile causality method.
Finance Research Letters, 85, 107888.
DOI: 10.1016/j.frl.2025.107888
Chai, H., & Wang, Y. (2026).
China–US relations and overseas capital allocation of Chinese firms.
Finance Research Letters, 93, 109621.
DOI: 10.1016/j.frl.2026.109621
Wang, X., Wang, J., & Johnson, L.
Geography and capital structure.
Canadian Journal of Administrative Sciences, 35(1), 107–122.
DOI: 10.1002/cjas.1383
Note: first published online in 2016; verify the final bibliographic year required by the target citation style.
Jimoh, S. O., Goh, S. K., & Tham, S. Y. (2026).
Between giants: The US-China trade war and ASEAN’s evolving role in global trade.
International Economics and Economic Policy, 23(2), 1–24.
DOI: 10.1007/s10368-026-00730-x
Vega-Gutiérrez, P. L., López-Iturriaga, F. J., & Rodríguez-Sanz, J. A. (2025).
Economic policy uncertainty and capital structure in Europe: An agency approach.
The European Journal of Finance, 31(1), 53–75.
DOI: 10.1080/1351847X.2024.2364044
Chowdhury, M. S. R., Aram, M., Javadi, S., & Nejadmalayeri, A. (2025).
Geopolitical risk and corporate capital structure.
Journal of Corporate Finance, 93, 102796.
DOI: 10.1016/j.jcorpfin.2025.102796
BƯỚC 2. KIỂM TRA, DIỄN GIẢI VÀ VIẾT KẾT QUẢ MÔ HÌNH CƠ SỞ
Bài mẫu: U.S.–China Tension and Corporate Capital Structure in ASEAN Countries
Mục tiêu đầu ra: Từ regression output hiện có, sinh viên phải viết được hoàn chỉnh phần Baseline Results / Main Results của paper, có:
- kiểm tra specification;
- kiểm tra dấu và ý nghĩa thống kê;
- tính economic significance;
- so sánh giữa các capital-structure measures;
- đối chiếu literature;
- giải thích ASEAN mechanism;
- tránh overclaim;
- tạo bảng kết quả đủ chuẩn để dùng trong manuscript.
Bước 2 không yêu cầu sinh viên chạy thêm hàng chục mô hình mới.
Trước hết phải:
- kiểm tra model cơ sở đã chạy đúng chưa;
- đọc đúng coefficient;
- viết đúng ý nghĩa kinh tế;
- so sánh đúng với literature;
- xác định baseline finding có đủ đáng tin để chuyển sang robustness hay chưa.
PHẦN A. THÀNH PHẨM CUỐI BƯỚC 2
Bước 2 chỉ hoàn thành khi nhóm có đủ:
- Baseline Regression Audit
- Main Results Table
- Coefficient Interpretation Table
- Economic Significance Table
- Cross-Outcome Comparison Table
- Literature Comparison Matrix
- Draft Baseline Results, 1.000–1.500 từ
- Draft Economic Interpretation, 400–700 từ
- Danh sách các vấn đề phải xử lý ở Bước 3
PHẦN B. TRƯỚC KHI ĐỌC KẾT QUẢ: PHẢI XÁC NHẬN MODEL ĐÚNG
1. Không được nhìn p-value trước khi kiểm tra specification
Sinh viên phải trả lời theo đúng thứ tự:
1.1. Main X là gì?
Phải ghi chính xác:
UCT = U.S.–China Tension Index
Nguồn:
Rogers, Sun, and Sun (2024)
Không được nhầm với:
- China–US Relations Tension Index của Chai and Wang (2026);
- GPR;
- EPU;
- Trade War dummy.
2. X variation ở cấp nào?
Nếu dùng annual UCT:
thì main X chỉ biến thiên theo:
year
không biến thiên theo:
- firm;
- country.
Do đó phải kiểm tra ngay
Baseline có:
Nếu:
YES
và annual UCT giống nhau cho toàn bộ firms trong từng năm:
DỪNG DIỄN GIẢI.
Model có vấn đề identification hoặc variable construction.
3. Baseline Model Audit
Sinh viên phải hoàn thành bảng:
| Item | Expected | Actual regression | Pass? |
|---|---|---|---|
| Sample | ASEAN-X | ||
| Unit | Firm-year | ||
| Main X | Annual UCT / FY-aligned UCT | ||
| Main Y timing | t+1 | ||
| Firm FE | Yes | ||
| Year FE | No nếu annual UCT direct effect | ||
| Country FE | Không cần nếu Firm FE | ||
| Industry × Year FE | Không nếu annual direct UCT | ||
| Country × Year FE | Không nếu annual direct UCT | ||
| Firm controls | Yes | ||
| Macro controls | Yes | ||
| UCT winsorized | No | ||
| Firm variables winsorized | 1–99 | ||
| VCE | Appropriate for design | ||
| Number of observations | Reported | ||
| Number of firms | Reported | ||
| Number of years | Reported | ||
| Number of countries | Reported |
Quy tắc
Nếu có bất kỳ dòng nào:
Pass = No
thì chưa được viết kết luận kinh tế.
PHẦN C. XÁC ĐỊNH CỘT BASELINE
4. Baseline table nên có cấu trúc như thế nào?
Một cấu trúc dễ viết:
| Variables | (1) Book Lev | (2) ST Lev | (3) LT Lev | (4) Debt Maturity | (5) Market Lev |
|---|---|---|---|---|---|
| UCT | |||||
| Tangibility | |||||
| Profitability | |||||
| Size | |||||
| Market-to-Book | |||||
| GDP Growth | |||||
| Inflation | |||||
| Firm FE | Yes | Yes | Yes | Yes | Yes |
| Year FE | No* | No* | No* | No* | No* |
| Observations | |||||
| Firms | |||||
| Adjusted R² / Within R² |
* nếu annual UCT direct-effect design.
Nếu actual paper dùng chỉ 4 DVs
Ví dụ:
- TDB;
- TDM;
- LDB;
- LDM;
thì thay đúng tên.
Không cần ép đủ 5 outcome nếu code không có.
PHẦN D. MAP TÊN BIẾN CODE SANG Ý NGHĨA KINH TẾ
5. Bảng bắt buộc
| Code variable | Manuscript name | Exact formula | Higher value means |
|---|---|---|---|
| TDB | |||
| TDM | |||
| LDB | |||
| LDM | |||
| STLEV | |||
| DM |
Không được viết Results nếu chưa map xong
Ví dụ:
Nếu:
thì manuscript phải gọi:
Total Market Leverage
Không được gọi chung:
“debt ratio”
PHẦN E. ĐỌC MAIN COEFFICIENT
6. Coefficient UCT cần ghi 5 thông tin
Mỗi dependent variable phải ghi:
- coefficient;
- standard error / t-statistic;
- p-value hoặc significance level;
- sign;
- economic meaning.
Bảng
| DV | β(UCT) | SE / t | p-value | Significance | Direction |
|---|---|---|---|---|---|
| Book Leverage | |||||
| ST Leverage | |||||
| LT Leverage | |||||
| Debt Maturity | |||||
| Market Leverage |
PHẦN F. VIẾT KHI UCT CÓ HỆ SỐ ÂM
7. Nếu UCT âm và có ý nghĩa
Ví dụ giả định:
với:
Không viết
U.S.–China tension causes ASEAN firms to reduce leverage by 0.6%.
Có ba lỗi:
- coefficient có thể không phải 0.6%;
- chưa chắc causal;
- chưa biết scale của UCT.
Viết đúng hơn
The coefficient on UCT is negative and statistically significant, indicating that periods of greater U.S.–China tension are associated with lower corporate leverage among ASEAN firms.
Nếu muốn mạnh hơn một chút
The negative association is consistent with the view that heightened bilateral tension increases financing uncertainty and induces firms to preserve debt capacity and financial flexibility.
Nếu chưa có strong identification
Dùng:
- associated with;
- related to;
- consistent with.
Không dùng:
- causes;
- leads to;
- results in
một cách mặc định.
PHẦN G. VIẾT KHI UCT DƯƠNG
8. Nếu UCT dương và có ý nghĩa
Đây không phải kết quả sai.
Vega-Gutiérrez et al. (2025) tìm thấy:
trong European firms.
Có thể giải thích
- internal cash flow suy giảm;
- firms cần external financing;
- debt được ưu tiên hơn equity;
- ASEAN trade-diversion opportunities tạo investment financing demand;
- local financial institutions duy trì credit;
- firms hedge liquidity risk bằng committed borrowing.
Đoạn mẫu
In contrast to the negative leverage response documented for geopolitical risk by Chowdhury et al. (2025), our estimated UCT coefficient is positive. This result is not necessarily inconsistent with the broader uncertainty literature. Vega-Gutiérrez et al. (2025) document a positive relationship between economic policy uncertainty and leverage and emphasize the role of agency and institutional conditions. In the ASEAN setting, elevated U.S.–China tension may also generate trade-diversion and investment opportunities that increase firms' external financing needs.
Sau đó phải kiểm tra
- ST debt tăng hay LT debt tăng?
- investment tăng hay giảm?
- cash flow có yếu đi không?
- country exposure heterogeneity có giải thích được không?
PHẦN H. VIẾT KHI UCT KHÔNG CÓ Ý NGHĨA
9. Nếu main coefficient insignificant
Không viết:
UCT has no effect on ASEAN firms.
Viết:
We find no statistically detectable average association between UCT and [DV] in the baseline specification.
Có thể do
- ASEAN responses offset each other;
- some countries gain from trade diversion;
- others suffer supply-chain disruption;
- debt effects differ by maturity;
- UCT effect concentrated in highly exposed firms;
- annual aggregate UCT too coarse.
Jimoh et al. rất quan trọng ở đây
Nếu ASEAN country responses are heterogeneous:
không đồng nghĩa:
Đoạn mẫu
The insignificant average coefficient may conceal economically meaningful cross-country heterogeneity. Jimoh et al. (2026) show that ASEAN economies respond asymmetrically to U.S.–China trade tensions depending on their pre-existing trade exposure. This motivates the exposure-based analyses developed in subsequent sections.
PHẦN I. ECONOMIC SIGNIFICANCE
10. Statistical significance chưa đủ
Sinh viên phải trả lời:
Một thay đổi có ý nghĩa thực tế của UCT làm leverage thay đổi bao nhiêu?
11. Cách 1: One-Standard-Deviation Change
Nếu:
và:
thì:
Nếu mean leverage:
thì relative effect:
12. Ví dụ minh họa
Giả sử:
thì:
tức leverage giảm:
1.5 percentage points
Relative to mean:
Viết
A one-standard-deviation increase in UCT is associated with a 1.5-percentage-point reduction in leverage, equivalent to approximately 5% of the sample mean.
Lưu ý
Đây chỉ là ví dụ.
Không dùng các số trên nếu output thực tế khác.
PHẦN J. CÁCH 2: INTERQUARTILE CHANGE
13. Q75 − Q25
Có thể tính:
Effect:
Tại sao useful?
UCT có thể có extreme spikes.
IQR comparison:
- dễ hiểu;
- ít phụ thuộc extreme values hơn 1 SD.
PHẦN K. ECONOMIC SIGNIFICANCE TABLE
14. Bảng bắt buộc
| DV | β(UCT) | SD(UCT) | Effect of +1 SD | Mean DV | % of mean |
|---|---|---|---|---|---|
| Book Leverage | |||||
| ST Leverage | |||||
| LT Leverage | |||||
| Debt Maturity | |||||
| Market Leverage |
Optional Panel B
| DV | β(UCT) | Q25 UCT | Q75 UCT | Effect Q25→Q75 |
|---|---|---|---|---|
PHẦN L. SO SÁNH TOTAL, SHORT-TERM VÀ LONG-TERM DEBT
15. Đây là phần phải viết chứ không chỉ report coefficients
Case A
Interpretation
UCT primarily affects:
- short-term borrowing;
- refinancing;
- rollover-sensitive debt.
Viết
The decline in total leverage appears to be concentrated in short-term borrowing. This pattern suggests that firms may respond to heightened bilateral tension by reducing refinancing exposure rather than uniformly cutting all debt financing.
16. Case B
and:
Interpretation
Firms shift toward longer maturity.
Possible mechanism:
- reduce rollover risk;
- lock in long-term financing;
- manage uncertainty.
17. Case C
and:
Interpretation
Broad deleveraging.
Supports:
- financial flexibility;
- tighter debt supply;
- lower financing demand.
18. Case D
but:
Interpretation
Potential maturity shortening.
Could imply:
- long-term lenders more reluctant;
- firms resort to short-term financing;
- credit horizon contracts.
This is economically different from total deleveraging.
PHẦN M. CROSS-OUTCOME COMPARISON TABLE
19. Bảng bắt buộc
| Outcome | Sign UCT | Significant? | Economic interpretation | Main mechanism suggested |
|---|---|---|---|---|
| Total leverage | ||||
| Short-term leverage | ||||
| Long-term leverage | ||||
| Debt maturity | ||||
| Market leverage |
Student task
Viết một paragraph:
“Taken together, the results indicate…”
không viết từng column rời rạc.
PHẦN N. CONTROLS: KHÔNG ĐƯỢC BỎ QUA HOÀN TOÀN
20. Không cần diễn giải mọi control
Results section không nên biến thành:
Tangibility positive. Profitability negative. Size positive…
Nhưng cần kiểm tra signs có hợp lý không.
21. Tangibility
Common expected:
because of collateral capacity.
Nếu positive/significant:
The positive coefficient on tangibility is broadly consistent with collateral-based explanations of leverage.
Reference:
- Rajan and Zingales (1995);
- Frank and Goyal (2009).
22. Profitability
Common pecking-order expectation:
If negative:
More profitable firms rely less on external debt, consistent with the pecking-order view that firms with greater internal funds have lower financing needs.
23. Size
Often:
If positive:
Larger firms exhibit greater leverage, consistent with better access to debt markets and lower perceived default risk.
24. Market-to-Book
Often:
but not universal.
Avoid overinterpreting.
PHẦN O. CONTROL SIGN CHECK TABLE
25. Bảng
| Control | Expected from literature | Actual sign | Significant? | Concern? |
|---|---|---|---|---|
| Tangibility | + | |||
| Profitability | − | |||
| Size | + | |||
| Market-to-Book | −/? | |||
| GDP Growth | ? | |||
| Inflation | ? |
Rule
Unexpected controls do not automatically invalidate model.
But if nearly all controls behave implausibly:
check:
- formulas;
- winsorization;
- scaling;
- sample;
- variable coding.
PHẦN P. DÙNG CHOWDHURY ET AL. (2025) ĐỂ VIẾT MAIN RESULTS
26. Đây là paper comparison chính
Sinh viên phải so sánh ít nhất:
- sign;
- total vs long-term debt;
- timing;
- mechanism;
- specification.
Nếu kết quả âm
Our findings are consistent with Chowdhury et al. (2025), who document a persistent negative relationship between geopolitical risk and corporate leverage. The similarity suggests that bilateral U.S.–China tensions may operate through broader geopolitical-financing channels, including higher distress risk, weaker debt demand, and tighter credit supply.
Nhưng phải thêm
Unlike Chowdhury et al., who study U.S. firms and a broad geopolitical-risk index, our analysis focuses on a bilateral U.S.–China tension measure and firms located in ASEAN economies.
PHẦN Q. DÙNG VEGA-GUTIÉRREZ ET AL. (2025)
27. Nếu finding khác dấu
Không né tránh.
Viết
Our results differ from Vega-Gutiérrez et al. (2025), who find that EPU is positively associated with leverage in European firms. The difference may reflect the nature of the shock and institutional setting: U.S.–China tension is a bilateral geopolitical and trade-related shock, whereas EPU captures domestic economic-policy uncertainty.
Nếu kết quả dương
Vega-Gutiérrez becomes supporting evidence.
Heterogeneity clue
- institutional quality;
- financial development;
- diversification.
These become natural Bước 4/5 tests.
PHẦN R. DÙNG JIMOH ET AL. (2026)
28. Jimoh không dùng để chứng minh direct leverage effect
Dùng để giải thích:
- ASEAN asymmetry;
- exposure heterogeneity;
- why average coefficient may vary;
- trade-diversion vs vulnerability.
Paragraph mẫu
The ASEAN context also warrants caution in interpreting the average coefficient as uniform across countries. Jimoh et al. (2026) show that the trade effects of U.S.–China tensions differ materially across ASEAN members depending on their pre-existing dependence on the U.S. and China. Accordingly, our baseline coefficient should be interpreted as an average regional association, with potentially substantial heterogeneity explored in subsequent exposure-based tests.
PHẦN S. DÙNG CHAI AND WANG (2026)
29. Dùng cho mechanism discussion
Their evidence:
tension can cause firms to:
- respond strategically;
- reallocate overseas capital;
- manage technology blockade;
- improve supply-chain security.
ASEAN translation
Heightened U.S.–China tensions may induce ASEAN firms to reorganize supply chains and investment locations, thereby altering external financing needs.
Important
Their result may support either direction for leverage:
- reallocation can require more financing;
- uncertainty can lead to deleveraging.
Do not claim their paper predicts one exact leverage sign.
PHẦN T. DÙNG PENG ET AL. (2025)
30. Use for time-varying/asymmetric interpretation
If coefficient is modest or unstable:
Peng et al. (2025) show that the financial effects of UCT are time-varying and asymmetric, suggesting that an average annual coefficient may mask periods in which bilateral tension is especially consequential.
This directly motivates later
- high vs low UCT;
- post-2018;
- COVID;
- nonlinear models;
- quantile/subsample analysis.
PHẦN U. DÙNG WANG ET AL.
31. Use for information environment
If country/institution heterogeneity appears important:
The importance of cross-country heterogeneity is also consistent with evidence that information environments and geographic frictions affect firms' financing choices.
Do not spend a whole paragraph on this paper in baseline results unless directly relevant.
PHẦN V. LITERATURE COMPARISON MATRIX
32. Bảng bắt buộc
| Paper | Main uncertainty/risk | Outcome | Main sign | Current result same/different? | What explains difference? |
|---|---|---|---|---|---|
| Chowdhury et al. (2025) | GPR | Leverage | Negative | ||
| Vega-Gutiérrez et al. (2025) | EPU | Leverage | Positive | ||
| Peng et al. (2025) | UCT | Stock markets | Time-varying/asymmetric | N/A | |
| Chai & Wang (2026) | China–US relations | Overseas allocation | Positive strategic response | N/A | |
| Jimoh et al. (2026) | Trade tension | ASEAN exports | Heterogeneous | N/A | |
| Wang et al. | Information/geography | Leverage | Geography matters | N/A |
PHẦN W. BASELINE RESULTS WRITING TEMPLATE
33. Paragraph 1 — Table introduction
Table X reports the baseline estimates of the relationship between U.S.–China tension and the capital-structure choices of ASEAN firms. The dependent variables capture different dimensions of debt financing, including [list actual measures]. All specifications include the firm-level controls described in Section 3 and firm fixed effects. Because the annual UCT measure varies only over time, the baseline direct-effect models do not include a full set of year fixed effects.
34. Paragraph 2 — Main coefficient
Nếu âm
Across the baseline specifications, the coefficient on UCT is [consistently negative / negative in X of Y models]. In Column (1), where the dependent variable is [DV], the estimated coefficient is [β] and is statistically significant at the [1/5/10]% level. This result indicates that periods of greater U.S.–China tension are associated with lower leverage among ASEAN firms.
Nếu dương
The coefficient on UCT is positive and statistically significant in [columns]. This suggests that ASEAN firms increase their reliance on debt during periods of elevated U.S.–China tension.
Nếu mixed
The effect of UCT differs across capital-structure dimensions. While [DV1] declines, [DV2] shows no statistically significant response, indicating that firms adjust particular components of debt rather than uniformly changing total leverage.
35. Paragraph 3 — Economic significance
The estimated effect is also economically meaningful. A one-standard-deviation increase in UCT is associated with a [X]-percentage-point change in [DV], corresponding to approximately [Y]% of its sample mean.
Must calculate.
No placeholder in final manuscript.
36. Paragraph 4 — Debt composition
The decomposition of debt provides additional insight into the financing response. The UCT coefficient is [more negative / only significant] for short-term leverage, whereas the coefficient for long-term leverage is [smaller / insignificant / opposite in sign]. This pattern suggests that the response is concentrated in refinancing-sensitive debt and is consistent with firms seeking to reduce rollover exposure during periods of elevated bilateral tension.
Replace according to actual results.
37. Paragraph 5 — Literature comparison
The negative leverage response is consistent with Chowdhury et al. (2025), who find that geopolitical risk reduces corporate leverage through both debt-demand and credit-supply channels. However, our evidence differs from Vega-Gutiérrez et al. (2025), who document a positive relationship between EPU and leverage in European firms. The contrast highlights that financing responses may depend on both the nature of the uncertainty shock and the institutional and trade environment in which firms operate.
38. Paragraph 6 — ASEAN interpretation
The ASEAN setting introduces an additional layer of heterogeneity. Jimoh et al. (2026) show that ASEAN members respond differently to U.S.–China trade tensions depending on their prior exposure to the U.S. and Chinese markets. The baseline estimate should therefore be interpreted as an average regional effect rather than a uniform response across countries. We investigate this possibility directly in subsequent exposure-based analyses.
PHẦN X. NẾU KẾT QUẢ MIXED THÌ VIẾT THẾ NÀO?
39. Không cố ép thành “robust negative effect”
Ví dụ:
- Book leverage: negative***
- ST leverage: negative**
- LT leverage: insignificant
- Market leverage: insignificant
Viết đúng
The evidence points to a selective rather than uniform deleveraging response. UCT is negatively associated with book leverage and short-term leverage, while the estimates for long-term and market leverage are statistically weaker. This pattern suggests that firms' immediate response is concentrated in shorter-horizon book debt.
Không viết
UCT significantly reduces all measures of capital structure.
PHẦN Y. SIGN CONSISTENCY TEST
40. Bảng bắt buộc
| Outcome | Sign | p<0.10? | Same direction as H1? | Same as Chowdhury? | Main conclusion |
|---|---|---|---|---|---|
Overall judgment
Choose one:
- Strongly consistent
- Moderately consistent
- Mixed
- No baseline support
- Opposite to H1
This wording is for internal assessment, not necessarily manuscript.
PHẦN Z. ĐỪNG CHỈ NHÌN SIGNIFICANCE
41. Check confidence intervals
If available:
must be considered.
A coefficient:
with wide CI crossing zero substantially is weak evidence even if another specification gives *.
42. Check coefficient stability
Across reasonable baseline columns:
does:
remain similar?
Example
| Model | β |
|---|---|
| Firm controls | −0.010 |
| + Macro controls | −0.009 |
| + Global controls | −0.002 |
If UCT collapses after global controls:
possible omitted global-shock problem.
This becomes a major issue for Bước 3.
PHẦN AA. TIME-SERIES UCT DIAGNOSTICS
43. Because annual UCT is time-only, this is mandatory
Plot or inspect:
- UCT by year;
- leverage mean by year;
- major UCT spikes;
- COVID period;
- 2018–2019 trade-war escalation.
Student questions
- Is coefficient driven by 2018–2020?
- Is COVID doing most of the work?
- Does removing 2020–2021 change coefficient?
- Is there a long time trend shared by UCT and leverage?
- Are only a few UCT spikes driving results?
Do not answer yet by running 30 tests.
Record these as Bước 3 robustness items.
PHẦN AB. MAIN RESULTS AUDIT TABLE
44. Bảng cuối cùng trước khi viết manuscript
| Check | Result | Concern level |
|---|---|---|
| UCT correctly sourced | Low/Medium/High | |
| UCT variation understood | ||
| Year FE identification correct | ||
| DV formulas confirmed | ||
| Lead timing confirmed | ||
| Winsorization correct | ||
| Main coefficient sign stable | ||
| Main coefficient statistically significant | ||
| Economic magnitude meaningful | ||
| Controls broadly plausible | ||
| Results consistent across debt measures | ||
| Global-shock concern | ||
| COVID concern | ||
| Few-cluster concern | ||
| Literature comparison coherent |
Rule
Nếu có:
High Concern
ở:
- UCT identification;
- DV construction;
- FE;
- timing;
phải sửa trước khi chuyển Bước 3.
PHẦN AC. STUDENT WORKFLOW — LÀM THEO ĐÚNG THỨ TỰ
45. Task 1 — Chụp / xuất baseline table
File:
02_BASELINE_RESULTS.xlsx
Copy từ Stata:
- coefficient;
- SE/t;
- p-value;
- N;
- firms;
- R²;
- FE;
- VCE.
46. Task 2 — Điền Regression Audit
Không viết prose trước.
47. Task 3 — Điền Coefficient Interpretation Table
Mỗi DV một dòng.
48. Task 4 — Tính Economic Significance
Tối thiểu:
- +1 SD UCT;
- effect relative to mean DV.
49. Task 5 — Đọc lại 3 paper trước khi viết
Mandatory:
- Chowdhury et al. (2025)
- Vega-Gutiérrez et al. (2025)
- Jimoh et al. (2026)
Optional supporting:
- Peng;
- Chai & Wang;
- Wang et al.
50. Task 6 — Viết Results 1.000–1.500 từ
Structure:
4.1. Baseline Results
Paragraph 1:
- table setup;
- variables;
- FE;
- VCE.
Paragraph 2:
- main UCT coefficient.
Paragraph 3:
- economic significance.
Paragraph 4:
- ST/LT/maturity comparison.
Paragraph 5:
- literature comparison.
Paragraph 6:
- ASEAN interpretation.
PHẦN AD. MẪU HOÀN CHỈNH NẾU KẾT QUẢ ÂM
51. Sample full baseline-results passage
Table X reports the baseline estimates of the association between U.S.–China tension and corporate capital structure among ASEAN firms. The dependent variables capture alternative dimensions of debt financing, including total leverage, short-term leverage, long-term leverage, and debt maturity. All models include the firm-level and macroeconomic controls described in Section 3 and firm fixed effects. Because the annual UCT series is common across firms within a given year, the direct-effect specifications do not include a full set of year fixed effects.
The coefficient on UCT is negative in the main leverage specification. In Column (1), the coefficient equals [β] and is statistically significant at the [X]% level. This finding indicates that higher U.S.–China tension is associated with lower corporate leverage among ASEAN firms. The result is consistent with the financial-flexibility view: when bilateral geopolitical and trade tensions intensify, firms may reduce reliance on debt in order to preserve borrowing capacity and limit exposure to future refinancing and distress risk.
The magnitude of the coefficient is economically meaningful. A one-standard-deviation increase in UCT is associated with a [X]-percentage-point decline in total leverage, equivalent to approximately [Y]% of the sample mean. Thus, the estimated relationship is not only statistically detectable but also economically relevant for firms' financing policies.
The decomposition of leverage provides additional insight. UCT has a [stronger/weaker] negative association with short-term leverage than with long-term leverage. If the effect is concentrated in short-term debt, the result suggests that firms respond to U.S.–China tension by reducing refinancing-sensitive obligations and limiting rollover exposure. If both short- and long-term leverage decline, the evidence instead points to a broader reduction in external debt financing.
These findings are consistent with Chowdhury et al. (2025), who document that geopolitical risk reduces corporate leverage and provide evidence of both lower debt demand and tighter credit supply. Our setting differs, however, in two important respects. First, UCT captures bilateral U.S.–China tensions rather than broad geopolitical risk. Second, our sample consists of ASEAN firms, which are economically integrated with both the U.S. and China and may therefore be simultaneously exposed to supply-chain disruptions and trade-diversion opportunities.
The evidence also differs from Vega-Gutiérrez et al. (2025), who report a positive relation between economic policy uncertainty and leverage among European firms. This contrast supports the view that financing responses to uncertainty depend on the type of shock and institutional environment. In ASEAN, U.S.–China tension can alter export demand, input costs, supply-chain configurations, and access to external finance. Moreover, Jimoh et al. (2026) show that ASEAN economies respond asymmetrically to bilateral trade tensions depending on their pre-existing exposure to the U.S. and China. Our baseline coefficient should therefore be interpreted as an average regional association, with potentially substantial cross-country and exposure-based heterogeneity examined in subsequent analyses.
PHẦN AE. MẪU HOÀN CHỈNH NẾU KẾT QUẢ DƯƠNG
52. Sample passage
Contrary to the baseline prediction, the coefficient on UCT is positive and statistically significant. The result suggests that ASEAN firms increase debt financing during periods of heightened U.S.–China tension. Although this finding differs from the negative geopolitical-risk–leverage relation documented by Chowdhury et al. (2025), it is consistent with evidence that uncertainty does not generate a uniformly negative leverage response. Vega-Gutiérrez et al. (2025), for example, find that economic policy uncertainty is positively related to leverage in European firms.
The ASEAN setting provides several plausible explanations. Heightened U.S.–China tensions can generate trade diversion, production relocation, and new investment opportunities for some ASEAN economies. Firms benefiting from these shifts may require additional external financing to expand production capacity and supply-chain networks. Chai and Wang (2026) likewise show that deteriorating China–U.S. relations can induce firms to undertake strategic overseas capital reallocation in response to technological and supply-chain pressures. Thus, an increase in leverage may reflect firms' financing of strategic adjustment rather than greater financial fragility alone.
This interpretation is particularly relevant given the cross-country asymmetry documented by Jimoh et al. (2026). The positive average coefficient may be driven by firms or countries positioned to benefit from trade diversion, while more vulnerable firms may respond by deleveraging. We therefore examine exposure-based heterogeneity in subsequent analyses.
PHẦN AF. MẪU NẾU BASELINE KHÔNG SIGNIFICANT
53. Sample passage
The baseline coefficient on UCT is not statistically distinguishable from zero. We therefore do not find evidence of a uniform average relationship between U.S.–China tension and [DV] across ASEAN firms. This result should not be interpreted as evidence that bilateral tensions are economically irrelevant. ASEAN economies differ substantially in their trade structures and exposure to the U.S. and China. Jimoh et al. (2026) show that U.S.–China trade tensions generate asymmetric export effects across ASEAN members, suggesting that positive responses in some economies may offset negative responses in others. The absence of a significant average effect therefore motivates exposure-based and country-level analyses rather than a conclusion of no effect.
PHẦN AG. WHAT NOT TO WRITE
54. Không viết
UCT causes lower leverage.
unless identification supports causal language.
55. Không viết
All ASEAN firms reduce leverage.
Baseline coefficient is an average effect.
56. Không viết
Our result confirms all previous studies.
Literature is mixed.
57. Không viết
The coefficient is economically significant because p < 0.01.
Statistical significance ≠ economic significance.
58. Không viết
Short-term leverage coefficient is more negative, therefore it is statistically different from long-term leverage.
Need formal coefficient-difference test if claiming statistically different effects.
59. Không viết
UCT is insignificant, therefore H1 is rejected and tension has no effect.
Use:
- baseline does not support H1;
- no detectable average effect.
PHẦN AH. TEST OF COEFFICIENT DIFFERENCES
60. Nếu muốn nói ST effect mạnh hơn LT effect
Need test:
Do not infer from:
- one significant;
- one insignificant.
Important statistical principle
“Significant” vs “not significant” does not prove coefficients differ.
Student output
| Comparison | β1 | β2 | Difference | p-value |
|---|---|---|---|---|
| ST vs LT | ||||
| Book vs Market |
This can be moved to robustness if implementation is complex.
PHẦN AI. RESULTS EVIDENCE MATRIX
61. Bảng sinh viên phải hoàn thành
| Claim to write | Regression evidence | Paper supporting interpretation | Exact use |
|---|---|---|---|
| UCT associated with lower leverage | Table X Col. 1 | Chowdhury et al. | Main comparison |
| Effect concentrated in ST debt | Table X Col. 2 vs 3 | Bui/Chowdhury/debt maturity literature | Refinancing |
| ASEAN effects may be heterogeneous | Baseline average | Jimoh et al. | Motivate heterogeneity |
| Opposite-sign literature exists | Vega-Gutiérrez et al. | Balanced discussion | |
| UCT effects can be time-varying | Peng et al. | Motivate robustness | |
| Corporate strategic adjustment channel | Chai & Wang | Mechanism | |
| Information environment matters | Wang et al. | Later heterogeneity |
PHẦN AJ. BƯỚC 2 EXCEL STRUCTURE
62. File đề xuất
02_RESULTS_EVIDENCE.xlsx
Sheet 1 — BASELINE_OUTPUT
| Model | DV | UCT coef | SE | t | p | N | Firms | Countries | FE | VCE |
|---|
Sheet 2 — ECON_SIGNIFICANCE
| DV | β | UCT SD | Mean DV | 1SD effect | % mean |
|---|
Sheet 3 — CONTROL_CHECK
| Variable | Expected | Actual | Significant | Comment |
|---|
Sheet 4 — LITERATURE_COMPARE
| Paper | X | Y | Sign | Sample | Similarity | Difference |
|---|
Sheet 5 — WRITING_NOTES
| Paragraph | Main message | Regression evidence | Citation | Draft sentence |
|---|
PHẦN AK. CITATIONS BẮT BUỘC TRONG BASELINE RESULTS
63. Tối thiểu
Results section baseline phải có:
Chowdhury et al. (2025)
→ closest capital-structure comparison.Vega-Gutiérrez et al. (2025)
→ competing uncertainty–leverage evidence.Jimoh et al. (2026)
→ ASEAN heterogeneity.
Depending discussion
Chai and Wang (2026)
→ strategic/supply-chain mechanism.Peng et al. (2025)
→ time-varying UCT effects.
Không cần nhồi tất cả vào một paragraph.
PHẦN AL. REFERENCES USED IN STEP 2
Chowdhury, M. S. R., Aram, M., Javadi, S., & Nejadmalayeri, A. (2025).
Geopolitical risk and corporate capital structure.
Journal of Corporate Finance, 93, 102796.
DOI: 10.1016/j.jcorpfin.2025.102796
Vega-Gutiérrez, P. L., López-Iturriaga, F. J., & Rodríguez-Sanz, J. A. (2025).
Economic policy uncertainty and capital structure in Europe: An agency approach.
The European Journal of Finance, 31(1), 53–75.
DOI: 10.1080/1351847X.2024.2364044
Jimoh, S. O., Goh, S. K., & Tham, S. Y. (2026).
Between giants: The US-China trade war and ASEAN’s evolving role in global trade.
International Economics and Economic Policy, 23(2), 1–24.
DOI: 10.1007/s10368-026-00730-x
Peng, C., Deng, H., Xie, J., & Liu, X. (2025).
US-China tension and stock market performance in US and China: New insights from time-varying quantile causality method.
Finance Research Letters, 85, 107888.
DOI: 10.1016/j.frl.2025.107888
Chai, H., & Wang, Y. (2026).
China–US relations and overseas capital allocation of Chinese firms.
Finance Research Letters, 93, 109621.
DOI: 10.1016/j.frl.2026.109621
Rogers, J. H., Sun, B., & Sun, T. (2024).
U.S.–China Tension.
SSRN Working Paper.
DOI: 10.2139/ssrn.4815838
Rajan, R. G., & Zingales, L. (1995).
What do we know about capital structure? Some evidence from international data.
The Journal of Finance, 50(5), 1421–1460.
Frank, M. Z., & Goyal, V. K. (2009).
Capital structure decisions: Which factors are reliably important?
Financial Management, 38(1), 1–37.
Cameron, A. C., Gelbach, J. B., & Miller, D. L. (2011).
Robust inference with multiway clustering.
Journal of Business & Economic Statistics, 29(2), 238–249.
Cameron, A. C., & Miller, D. L. (2015).
A practitioner's guide to cluster-robust inference.
Journal of Human Resources, 50(2), 317–372.
PHẦN AM. CHECKLIST CUỐI BƯỚC 2
Đã xác nhận exact UCT variable.
Đã xác nhận UCT annual hay fiscal-year aligned.
Đã xác nhận Year FE không absorb main UCT.
Đã map tất cả DVs từ code sang manuscript name.
Đã xác nhận exact formulas của leverage measures.
Đã xác nhận outcome timing t+1.
Đã xác nhận sample ASEAN-X.
Đã xác nhận firm FE.
Đã xác nhận VCE.
Đã ghi số firms.
Đã ghi số observations.
Đã ghi số countries.
Đã ghi số years.
Đã ghi β(UCT) cho từng DV.
Đã ghi SE/t/p cho từng DV.
Đã xác định sign consistency.
Đã tính +1 SD economic effect.
Đã tính effect relative to mean DV.
Đã so ST và LT debt.
Đã diễn giải Debt Maturity nếu có.
Đã kiểm tra signs của controls.
Đã đọc Chowdhury et al. trước khi viết.
Đã đọc Vega-Gutiérrez et al. trước khi viết.
Đã đọc Jimoh et al. trước khi viết.
Đã đối chiếu kết quả với Chowdhury.
Đã trình bày evidence trái dấu từ Vega-Gutiérrez.
Đã giải thích ASEAN heterogeneity bằng Jimoh.
Không gọi association là causality.
Không nói insignificant = no effect.
Không nói significant vs insignificant = statistically different.
Đã hoàn thành Literature Comparison Matrix.
Đã hoàn thành Results Evidence Matrix.
Đã viết Draft Baseline Results 1.000–1.500 từ.
Đã ghi danh sách robustness issues cho Bước 3.
PHẦN AN. CHUYỂN SANG BƯỚC 3
Sau Bước 2, sinh viên phải có một danh sách vấn đề rõ ràng cần kiểm tra.
Ví dụ:
- UCT coefficient có bị driven bởi COVID không?
- Có bị driven bởi 2018–2019 trade-war escalation không?
- Có robust với alternative leverage measures không?
- Có robust với alternative UCT scaling không?
- Có robust với global EPU/GPR controls không?
- Có robust với time trend không?
- Có robust khi loại từng ASEAN country không?
- Có different effect theo U.S./China exposure không?
- Có few-cluster inference issue không?
- Có cần fiscal-year-aligned UCT không?
Các câu hỏi này sẽ trở thành:
BƯỚC 3. ROBUSTNESS, ALTERNATIVE SPECIFICATIONS VÀ IDENTIFICATION CHECKS
BƯỚC 3. ROBUSTNESS, ALTERNATIVE SPECIFICATIONS VÀ IDENTIFICATION CHECKS
Bài mẫu: U.S.–China Tension and Corporate Capital Structure in ASEAN Countries
Mục tiêu đầu ra: Kiểm tra xem baseline finding ở Bước 2 có:
- phụ thuộc vào một cách đo leverage duy nhất hay không;
- phụ thuộc vào một cách đo UCT duy nhất hay không;
- bị chi phối bởi COVID hoặc trade-war escalation hay không;
- chỉ phản ánh global uncertainty/global financial conditions hay không;
- bị chi phối bởi một quốc gia ASEAN cụ thể hay không;
- nhạy với cách xử lý fixed effects và standard errors hay không;
- mạnh hơn ở firms/countries có U.S./China exposure cao hay không;
- đủ đáng tin để chuyển sang Bước 4 về endogeneity/advanced identification hay chưa.
Quy tắc của Bước 3
Robustness không có nghĩa là chạy thật nhiều regressions.
Mỗi robustness test phải trả lời một concern cụ thể.
Sinh viên phải viết được:
Concern → Test → Result → Interpretation
chứ không chỉ:
“We perform several robustness tests.”
PHẦN A. THÀNH PHẨM CUỐI BƯỚC 3
Bước 3 chỉ hoàn thành khi nhóm có đủ:
- Robustness Planning Matrix
- Alternative Capital-Structure Table
- Alternative UCT Table
- Global-Shock Control Table
- Event-Period Sensitivity Table
- Country Sensitivity Table
- Exposure-Based Robustness Table
- Inference / VCE Robustness Table
- Robustness Summary Matrix
- Draft Robustness Section, 1.500–2.200 từ
- Danh sách vấn đề endogeneity còn lại cho Bước 4
PHẦN B. NGUYÊN TẮC ƯU TIÊN
1. Không chạy robustness theo thứ tự ngẫu nhiên
Chạy theo thứ tự:
Priority 1
Identification của UCT
Annual UCT có chỉ time-series variation hay không?
Priority 2
Alternative Capital Structure Measures
Kết quả có phụ thuộc một DV không?
Priority 3
Alternative UCT Construction
Annual, standardized, log, fiscal-year aligned.
Priority 4
Global Time-Shock Controls
Kết quả có chỉ phản ánh:
- Global EPU;
- GPR;
- VIX;
- U.S. interest rates;
- global recession?
Priority 5
Event Sensitivity
- 2018–2019;
- COVID;
- extreme UCT years.
Priority 6
ASEAN Country Sensitivity
Có một nước drive toàn bộ result không?
Priority 7
Exposure-Based Tests
Tác động có mạnh hơn nơi exposure cao không?
Priority 8
Alternative Inference
VCE / small-cluster / time-series concerns.
PHẦN C. ROBUSTNESS PLANNING MATRIX
2. Bảng bắt buộc trước khi chạy
| Concern | Why it matters | Test | Expected evidence if baseline is robust | Priority |
|---|---|---|---|---|
| UCT only varies over time | Year FE cannot be used in annual direct model | Verify unique UCT/year | One value per year confirmed | 1 |
| DV measurement | Leverage can be measured differently | Alternative leverage DVs | Sign broadly consistent | 2 |
| UCT scaling | Index distribution may drive coefficient | Standardized/log/fiscal-year UCT | Similar direction | 3 |
| Global uncertainty | UCT may proxy for GPR/EPU | Add global controls | UCT remains meaningful | 4 |
| COVID | Large common shock | Exclude 2020–2021 | Similar direction | 5 |
| Trade-war escalation | Few years may drive result | Exclude 2018–2019 / event split | Coefficient not entirely event-driven | 5 |
| One ASEAN country | Large country may dominate | Leave-one-country-out | Direction stable | 6 |
| ASEAN heterogeneity | Average effect may hide exposure | UCT × exposure | Stronger where exposure high | 7 |
| SE choice | Effective variation is limited | Alternative VCE | Inference broadly stable | 8 |
PHẦN D. ROBUSTNESS 1 — ALTERNATIVE CAPITAL-STRUCTURE MEASURES
3. Tại sao đây là test đầu tiên?
Capital structure không phải một concept chỉ có một ratio.
Main finding phải được kiểm tra trên:
- total debt;
- short-term debt;
- long-term debt;
- market leverage;
- maturity.
Core reference
Chowdhury et al. (2025)
Paper này dùng:
- market leverage;
- book leverage;
- total debt;
- long-term debt;
để kiểm tra robustness của GPR–capital structure relation.
4. Required alternative outcomes
4.1. Book Leverage
hoặc exact code.
4.2. Short-Term Book Leverage
4.3. Long-Term Book Leverage
4.4. Market Leverage
theo exact code.
4.5. Long-Term Market Leverage
Nếu code có:
phải ghi exact formula.
4.6. Debt Maturity
for:
PHẦN E. ALTERNATIVE CAPITAL-STRUCTURE TABLE
5. Bảng bắt buộc
| Model | DV | β(UCT) | SE | p | Sign | Same as baseline? | Economic meaning |
|---|---|---|---|---|---|---|---|
| 1 | Book leverage | ||||||
| 2 | Market leverage | ||||||
| 3 | ST leverage | ||||||
| 4 | LT leverage | ||||||
| 5 | Debt maturity |
Robustness standard
Không yêu cầu:
tất cả coefficients đều significant.
Yêu cầu:
- direction có logic;
- pattern không contradictory;
- main conclusion không phụ thuộc duy nhất một ratio.
6. Cách viết
Nếu direction consistent
The baseline result is robust to alternative capital-structure measures. UCT remains negatively associated with both book- and market-based leverage measures, suggesting that the finding is not driven by a particular valuation convention.
Nếu chỉ ST debt robust
The effect is concentrated in short-term debt, indicating that U.S.–China tensions primarily affect refinancing-sensitive borrowing rather than firms' long-horizon debt positions.
Nếu book negative nhưng market insignificant
Không gọi result “fully robust”.
Viết:
The negative association is stronger for book-based leverage, while the market-leverage estimate is statistically weaker.
PHẦN F. ROBUSTNESS 2 — ALTERNATIVE UCT SCALING
7. Why test scaling?
UCT may be:
- highly skewed;
- characterized by large spikes;
- difficult to interpret in raw units.
Need to check result does not depend on raw scaling.
8. Standardized UCT
Construct:
Advantage
Coefficient directly means:
effect of a one-standard-deviation increase in UCT.
Expected
Sign should match raw UCT.
9. Log UCT
If:
and distribution is right-skewed:
Purpose
Reduce influence of extreme spikes.
Important
Do not log mechanically.
First check:
- minimum;
- zeros;
- distribution.
10. Change in UCT
Alternative:
Interpretation
Captures:
escalation/de-escalation
rather than level of tension.
Caution
This asks a different economic question.
Use as robustness, not substitute main measure unless theory supports it.
PHẦN G. ASYMMETRIC UCT
11. Motivation from Peng et al. (2025)
Peng et al. show UCT effects on stock markets are:
- time-varying;
- asymmetric.
Therefore test:
Positive tension changes
Negative tension changes
or separate increase/decrease dummies.
Question
Does capital structure respond more strongly to:
- escalation;
- de-escalation?
Do not expect symmetry automatically.
PHẦN H. FISCAL-YEAR-ALIGNED UCT
12. This is a particularly important robustness test
Use monthly UCT.
For each firm:
Why?
Annual calendar UCT may not align with firm fiscal reporting periods.
13. Required diagnostics before regression
| Diagnostic | Result |
|---|---|
| Number of fiscal-year-end months | |
| % December year-end | |
| % June year-end | |
| % March year-end | |
| Within-year SD of FY-UCT | |
| Minimum FY-UCT within year | |
| Maximum FY-UCT within year |
Rule
If:
do not claim FY-aligned UCT gives independent identification with Year FE.
14. FY-aligned model
If enough variation:
If Year FE absorbs most variation
Report this limitation.
Do not force the model.
PHẦN I. ALTERNATIVE UCT TABLE
15. Bảng bắt buộc
| UCT measure | β | SE | p | FE | Interpretation |
|---|---|---|---|---|---|
| Raw annual UCT | Firm FE | Baseline | |||
| Standardized UCT | Firm FE | +1 SD | |||
| Ln(1+UCT) | Firm FE | Log scaling | |||
| ΔUCT | Firm FE | Escalation | |||
| FY-aligned UCT | Firm + Year FE? | Timing robustness |
PHẦN J. ROBUSTNESS 3 — GLOBAL UNCERTAINTY CONTROLS
16. Why this is essential?
Annual UCT is time-series.
Without Year FE, UCT may capture:
- global uncertainty;
- global geopolitical shocks;
- global financial conditions.
This is one of the main weaknesses of the direct-effect model.
17. Add Global EPU
Model:
Question
Does UCT retain explanatory power beyond general policy uncertainty?
18. Add GPR
Why?
Chowdhury et al. show GPR affects leverage.
UCT may simply proxy broader geopolitical tension.
19. Add Global Financial Conditions
Potential controls:
- VIX;
- U.S. policy rate / Fed Funds Rate;
- global credit spread;
- world GDP growth.
Do not add everything at once first.
Run sequentially.
PHẦN K. GLOBAL CONTROL SEQUENCE
20. Recommended sequence
Model 1
Baseline.
Model 2
- Global EPU.
Model 3
- GPR.
Model 4
- VIX.
Model 5
- U.S. interest rate.
Model 6
- selected parsimonious global set.
Why sequential?
To see:
which global variable changes β(UCT).
PHẦN L. GLOBAL-SHOCK TABLE
21. Bảng bắt buộc
| Model | Added global control | β(UCT) | p | % change in β | Concern |
|---|---|---|---|---|---|
| 1 | None | ||||
| 2 | Global EPU | ||||
| 3 | GPR | ||||
| 4 | VIX | ||||
| 5 | U.S. rate | ||||
| 6 | Parsimonious set |
Coefficient change
Interpretation
If UCT coefficient:
- retains sign;
- remains economically meaningful;
- does not collapse,
support for incremental UCT effect strengthens.
PHẦN M. IF UCT COLLAPSES AFTER GPR/EPU
22. Do not hide it
If:
becomes:
- near zero;
- insignificant;
- changes sign
after adding GPR/EPU:
write:
The baseline UCT association is sensitive to broader measures of global uncertainty, suggesting that part of the estimated relation may reflect common geopolitical or policy-uncertainty conditions.
This is valuable evidence.
Do not selectively omit the model.
PHẦN N. ROBUSTNESS 4 — TIME TREND
23. Linear time trend
Because no Year FE in annual direct model, consider:
Model:
Purpose
Check UCT is not simply capturing:
- long-run deleveraging;
- secular leverage trend.
24. Quadratic trend
Optional:
Caution
Too-flexible trends can absorb much UCT variation.
Use only as sensitivity.
PHẦN O. ROBUSTNESS 5 — COVID PERIOD
25. Why COVID is critical?
COVID simultaneously affected:
- U.S.–China relations;
- global trade;
- credit markets;
- firm leverage;
- ASEAN economies.
Annual UCT coefficient may be heavily driven by:
26. Exclude COVID years
Run:
- exclude 2020;
- exclude 2020–2021.
Bảng
| Sample | β(UCT) | p | Sign |
|---|---|---|---|
| Full sample | |||
| Exclude 2020 | |||
| Exclude 2020–2021 |
Interpretation
If stable:
result is not driven solely by COVID.
If collapses:
COVID period is economically important.
Do not call it a failed robustness test automatically.
PHẦN P. ROBUSTNESS 6 — TRADE WAR PERIOD
27. 2018–2019 escalation
UCT may have large spikes around:
- tariff escalation;
- technology restrictions.
Test:
Exclude 2018–2019
Does baseline persist?
Pre/Post 2018
Estimate separately:
and:
Caution
Subsamples have fewer time clusters.
Interpret statistical significance carefully.
28. Why this follows Peng and Jimoh
Peng:
UCT effects are time-varying.
Jimoh:
ASEAN trade response differs around U.S.–China tension escalation.
Therefore event sensitivity is economically motivated.
PHẦN Q. EVENT-SENSITIVITY TABLE
29. Bảng bắt buộc
| Sample | Years | β(UCT) | SE | p | Interpretation |
|---|---|---|---|---|---|
| Full | |||||
| Exclude COVID | |||||
| Exclude 2018–2019 | |||||
| Pre-2018 | |||||
| Post-2018 | |||||
| Exclude top UCT year |
PHẦN R. EXTREME UCT YEAR CHECK
30. Leave-one-high-UCT-year-out
Identify:
- top 1 UCT year;
- top 2 UCT years.
Re-estimate without each.
Purpose
Check whether:
is driven by one extreme annual observation.
This is especially important
because effective UCT variation is:
number of years
not number of firm-year observations.
PHẦN S. ROBUSTNESS 7 — LEAVE-ONE-COUNTRY-OUT
31. Why necessary in ASEAN?
ASEAN sample can be dominated by:
- Malaysia;
- Singapore;
- Thailand;
- Indonesia;
- Vietnam
depending Compustat coverage.
One country may contribute a large share of observations.
32. Procedure
For each country:
- drop country;
- re-estimate baseline;
- record β(UCT).
Bảng
| Excluded country | Remaining countries | β(UCT) | p | Sign change? |
|---|---|---|---|---|
| Indonesia | ||||
| Malaysia | ||||
| Philippines | ||||
| Singapore | ||||
| Thailand | ||||
| Vietnam |
Adjust to actual ASEAN-X sample.
33. Interpretation
Robust
If sign remains broadly stable.
Concern
If dropping one country:
- reverses sign;
- reduces coefficient dramatically.
Then paper should say:
regional result is disproportionately influenced by Country X.
Not hide it.
PHẦN T. COUNTRY WEIGHTING
34. Why consider weighting?
If Singapore has many firms and Brunei has very few:
firm-level regression weights regional result toward countries with more listed firms.
This may be appropriate.
But it is not the same as:
equal ASEAN-country effect.
Sensitivity
Possible:
- unweighted firm-level baseline;
- country-balanced weighting;
- country-year aggregate regression.
Only use if econometrically justified.
PHẦN U. COUNTRY-YEAR AGGREGATED SENSITIVITY
35. Optional but useful
Collapse firm outcomes to:
e.g.:
- mean leverage;
- median leverage.
Then estimate:
Why?
Shows result does not purely arise from thousands of repeated firm observations.
Caution
UCT still only has time variation.
Inference must treat the time dimension carefully.
This is a sensitivity check, not automatically superior.
PHẦN V. ROBUSTNESS 8 — ASEAN EXPOSURE DESIGN
36. This is the most economically informative robustness test
Core source:
Jimoh et al. (2026)
They show ASEAN responses depend on pre-existing U.S./China trade dependence.
37. Pre-Tension U.S. Exposure
Construct before major escalation.
For example:
or:
depending data/theory.
Important
Choose formula before seeing interaction results.
38. Pre-Tension China Exposure
39. Combined Exposure
But
Do not use combined measure only.
US and China exposure may have different economic effects.
PHẦN W. EXPOSURE REGRESSIONS
40. U.S. Exposure
Main UCT effect:
absorbed by Year FE.
41. China Exposure
42. Interpretation
Negative interaction
Higher exposure amplifies deleveraging.
Positive interaction
Highly exposed countries/firms increase leverage more.
Could reflect:
- trade diversion;
- financing of relocation;
- new investment opportunities.
PHẦN X. COUNTRY-EXPOSURE IDENTIFICATION WARNING
43. Only ASEAN-X independent exposure clusters
If:
then exposure varies across only:
Even with thousands of firms:
effective cross-sectional treatment variation is small.
Therefore mandatory
- report number of countries;
- leave-one-country-out;
- alternative exposure measures;
- cautious p-value interpretation.
Do not write
“The interaction is precisely estimated because N = 30,000 firm-years.”
That is misleading.
PHẦN Y. FIRM-LEVEL EXPOSURE
44. Strongest extension if data exist
Potential:
- geographic segment sales;
- customer country;
- supplier country;
- tariff exposure;
- textual U.S./China exposure.
Then:
Model can include
because interaction varies across firms within country-year.
This is much stronger.
45. Firm-level model
Interpretation
Within the same:
- country;
- year;
- macro environment;
more-exposed firms respond differently.
PHẦN Z. EXPOSURE-BASED ROBUSTNESS TABLE
46. Bảng bắt buộc
| Exposure | Level | Pre-determined? | Interaction β | p | FE | Interpretation |
|---|---|---|---|---|---|---|
| U.S. trade exposure | Country | Yes | Firm + Year + Ind×Year | |||
| China trade exposure | Country | Yes | ||||
| Combined exposure | Country | Yes | ||||
| Firm U.S. exposure | Firm | Yes | Firm + Country×Year | |||
| Firm China exposure | Firm | Yes |
PHẦN AA. ROBUSTNESS 9 — ALTERNATIVE SAMPLE RESTRICTIONS
47. Require at least 2 observations per firm
Firm FE needs within-firm variation.
Drop firms with only one usable year.
Report sample change.
48. Require at least 3 observations per firm
More stringent robustness.
Why?
Reduce influence of very short panels.
49. Balanced panel?
Not recommended as main robustness automatically.
Balanced panel can introduce:
survivorship bias.
If used:
state clearly.
Prefer unbalanced panel baseline.
50. Exclude very small firms
If Compustat contains micro/shell firms:
test minimum assets or sales threshold if economically justified.
Do not choose threshold after looking at significance.
PHẦN AB. ROBUSTNESS 10 — ALTERNATIVE INDUSTRY SCREEN
51. Financial firms
Confirm exclusion:
52. Utilities
Confirm exclusion:
Sensitivity
Possible:
- baseline excludes both;
- robustness adds utilities back;
- financial firms remain excluded.
Why?
Utilities may have regulated capital structures.
If result unchanged, industry-screen concern is weaker.
PHẦN AC. ROBUSTNESS 11 — CONTROL SET SENSITIVITY
53. Parsimonious model
UCT + core controls:
- Tangibility;
- Profitability;
- Size;
- Market-to-Book.
54. Expanded firm-control model
Add if justified:
- Cash;
- Sales Growth;
- asset volatility;
- dividend status.
Caution
Do not add potential mediators mechanically.
55. Macro-control model
Add:
- GDP Growth;
- Inflation.
56. Global-control model
Add selected:
- GPR;
- Global EPU;
- VIX;
- U.S. rates.
PHẦN AD. CONTROL-SEQUENCE TABLE
57. Bảng
| Model | Firm controls | Macro controls | Global controls | β(UCT) | p |
|---|---|---|---|---|---|
| 1 | Core | No | No | ||
| 2 | Core | Yes | No | ||
| 3 | Expanded | Yes | No | ||
| 4 | Core | Yes | Global EPU | ||
| 5 | Core | Yes | GPR | ||
| 6 | Core | Yes | Parsimonious global set |
Goal
Show UCT coefficient is not:
- a control-set artifact.
PHẦN AE. ROBUSTNESS 12 — VCE / INFERENCE
58. Annual UCT direct model
Main treatment variation:
Therefore inference cannot be judged only from:
Need to report
- number of years;
- number of firms;
- number of countries.
59. Two-way Firm and Year Clustering
Possible robustness:
Reference:
Cameron, Gelbach, and Miller (2011).
Why?
- serial dependence within firms;
- common shocks within years.
60. Firm clustering only
Can be reported as sensitivity.
But it does not fully address:
- common-year treatment.
Do not present it as obviously superior.
61. Year clustering only
Treatment varies by year.
But number of years can be small.
Use cautiously.
62. Driscoll–Kraay SE
Possible sensitivity for panel data with:
- heteroskedasticity;
- serial correlation;
- cross-sectional dependence.
Reference:
Driscoll and Kraay (1998).
Important
Driscoll–Kraay does not fix:
- omitted variables;
- weak identification;
- Year FE collinearity.
It is an inference robustness check only.
63. Wild Cluster Bootstrap
If using country-level exposure design with few countries:
consider wild-cluster bootstrap.
Reference family:
- Cameron, Gelbach, and Miller;
- small-cluster inference literature.
Critical warning
With:
no bootstrap method magically creates many independent clusters.
Interpret p-values cautiously.
PHẦN AF. INFERENCE ROBUSTNESS TABLE
64. Bảng bắt buộc
| VCE / inference | β(UCT) | SE | p | Main conclusion |
|---|---|---|---|---|
| Baseline VCE | ||||
| Cluster firm | ||||
| Cluster year | ||||
| Two-way firm + year | ||||
| Driscoll–Kraay | ||||
| Other justified method |
Rule
Do not select the VCE with the smallest p-value.
Choose baseline VCE based on:
dependence structure, not significance.
PHẦN AG. ROBUSTNESS 13 — ALTERNATIVE TIMING
65. Contemporaneous
66. One-year forward
Baseline:
67. Two-year forward
Interpretation
Can reveal persistence.
68. Lagged UCT
Economically similar timing to lead-Y form.
PHẦN AH. TIMING TABLE
69. Bảng
| Timing | β(UCT) | p | Interpretation |
|---|---|---|---|
| UCT_t → Lev_t | Contemporaneous | ||
| UCT_t → Lev_t+1 | Baseline | ||
| UCT_t → Lev_t+2 | Persistence | ||
| UCT_t−1 → Lev_t | Lagged shock |
PHẦN AI. ROBUSTNESS 14 — SUBPERIODS
70. Potential periods
Depending sample years:
- pre-2018;
- 2018 onward;
- pre-COVID;
- COVID;
- post-COVID.
Do not over-split
Each split reduces:
- time variation;
- power;
- year clusters.
Only use theoretically important splits.
PHẦN AJ. ROBUSTNESS 15 — HIGH UCT REGIME
71. High Tension Dummy
Define ex ante:
if UCT above:
- median;
- 75th percentile;
- predetermined threshold.
Interaction
or compare high vs low regimes.
Motivation
Peng et al. suggest nonlinearity/time variation.
Caution
Median split is simple but loses information.
Prefer continuous/nonlinear tests if feasible.
PHẦN AK. QUADRATIC UCT
72. Nonlinearity
Estimate:
Question
Does effect become stronger at extreme tension?
Need
- center/standardize UCT;
- interpret marginal effects.
Do not interpret $\beta_1$ alone.
PHẦN AL. ROBUSTNESS SUMMARY MATRIX
73. Bảng cuối cùng
| Robustness family | Main test | Sign preserved? | Statistical evidence | Economic magnitude | Overall judgment |
|---|---|---|---|---|---|
| Alternative leverage | |||||
| Alternative UCT | |||||
| Global controls | |||||
| Time trend | |||||
| COVID | |||||
| Trade-war years | |||||
| Extreme UCT years | |||||
| Leave-one-country-out | |||||
| Exposure design | |||||
| VCE | |||||
| Alternative timing |
Overall judgment choose
- Highly robust
- Broadly robust
- Moderately robust
- Sensitive
- Not robust
Do not choose based only on significance count.
PHẦN AM. HOW TO JUDGE ROBUSTNESS
74. Highly robust
If:
- direction stable;
- magnitude reasonably stable;
- alternative DVs support same economic story;
- global controls do not erase coefficient;
- no single country/year drives result;
- exposure tests support mechanism;
- inference does not depend on one VCE.
75. Moderately robust
If:
- sign stable;
- some specifications insignificant;
- economic interpretation remains coherent.
This is common.
76. Sensitive
If coefficient:
- flips sign often;
- collapses with global controls;
- disappears when one country/year removed.
Then paper needs stronger identification or narrower claim.
PHẦN AN. HOW TO WRITE THE ROBUSTNESS SECTION
77. Recommended structure
5.1. Alternative Measures
- alternative leverage;
- alternative UCT.
5.2. Controlling for Global Conditions
- EPU;
- GPR;
- VIX;
- trend.
5.3. Event and Sample Sensitivity
- COVID;
- 2018–2019;
- leave-one-country-out;
- extreme years.
5.4. Exposure-Based Tests
- U.S. exposure;
- China exposure;
- firm exposure if available.
5.5. Alternative Inference
- VCE;
- DK;
- clustering.
PHẦN AO. SAMPLE WRITING — ALTERNATIVE MEASURES
78. Paragraph template
We first assess whether the baseline finding depends on the measurement of corporate capital structure. Re-estimating the model using [alternative measures] produces coefficients on UCT that remain [negative/positive] and economically comparable to the baseline estimate. The effect is particularly pronounced for [short-term/long-term] debt, suggesting that bilateral tension affects [refinancing-sensitive / longer-horizon] financing decisions rather than merely altering the denominator used to construct leverage.
PHẦN AP. SAMPLE WRITING — GLOBAL CONTROLS
79. Paragraph template
A central concern with the annual UCT specification is that the index varies only over time and may therefore proxy for broader global shocks. To address this issue, we sequentially control for global economic policy uncertainty, geopolitical risk, global financial-market volatility, and U.S. monetary conditions. The UCT coefficient remains [sign/direction] after the inclusion of these controls, indicating that the baseline relationship is not explained solely by general global uncertainty or financial conditions.
If coefficient weakens
The magnitude of the UCT coefficient declines after controlling for [GPR/EPU], suggesting that broader global uncertainty accounts for part, but not all, of the baseline association.
PHẦN AQ. SAMPLE WRITING — COVID
80. Paragraph template
We next examine whether the baseline result is driven by the COVID-19 period, when ASEAN firms simultaneously experienced severe trade, financing, and operating disruptions. Excluding 2020 and 2021 yields a UCT coefficient of [β], compared with [baseline β] in the full sample. The sign and economic magnitude remain [similar/different], indicating that [the result is not solely attributable to COVID / the COVID period is an important component of the estimated relationship].
PHẦN AR. SAMPLE WRITING — LEAVE-ONE-COUNTRY-OUT
81. Paragraph template
Because the number of listed firms differs substantially across ASEAN economies, we examine whether the regional result is disproportionately driven by any single country. We re-estimate the baseline specification after sequentially excluding each country from the sample. The UCT coefficient retains the same sign in [X of Y] estimations and remains within [range] of the full-sample coefficient, suggesting that the baseline result is not attributable to a single ASEAN economy.
PHẦN AS. SAMPLE WRITING — EXPOSURE
82. Paragraph template
We further exploit differences in ASEAN countries' pre-existing exposure to the United States and China. Motivated by Jimoh et al. (2026), we interact UCT with predetermined measures of U.S. and China trade dependence. The interaction coefficient is [sign], indicating that firms in economies with greater [U.S./China] exposure experience [larger/smaller] financing responses when bilateral tensions intensify. This finding supports the view that UCT operates through economically meaningful trade and supply-chain exposure rather than merely capturing a common global time trend.
PHẦN AT. SAMPLE WRITING — ALTERNATIVE VCE
83. Paragraph template
Finally, we examine the sensitivity of statistical inference to alternative treatments of within-firm and common-year dependence. The main coefficient remains [statistically supported / similar in magnitude] under [two-way firm-and-year clustering / Driscoll–Kraay standard errors / other justified method]. Although the relatively limited number of time or country clusters requires caution in interpreting conventional p-values, the consistency of coefficient signs and magnitudes across inference procedures supports the economic interpretation of the results.
PHẦN AU. WHAT COUNTS AS A FAILED ROBUSTNESS TEST?
84. A robustness test is not “failed” merely because p becomes 0.12
Need ask:
- Did sign reverse?
- Did magnitude collapse?
- Did confidence interval widen because sample shrank?
- Did identifying variation become much smaller?
Example
Baseline:
Leave-out sample:
This may still show strong coefficient stability despite weaker precision.
PHẦN AV. WHAT COUNTS AS A SERIOUS PROBLEM?
85. Serious concern examples
Problem 1
baseline but:
after GPR control.
Problem 2
Removing Vietnam changes coefficient from:
to:
Problem 3
Only 2020 drives result.
Problem 4
Annual UCT becomes insignificant once a simple trend is added.
Problem 5
Different leverage measures tell economically incompatible stories.
These must be investigated before Bước 4.
PHẦN AW. DO NOT P-HACK
86. Prohibited workflow
Do not:
- run 40 specifications;
- select only significant ones;
- call them robustness.
Correct workflow
Define:
- concern;
- test;
- expected interpretation
before seeing result.
PHẦN AX. DO NOT CHANGE SAMPLE SILENTLY
87. Every robustness regression must report
- N;
- firms;
- countries;
- years.
If N changes:
explain why.
Example:
Debt maturity sample is smaller because zero-debt firms are excluded.
PHẦN AY. DO NOT CHANGE VARIABLE DEFINITIONS ACROSS TABLES
88. Example
Do not use:
in baseline and:
in robustness without saying so.
Alternative definitions are allowed only if the test explicitly concerns variable measurement.
PHẦN AZ. ROBUSTNESS EVIDENCE MATRIX
89. Bảng bắt buộc
| Concern | Test | Regression evidence | Main result | Citation supporting test | Manuscript sentence |
|---|---|---|---|---|---|
| Alternative leverage | Chowdhury et al. | ||||
| UCT scaling | Peng et al. | ||||
| Global uncertainty | Chowdhury / uncertainty literature | ||||
| COVID | |||||
| ASEAN heterogeneity | Jimoh et al. | ||||
| Supply-chain exposure | Chai & Wang | ||||
| Alternative VCE | Cameron et al. |
PHẦN BA. EXCEL FILE CHO BƯỚC 3
90. File đề xuất
03_ROBUSTNESS_EVIDENCE.xlsx
Sheet 1 — PLAN
| ID | Concern | Test | Priority | Status |
|---|
Sheet 2 — ALT_DV
| DV | β | SE | p | N | Firms | Countries |
|---|
Sheet 3 — ALT_UCT
| UCT measure | β | SE | p | N | FE |
|---|
Sheet 4 — GLOBAL_CONTROLS
| Added control | β(UCT) | p | % change | Comment |
|---|
Sheet 5 — EVENTS
| Sample restriction | β | p | N | Years |
|---|
Sheet 6 — COUNTRY_LOO
| Country excluded | β | p | N |
|---|
Sheet 7 — EXPOSURE
| Exposure | β interaction | p | FE | VCE |
|---|
Sheet 8 — VCE
| Inference method | β | SE | p |
|---|
Sheet 9 — SUMMARY
| Test | Supports baseline? | Notes |
|---|
PHẦN BB. REFERENCES BẮT BUỘC CHO BƯỚC 3
91. Main empirical papers
Chowdhury, M. S. R., Aram, M., Javadi, S., & Nejadmalayeri, A. (2025).
Geopolitical risk and corporate capital structure.
Journal of Corporate Finance, 93, 102796.
DOI: 10.1016/j.jcorpfin.2025.102796
Jimoh, S. O., Goh, S. K., & Tham, S. Y. (2026).
Between giants: The US-China trade war and ASEAN’s evolving role in global trade.
International Economics and Economic Policy, 23(2), 1–24.
DOI: 10.1007/s10368-026-00730-x
Peng, C., Deng, H., Xie, J., & Liu, X. (2025).
US-China tension and stock market performance in US and China: New insights from time-varying quantile causality method.
Finance Research Letters, 85, 107888.
DOI: 10.1016/j.frl.2025.107888
Chai, H., & Wang, Y. (2026).
China–US relations and overseas capital allocation of Chinese firms.
Finance Research Letters, 93, 109621.
DOI: 10.1016/j.frl.2026.109621
Vega-Gutiérrez, P. L., López-Iturriaga, F. J., & Rodríguez-Sanz, J. A. (2025).
Economic policy uncertainty and capital structure in Europe: An agency approach.
The European Journal of Finance, 31(1), 53–75.
DOI: 10.1080/1351847X.2024.2364044
92. Inference references
Cameron, A. C., Gelbach, J. B., & Miller, D. L. (2011).
Robust inference with multiway clustering.
Journal of Business & Economic Statistics, 29(2), 238–249.
Cameron, A. C., & Miller, D. L. (2015).
A practitioner's guide to cluster-robust inference.
Journal of Human Resources, 50(2), 317–372.
Driscoll, J. C., & Kraay, A. C. (1998).
Consistent covariance matrix estimation with spatially dependent panel data.
The Review of Economics and Statistics, 80(4), 549–560.
PHẦN BC. CHECKLIST CUỐI BƯỚC 3
Đã xác định robustness concerns trước khi chạy.
Đã chạy alternative leverage measures.
Đã kiểm tra Book vs Market leverage.
Đã kiểm tra ST vs LT leverage.
Đã kiểm tra Debt Maturity nếu có.
Đã chạy standardized UCT.
Đã kiểm tra log UCT nếu phù hợp.
Đã kiểm tra ΔUCT nếu có economic rationale.
Đã kiểm tra asymmetric UCT nếu phù hợp.
Đã xây fiscal-year-aligned UCT nếu monthly data có sẵn.
Đã kiểm tra within-year variation trước khi dùng Year FE.
Đã thêm Global EPU.
Đã thêm GPR.
Đã thêm VIX/global financial control nếu có.
Đã thêm U.S. interest-rate control nếu relevant.
Đã kiểm tra time trend.
Đã exclude COVID period.
Đã kiểm tra 2018–2019 sensitivity.
Đã kiểm tra top UCT year.
Đã thực hiện leave-one-country-out.
Đã báo N sau mỗi country exclusion.
Đã xây predetermined U.S. exposure nếu có data.
Đã xây predetermined China exposure nếu có data.
Đã tách U.S. và China exposure thay vì chỉ combined.
Đã kiểm tra firm-level exposure nếu available.
Đã kiểm tra alternative sample restrictions.
Đã kiểm tra control-set sensitivity.
Đã kiểm tra alternative timing.
Đã kiểm tra VCE/inference sensitivity.
Đã ghi number of year clusters.
Đã ghi number of country clusters.
Không chọn VCE theo p-value.
Không chọn specifications theo significance.
Đã hoàn thành Robustness Summary Matrix.
Đã hoàn thành Robustness Evidence Matrix.
Đã viết Draft Robustness Section 1.500–2.200 từ.
Đã xác định các endogeneity concerns còn lại cho Bước 4.
PHẦN BD. KHI NÀO ĐƯỢC CHUYỂN SANG BƯỚC 4?
Chuyển sang Bước 4 nếu:
- main sign tương đối ổn định;
- result không chỉ do một DV;
- result không chỉ do COVID;
- result không chỉ do một country;
- UCT không hoàn toàn biến mất sau global controls;
- exposure tests có economic logic;
- inference limitations đã được xác định rõ.
Nếu những điều này chưa đạt:
không nên nhảy ngay sang IV/GMM để “cứu significance”.
Phải quay lại:
- variable construction;
- sample;
- identification;
- theory.
PHẦN BE. CÁC VẤN ĐỀ CHUYỂN SANG BƯỚC 4
Bước 4 sẽ xử lý các vấn đề không thể giải quyết chỉ bằng robustness:
- Reverse causality có thực sự đáng lo không?
- Omitted global shocks còn lại.
- Measurement error của UCT.
- Causal interpretation.
- Exposure endogeneity.
- Dynamic capital-structure adjustment.
- Instrumental variables.
- 2SLS.
- Dynamic GMM.
- Event/exposure identification.
- Alternative quasi-experimental designs.
- Mechanisms:
- debt demand;
- credit supply;
- financial constraints;
- trade diversion;
- supply-chain disruption.
Tên bước tiếp theo:
BƯỚC 4. ENDOGENEITY, IDENTIFICATION VÀ ADVANCED ANALYSES
BƯỚC 4. ENDOGENEITY, IDENTIFICATION VÀ ADVANCED ANALYSES
Bài mẫu: U.S.–China Tension and Corporate Capital Structure in ASEAN Countries
Mục tiêu đầu ra: Sau khi baseline và robustness ở Bước 2–3 đã tương đối ổn định, sinh viên phải đánh giá mức độ có thể tiến từ association sang stronger identification / causal interpretation, đồng thời xây dựng các advanced analyses có logic kinh tế rõ ràng.
Nguyên tắc lớn của Bước 4
Không chạy IV, 2SLS hoặc GMM chỉ vì baseline coefficient chưa đủ significant.
Mỗi advanced method phải trả lời một endogeneity concern cụ thể.
Trình tự ưu tiên của bài này:
- Xác định endogeneity concern thật sự là gì.
- Exposure-based identification.
- Placebo / falsification.
- Event / pre-trend analysis.
- Dynamic capital-structure specification.
- IV/2SLS chỉ khi có instrument credible.
- Mechanism tests.
- Heterogeneity có lý thuyết.
Nếu không có thiết kế đủ mạnh, manuscript phải giữ ngôn ngữ “is associated with” thay vì “causes”.
PHẦN A. THÀNH PHẨM CUỐI BƯỚC 4
Bước 4 chỉ được coi là hoàn thành khi nhóm có đủ:
- Endogeneity Diagnosis Table
- Identification Strategy Ranking
- Exposure-Based Identification Table
- Placebo / Falsification Table
- Pre-Trend / Event Evidence Table nếu có
- Dynamic Capital Structure Table
- IV Candidate Screening Table
- 2SLS Table nếu có instrument đạt yêu cầu
- Mechanism Table
- Heterogeneity Table
- Advanced Results Evidence Matrix
- Draft Endogeneity & Advanced Analyses, 1.800–2.800 từ
- Causal-language decision: paper được phép viết ở mức nào?
PHẦN B. ENDOGENEITY CỦA BÀI NÀY ĐẾN TỪ ĐÂU?
1. Không được viết chung chung
Không viết:
“There may be an endogeneity problem.”
Phải xác định cụ thể concern nào đang tồn tại.
2. Concern 1 — Firm-Level Reverse Causality
Baseline:
Một ASEAN firm riêng lẻ gần như không thể làm U.S.–China Tension Index thay đổi. Vì vậy reverse causality kiểu:
không phải concern lớn nhất của bài này.
Điều cần lo hơn là các aggregate shocks có thể đồng thời làm UCT tăng và thay đổi capital structure của firms.
3. Concern 2 — Omitted Global Shocks
Đây là concern quan trọng nhất của annual UCT.
Annual UCT:
có thể correlated với:
- Global EPU;
- GPR;
- VIX;
- U.S. monetary policy;
- global trade cycle;
- global recession;
- COVID;
- commodity-price shocks.
Do annual direct-effect model không thể dùng full Year FE nếu muốn estimate main UCT, omitted time shocks là vấn đề lớn nhất.
4. Concern 3 — Measurement Error
UCT là textual news-based index nên có thể chịu ảnh hưởng của:
- media attention;
- keyword construction;
- newspaper coverage;
- annual aggregation;
- U.S.-centric news sources.
Measurement error có thể làm coefficient bị attenuated hoặc noisy.
5. Concern 4 — Anticipation
Firms có thể phản ứng trước khi annual UCT fully materializes, ví dụ trước:
- tariff announcements;
- technology restrictions;
- sanctions;
- major diplomatic deterioration.
Vì vậy cần xem monthly UCT, lead/lag structure và event-based design.
6. Concern 5 — Dynamic Leverage
Leverage rất persistent. Firms không điều chỉnh debt ratio tức thời mà có thể tiến dần về target leverage.
Do đó static FE model có thể bỏ qua adjustment dynamics.
PHẦN C. ENDOGENEITY DIAGNOSIS TABLE
| Concern | Serious for this paper? | Why? | Baseline response | Advanced response |
|---|---|---|---|---|
| Firm-level reverse causality | Low/Medium | UCT aggregate | Lead outcome | Exposure/event design |
| Omitted global shocks | High | UCT time-only | Global controls | Exposure + stronger FE |
| Measurement error | Medium | Textual index | Alt UCT | Alternative measure/event data |
| Anticipation | Medium | Firms anticipate shocks | Timing robustness | Monthly/event study |
| Country omitted shocks | High in exposure model | ASEAN heterogeneity | Macro controls | Country×Year FE with firm exposure |
| Exposure endogeneity | Medium/High | Current exposure may react to UCT | Lag exposure | Pre-determined exposure |
| Dynamic leverage | High | Capital structure persistent | Lead Y | Dynamic model / GMM |
PHẦN D. IDENTIFICATION STRATEGY RANKING
7. Xếp hạng theo độ thuyết phục
Strategy 1 — Firm-Level Exposure × UCT
Ưu tiên mạnh nhất nếu có credible firm-level exposure.
Strategy 2 — Predetermined Country Exposure × UCT
Khá hữu ích nhưng bị giới hạn bởi số ASEAN countries nhỏ.
Strategy 3 — Fiscal-Year-Aligned UCT
Cải thiện timing, nhưng không tự động tạo causal identification.
Strategy 4 — Event / Exposure Design
Mạnh nếu event và pre-trends hợp lý.
Strategy 5 — IV / 2SLS
Chỉ dùng khi instrument thực sự đạt relevance và exclusion.
Strategy 6 — Dynamic GMM
Hữu ích cho dynamic leverage và endogenous firm controls, nhưng không tự động làm UCT exogenous.
PHẦN E. STRATEGY 1 — FIRM-LEVEL EXPOSURE × UCT
8. Preferred specification
Giả sử có exposure đo trước tension shock:
Estimate:
Tại sao mạnh?
Country × Year FE hấp thụ:
- GDP;
- inflation;
- exchange rate;
- domestic credit conditions;
- domestic policy;
- country-specific COVID effects;
- common UCT main effect.
Industry × Year FE hấp thụ:
- global sector demand;
- commodity cycles;
- sector credit cycles;
- technology shocks.
Firm FE hấp thụ:
- time-invariant firm heterogeneity.
Identification trở thành:
Trong cùng một country-year, firms có exposure U.S./China cao hơn có điều chỉnh capital structure khác firms ít exposure hay không?
PHẦN F. FIRM EXPOSURE OPTIONS
9. Các exposure có thể sử dụng nếu dữ liệu cho phép
Geographic Sales Exposure
Supply-Chain Exposure
- major customers in U.S.;
- suppliers in China;
- customer/supplier links.
Tariff Exposure
- U.S. tariff exposure;
- China retaliatory tariff exposure;
- HS-product exposure mapped to firms/industries.
Segment Exposure
Compustat geographic segments nếu đủ coverage.
Textual Exposure
Firm filings / earnings calls liên quan:
- China;
- U.S.;
- tariffs;
- export controls;
- supply-chain risk.
Quy tắc
Không gọi một proxy là “firm U.S./China exposure” nếu không có source và economic meaning rõ.
PHẦN G. PREDETERMINED EXPOSURE
10. Exposure phải precede major UCT shocks
Ưu tiên đo:
- trước 2018;
- đầu sample;
- trung bình các năm pre-tension.
Ký hiệu:
Tại sao?
Nếu dùng current exposure:
thì chính UCT có thể đã làm firms thay đổi exposure, tạo reverse causality trong interaction.
PHẦN H. EXPOSURE TABLE
| Exposure | Source | Level | Measurement period | Predetermined? | Coverage | Main concern |
|---|---|---|---|---|---|---|
| U.S. sales share | Firm | |||||
| China sales share | Firm | |||||
| U.S. trade exposure | Country | |||||
| China trade exposure | Country | |||||
| Tariff exposure | Industry/Firm | |||||
| Supply-chain exposure | Firm |
PHẦN I. STRATEGY 2 — COUNTRY EXPOSURE × UCT
11. Model
Main source
Jimoh et al. (2026) cho logic pre-existing ASEAN exposure.
Main benefit
Year FE hấp thụ:
- UCT main effect;
- global shocks;
- common time trend.
Main weakness
Nếu final sample là ASEAN-6 thì chỉ có sáu independent country exposure values.
Do đó p-values phải được diễn giải rất thận trọng.
PHẦN J. TÁCH U.S. VÀ CHINA EXPOSURE
12. Không chỉ dùng combined exposure
Estimate riêng:
và:
Tại sao?
ASEAN firms có thể:
- hưởng lợi từ U.S. trade diversion;
- chịu rủi ro từ China input dependence.
Jimoh et al. cho thấy phản ứng ASEAN bất đối xứng theo exposure.
PHẦN K. STRATEGY 3 — PLACEBO / FALSIFICATION
13. Future UCT placebo
Test:
và:
Mong đợi
Future UCT không nên dự báo mạnh leverage đã được quyết định trước đó.
Nếu significant mạnh
Có thể chỉ ra:
- common trends;
- omitted time shocks;
- serial correlation;
- spurious association.
Caution
UCT có thể persistent, nên future UCT placebo phải được diễn giải cùng autocorrelation của UCT.
PHẦN L. PLACEBO TABLE
| Model | X | Y | β | p | Expected |
|---|---|---|---|---|---|
| Baseline | UCT_t | Lev_t+1 | Main | ||
| Placebo 1 | UCT_t+1 | Lev_t | Weak/insignificant | ||
| Placebo 2 | UCT_t+2 | Lev_t | Weak/insignificant |
PHẦN M. STRATEGY 4 — PRE-TREND / EVENT STUDY
14. Khi nào dùng?
Khi có:
- một major bilateral tension event;
- pre-existing exposure;
- sufficient pre/post years.
Event phải được xác định trước khi nhìn regression significance.
15. Event-time model
Pre-trend condition
Với:
coefficients phải gần zero nếu parallel-trend interpretation hợp lý.
Post coefficients
Cho thấy dynamic financing response sau shock.
PHẦN N. EVENT EVIDENCE TABLE
| Event | Exact date | Source | Why economically relevant? | Exposure | Pre-trend passes? |
|---|---|---|---|---|---|
Rule
Không được tự ghi “2018 trade war” mà không xác định exact event/window và source.
PHẦN O. DIFFERENCE-IN-DIFFERENCES LOGIC
16. Post dummy đơn thuần không phải DiD
Sai:
rồi gọi là DiD.
Cần treated/comparison variation:
hoặc continuous treatment intensity.
Prefer continuous exposure
thường tốt hơn median split vì giữ lại information.
PHẦN P. STRATEGY 5 — DYNAMIC CAPITAL STRUCTURE
17. Leverage persistence
Model:
Interpretation
= leverage persistence.
Approximate speed of adjustment:
Ví dụ nếu:
thì:
xấp xỉ 30% adjustment per period.
PHẦN Q. DYNAMIC FE BIAS
18. Không thêm lagged DV vào FE rồi bỏ qua vấn đề
Lagged leverage + firm FE có thể tạo dynamic-panel bias khi T không lớn.
Do đó nếu dynamic model là trọng tâm, xem xét estimator phù hợp.
PHẦN R. SYSTEM GMM
19. GMM giúp gì?
System GMM có thể giúp với:
- lagged dependent variable;
- endogenous/predetermined firm controls;
- dynamic adjustment.
GMM không tự động giải quyết gì?
GMM không tự động làm aggregate UCT exogenous.
Internal instruments như:
- lagged leverage;
- lagged profitability;
- lagged firm controls
không tạo exogenous variation trong UCT.
Critical rule
System GMM là dynamic-panel method, không phải phép biến UCT thành causal shock.
PHẦN S. GMM DIAGNOSTICS
20. Nếu dùng System GMM, bắt buộc báo
- AR(1)
- AR(2)
- Hansen/Sargan
- number of instruments
- number of groups/firms
- instrument collapse
- lag range
- endogenous/predetermined/exogenous classification.
Instrument proliferation
Không để:
PHẦN T. GMM REPORTING TABLE
| Item | Result | Acceptable? |
|---|---|---|
| Lagged leverage coefficient | ||
| UCT coefficient | ||
| AR(1) p | ||
| AR(2) p | ||
| Hansen p | ||
| Instruments | ||
| Groups | ||
| Instruments < Groups? | ||
| Collapsed instruments? |
PHẦN U. CÁCH DÙNG BUI ET AL. (2026)
21. Closest direct paper
Bui et al. (2026) sử dụng:
- baseline regressions;
- 2SLS;
- dynamic GMM
cho Vietnamese listed firms.
Sinh viên phải đọc trực tiếp phần:
- exact IV construction;
- first stage;
- exclusion argument;
- GMM specification;
- instrument count;
- diagnostic tests.
Không được viết
“Following Bui et al. (2026), we use instrument X”
nếu chưa verify exact instrument trong paper.
PHẦN V. STRATEGY 6 — IV / 2SLS
22. Hai điều kiện cốt lõi
Relevance
Exclusion
Instrument:
không được trực tiếp ảnh hưởng ASEAN firm leverage ngoài kênh UCT.
Quan trọng
Exclusion thường khó bảo vệ hơn relevance.
PHẦN W. NHỮNG BIẾN KHÔNG TỰ ĐỘNG LÀ IV HỢP LỆ
23. Lagged UCT
Không tự động valid vì:
có thể trực tiếp ảnh hưởng:
qua persistent financing effects.
24. GPR
Không phải IV tốt mặc định vì Chowdhury et al. trực tiếp cho thấy:
Exclusion dễ vi phạm.
25. Global EPU
EPU cũng trực tiếp ảnh hưởng corporate financing.
26. VIX
VIX ảnh hưởng trực tiếp financing costs và market conditions.
Kết luận
Các biến trên thường phù hợp làm controls, không phải clean instruments.
PHẦN X. IV CANDIDATE SCREENING TABLE
| Candidate Z | Why correlated with UCT? | Could directly affect ASEAN leverage? | Exclusion credible? | Decision |
|---|---|---|---|---|
| Lagged UCT | Persistence | Yes | Weak | |
| GPR | Global tension | Yes | No/Weak | |
| Global EPU | Uncertainty | Yes | No/Weak | |
| VIX | Risk sentiment | Yes | No | |
| Bilateral political event | ||||
| External diplomatic shock |
Rule
Không chạy 2SLS trước khi bảng này có một candidate thực sự defendable.
PHẦN Y. MỘT IV TỐT HƠN PHẢI TRÔNG NHƯ THẾ NÀO?
27. Concept
Candidate tốt phải:
- làm U.S.–China tension thay đổi;
- external đối với ASEAN firms;
- không trực tiếp làm ASEAN leverage thay đổi ngoài kênh UCT.
Potential research directions có thể là:
- narrowly defined bilateral political events;
- institutional scheduling shocks;
- diplomatic events có timing exogenous.
Nhưng mỗi candidate phải có exclusion argument riêng. Không có instrument mặc định trong hướng dẫn này.
PHẦN Z. 2SLS FIRST STAGE
28. Nếu có credible IV
First stage:
Báo:
- instrument coefficient;
- first-stage F;
- partial R²;
- weak-IV diagnostics.
Cảnh báo rất quan trọng
Nếu UCT và IV đều chỉ time-series thì effective first-stage sample gần với số năm, không phải số firm-years.
PHẦN AA. SECOND STAGE
29. Model
Không được chỉ báo second-stage p-value
Phải báo first stage đầy đủ.
PHẦN AB. WEAK-IV WARNING
30. Không dùng “F > 10” một cách máy móc
Phải xem:
- effective sample dimension;
- clustered inference;
- number of endogenous regressors;
- robust weak-IV diagnostics.
Weak IV có thể làm 2SLS tệ hơn OLS/FE.
PHẦN AC. IV RESULTS TABLE
| Item | First stage | Second stage |
|---|---|---|
| Instrument coefficient | ||
| SE | ||
| First-stage F | ||
| Partial R² | ||
| UCT coefficient | ||
| SE | ||
| p-value | ||
| N | ||
| FE | ||
| VCE |
PHẦN AD. STRATEGY 7 — MECHANISM: DEBT DEMAND
31. Logic
Candidate outcomes
- CAPEX;
- investment;
- acquisitions;
- financing deficit;
- sales growth.
Test 1
Nếu:
đây là evidence consistent với lower financing demand.
Caution
Không tự động gọi đây là formal mediation.
PHẦN AE. FINANCING DEFICIT
32. Nếu dữ liệu cho phép
Conceptual formula:
Exact formula phải lấy từ validated literature.
Question:
Does UCT change firms' external financing need?
PHẦN AF. STRATEGY 8 — CREDIT SUPPLY
33. Logic
Leverage có thể giảm vì lenders tighten debt supply.
Potential evidence:
- bank loan growth;
- lending spread;
- interest expense / debt;
- debt issuance;
- syndicated loan availability;
- country credit growth.
Stronger story
Nếu UCT:
- làm debt quantity giảm;
- đồng thời borrowing cost tăng;
credit-supply tightening interpretation mạnh hơn.
PHẦN AG. COUNTRY CREDIT CONDITIONS
34. Possible proxies
Caution
Country credit growth vẫn phản ánh cả:
- credit supply;
- credit demand.
Không gọi đó là pure supply.
PHẦN AH. STRATEGY 9 — FINANCIAL CONSTRAINTS
35. Motivation
Bui et al. cho evidence effect mạnh hơn ở financially constrained firms.
Possible measures:
- SA Index;
- WW Index;
- Size;
- dividend status;
- credit rating.
Prefer continuous interaction
thay vì split median nếu có thể.
PHẦN AI. SA / WW INDEX RULE
36. Không dùng công thức từ trí nhớ
Sinh viên phải verify:
- exact formula;
- sign direction;
- data inputs;
- original citation.
Không reuse từ project khác nếu variable definitions khác.
PHẦN AJ. STRATEGY 10 — FINANCIAL FLEXIBILITY
37. Cash holdings
Test:
Possible interpretation:
- high cash cushions financing shocks;
- or high cash enables more aggressive deleveraging.
Sign phải được theory dẫn dắt.
38. Existing leverage
Test:
Highly leveraged firms may face stronger:
- distress risk;
- refinancing pressure;
- covenant constraints.
PHẦN AK. STRATEGY 11 — DEBT MATURITY / ROLLOVER CHANNEL
39. Nếu baseline effect tập trung vào ST debt
This suggests rollover-risk management.
Test:
and examine:
Evidence pattern
- ST leverage falls;
- debt maturity rises;
supports shift away from short-term refinancing exposure.
PHẦN AL. STRATEGY 12 — TRADE DIVERSION
40. ASEAN opportunity channel
Particularly relevant for
- export manufacturing;
- sectors receiving relocated FDI;
- countries benefiting from U.S.–China trade diversion.
Main source
Jimoh et al. (2026) and ASEAN fragmentation literature.
PHẦN AM. TRADE-DIVERSION TESTS
41. Candidate proxies
- export growth;
- FDI inflows;
- U.S. market exposure;
- tariff-benefit industry;
- manufacturing exposure.
Positive UCT interaction is economically possible
It may indicate opportunity channel dominates risk channel for selected firms.
PHẦN AN. STRATEGY 13 — SUPPLY-CHAIN DISRUPTION
42. Mechanism
Chai and Wang (2026)
Useful for:
- technological blockade;
- supply-chain security;
- strategic overseas reallocation.
Candidate proxies
- China input dependence;
- imported intermediate inputs;
- supplier exposure;
- industry import dependence;
- inventory changes.
PHẦN AO. MECHANISM TABLE
| Mechanism | Proxy / outcome | Expected UCT effect | Regression evidence | Interpretation |
|---|---|---|---|---|
| Debt demand | Investment | − | ||
| Debt demand | Financing deficit | |||
| Credit supply | Credit growth | − | ||
| Credit supply | Borrowing cost | + | ||
| Financial constraints | SA/WW interaction | Stronger | ||
| Financial flexibility | Cash interaction | |||
| Rollover | ST debt / maturity | |||
| Trade diversion | Export/FDI exposure | Opportunity + | ||
| Supply chain | China-input exposure | Risk − |
PHẦN AP. FORMAL MEDIATION — DÙNG CẨN THẬN
43. Không gọi mọi channel test là mediation
Safer language:
- mechanism-consistent evidence;
- channel evidence;
- cross-sectional evidence.
Chỉ dùng formal mediation nếu assumptions và method thực sự phù hợp.
PHẦN AQ. HETEROGENEITY PRIORITY
44. Thứ tự đề xuất
- U.S. exposure
- China exposure
- Financial constraints
- Cash holdings
- Existing leverage / short-term debt dependence
- Institutional quality
- Financial development
- Industry exposure
Không cần chạy tất cả.
PHẦN AR. INSTITUTIONAL QUALITY HETEROGENEITY
45. Motivation
Vega-Gutiérrez et al. cho thấy institutional quality và financial development moderate uncertainty–capital structure relation.
ASEAN test:
or:
Prefer structural/pre-period measure
nếu mục tiêu là long-run institutional environment.
Nếu IQ là country-year
Interaction biến thiên country-year, nên không dùng Country × Year FE vì sẽ absorb interaction.
PHẦN AS. HETEROGENEITY TABLE
| Moderator | Level | Interaction β | p | Predicted sign | Supported? |
|---|---|---|---|---|---|
| U.S. exposure | Country/Firm | ||||
| China exposure | Country/Firm | ||||
| Financial constraints | Firm | ||||
| Cash | Firm | ||||
| Existing leverage | Firm | ||||
| Institutional quality | Country | ||||
| Financial development | Country | ||||
| Industry exposure | Industry |
PHẦN AT. INTERACTION INTERPRETATION
46. Không chỉ đọc β interaction
Với:
marginal effect:
Báo ít nhất
- P25 exposure;
- P50;
- P75.
PHẦN AU. MARGINAL EFFECT TABLE
| Exposure level | Marginal effect of UCT | SE | p |
|---|---|---|---|
| P25 | |||
| P50 | |||
| P75 |
PHẦN AV. MULTIPLE HYPOTHESIS PROBLEM
47. Nếu chạy quá nhiều heterogeneity tests
Risk:
- false positives;
- data mining.
Good practice:
- pre-specify a few theoretically motivated moderators;
- treat additional tests as exploratory;
- consider multiple-testing adjustments where appropriate.
PHẦN AW. ADVANCED RESULTS EVIDENCE MATRIX
| Claim | Model/Table | Evidence | Supporting paper | Strength |
|---|---|---|---|---|
| UCT effect stronger with China exposure | Jimoh / Chai-Wang | |||
| UCT affects debt demand | Chowdhury | |||
| Credit supply channel | Chowdhury / credit literature | |||
| Constrained firms react more | Bui et al. | |||
| Institutions moderate effect | Vega-Gutiérrez | |||
| Dynamic leverage response | Capital-structure literature |
Strength:
- Strong
- Moderate
- Suggestive
- Exploratory
PHẦN AX. SAMPLE WRITING — ENDOGENEITY MOTIVATION
48. Paragraph template
Although the baseline regressions establish a robust association between U.S.–China tension and corporate capital structure, several identification concerns remain. The most important concern is not conventional firm-level reverse causality, since the leverage decision of an individual ASEAN firm is unlikely to affect aggregate U.S.–China tensions, but rather omitted global shocks that may be correlated with both UCT and corporate financing conditions. This concern is particularly relevant because the annual UCT index varies primarily over time and therefore cannot be combined with a full set of year fixed effects in the direct-effect specification.
PHẦN AY. SAMPLE WRITING — EXPOSURE IDENTIFICATION
49. Paragraph template
To obtain stronger identification, we exploit cross-sectional differences in firms' pre-existing economic exposure to the United States and China. We interact UCT with predetermined exposure measures and include firm and year fixed effects. Where firm-level exposure is available, we further include country-by-year and industry-by-year fixed effects, so that identification comes from differences in exposure among firms facing the same country-level macroeconomic environment and the same global industry conditions. This design substantially reduces the concern that the estimated effect merely captures common time shocks.
PHẦN AZ. SAMPLE WRITING — PLACEBO
50. Paragraph template
We conduct falsification tests using future values of UCT. If the baseline relationship reflects a genuine temporal response of capital structure to bilateral tension, future tension should not systematically explain leverage choices made before that tension is realized. The coefficients on lead UCT are [small/insignificant], providing evidence against a purely spurious common-trend interpretation.
PHẦN BA. SAMPLE WRITING — DYNAMIC MODEL
51. Paragraph template
Corporate leverage is highly persistent and may adjust gradually toward a target. We therefore estimate dynamic specifications that include lagged leverage. Because the inclusion of a lagged dependent variable in a fixed-effects panel can generate dynamic-panel bias, we treat the dynamic estimates as a separate specification and use an appropriate dynamic-panel estimator where necessary. Importantly, the dynamic estimator addresses persistence in leverage and the endogeneity of selected firm-level controls; it does not by itself establish the exogeneity of the aggregate UCT measure.
PHẦN BB. SAMPLE WRITING — IV
52. Chỉ dùng nếu instrument credible
We additionally estimate an instrumental-variable specification using [INSTRUMENT], which is expected to affect U.S.–China tension but has no direct effect on ASEAN firms' capital structure conditional on the included controls. The first-stage coefficient is [ ], with a weak-instrument statistic of [ ]. The second-stage coefficient on predicted UCT is [ ]. The validity of this specification therefore depends critically on the exclusion restriction that [INSTRUMENT] affects ASEAN financing decisions only through bilateral U.S.–China tension.
Nếu không defend được câu cuối
Không dùng IV.
PHẦN BC. SAMPLE WRITING — DEBT DEMAND
53. Paragraph
The decline in leverage may reflect lower demand for external financing if heightened bilateral tension causes firms to postpone investment. Consistent with this mechanism, UCT is associated with [lower investment / lower capital expenditures], and the leverage response is stronger among firms with [relevant characteristic]. We interpret this evidence as consistent with a debt-demand channel rather than as formal causal mediation.
PHẦN BD. SAMPLE WRITING — CREDIT SUPPLY
54. Paragraph
We also examine whether tighter credit conditions contribute to the leverage response. UCT is associated with [lower credit growth / higher borrowing costs / reduced short-term debt availability], suggesting that external financing supply becomes less favorable during periods of heightened bilateral tension. This interpretation is consistent with the credit-supply channel documented in the geopolitical-risk capital-structure literature.
PHẦN BE. SAMPLE WRITING — TRADE DIVERSION
55. Paragraph
The financing response is not uniformly negative across ASEAN. Firms with greater exposure to trade-diversion opportunities exhibit a [less negative/more positive] leverage response when UCT rises. This pattern is consistent with evidence that U.S.–China trade tensions can redirect production, exports, and investment toward selected ASEAN economies. For these firms, higher tension may create expansion opportunities that increase external financing needs even as uncertainty rises.
PHẦN BF. CAUSAL-LANGUAGE DECISION
56. Level 1 — Association only
Nếu chỉ có:
- FE baseline;
- robustness.
Allowed:
- associated with;
- related to;
- linked to.
57. Level 2 — Stronger identification
Nếu có:
- predetermined exposure;
- strong FE;
- placebo/pre-trend;
- robust inference.
Có thể viết:
- evidence is consistent with an effect;
- suggests that UCT influences;
- stronger evidence that…
Vẫn tránh absolute causality.
58. Level 3 — Causal claim
Chỉ khi design plausibly establishes:
- treatment exogeneity;
- valid parallel trends hoặc valid IV;
- no obvious alternative channel;
- robust inference.
PHẦN BG. THỨ TỰ ADVANCED ANALYSES CHO SINH VIÊN
59. Recommended order
Analysis 1
Predetermined firm/country exposure interaction.
Analysis 2
Placebo lead UCT.
Analysis 3
Pre-trend/event test nếu feasible.
Analysis 4
Financial-constraint heterogeneity.
Analysis 5
ST vs LT debt / rollover mechanism.
Analysis 6
Investment / debt-demand mechanism.
Analysis 7
Credit-supply proxy.
Analysis 8
Dynamic leverage.
Analysis 9
IV/2SLS chỉ nếu có credible instrument.
PHẦN BH. MINIMUM STRONG PACKAGE
60. Một paper tốt không cần chạy tất cả
Minimum đề xuất:
- Predetermined exposure interaction
- Placebo lead
- Financial constraints
- ST vs LT debt / maturity
- Investment hoặc credit-supply mechanism
- Dynamic leverage robustness
IV là optional, không phải bắt buộc.
PHẦN BI. WHY IV IS OPTIONAL
61. Weak IV có thể làm paper tệ hơn
Reviewer thường chấp nhận hơn một paper viết:
“We cannot claim full causality, but exposure and falsification tests strengthen identification.”
so với một paper dùng instrument khó bảo vệ rồi claim causality.
Quality order
Credible design > fancy estimator.
PHẦN BJ. METHOD DECISION MATRIX
| Method | Problem addressed | Appropriate here? | Required condition | Main limitation |
|---|---|---|---|---|
| Firm FE | Time-invariant heterogeneity | Yes | Within-firm variation | Omitted time shocks |
| Exposure × UCT | Common time shocks | Strong | Predetermined exposure | Exposure measurement |
| Country×Year FE | Country-year shocks | With firm exposure | Firm-level interaction variation | Cannot use with country exposure |
| Event study | Dynamic shock response | If event valid | Pre-trends | Event confounding |
| 2SLS | Endogenous UCT | Only with valid IV | Relevance + exclusion | Hard IV |
| System GMM | Dynamic leverage | Yes for persistence | Valid internal instruments | Instrument proliferation |
| Driscoll–Kraay | Cross-sectional dependence | Sensitivity | Sufficient T | Inference only |
PHẦN BK. EXCEL FILE CHO BƯỚC 4
62. File đề xuất
04_ADVANCED_EVIDENCE.xlsx
Sheet 1 — ENDOGENEITY
| Concern | Severity | Test | Result | Remaining concern |
|---|
Sheet 2 — EXPOSURE
| Exposure | Level | Pre-period | β interaction | SE | p | FE | VCE |
|---|
Sheet 3 — PLACEBO
| Test | β | p | Pass? |
|---|
Sheet 4 — EVENT
| Event time | β | SE | p | Pre/Post |
|---|
Sheet 5 — DYNAMIC
| Model | Lag Lev β | UCT β | AR1 | AR2 | Hansen | Instruments | Groups |
|---|
Sheet 6 — IV
| Instrument | First-stage β | F | Partial R² | 2SLS β | p | Exclusion note |
|---|
Sheet 7 — MECHANISMS
| Mechanism | Proxy | β(UCT) / interaction | p | Interpretation |
|---|
Sheet 8 — HETEROGENEITY
| Moderator | β interaction | p | Marginal effect low | Marginal effect high |
|---|
Sheet 9 — CLAIMS
| Manuscript claim | Evidence | Strength | Citation |
|---|
PHẦN BL. REFERENCES CẦN DÙNG
63. Closest empirical references
Chowdhury, M. S. R., Aram, M., Javadi, S., & Nejadmalayeri, A. (2025).
Geopolitical risk and corporate capital structure.
Journal of Corporate Finance, 93, 102796.
DOI: 10.1016/j.jcorpfin.2025.102796
Use for:
- endogeneity logic;
- debt demand;
- credit supply;
- leverage outcomes;
- heterogeneity.
Bui Tien Thinh, Le Thi My Nuong, Ha Thi Thuy, & Dao Nhat Minh. (2026).
Corporate leverage and U.S.-China tensions: Evidence from Vietnam.
Cogent Economics & Finance, 14(1), 2613610.
DOI: 10.1080/23322039.2026.2613610
Use for:
- closest UCT–leverage evidence;
- 2SLS/GMM precedent;
- financial constraints;
- Vietnam benchmark.
Important: verify exact instrument and exact GMM implementation before using.
Jimoh, S. O., Goh, S. K., & Tham, S. Y. (2026).
Between giants: The US-China trade war and ASEAN’s evolving role in global trade.
International Economics and Economic Policy, 23(2), 1–24.
DOI: 10.1007/s10368-026-00730-x
Use for:
- predetermined ASEAN exposure;
- U.S./China asymmetry;
- trade-diversion channel.
Chai, H., & Wang, Y. (2026).
China–US relations and overseas capital allocation of Chinese firms.
Finance Research Letters, 93, 109621.
DOI: 10.1016/j.frl.2026.109621
Use for:
- strategic reallocation;
- technological blockade;
- supply-chain channel.
Vega-Gutiérrez, P. L., López-Iturriaga, F. J., & Rodríguez-Sanz, J. A. (2025).
Economic policy uncertainty and capital structure in Europe: An agency approach.
The European Journal of Finance, 31(1), 53–75.
DOI: 10.1080/1351847X.2024.2364044
Use for:
- institutional heterogeneity;
- financial development;
- competing leverage response.
PHẦN BM. DYNAMIC / ECONOMETRIC REFERENCES
64. Dynamic capital structure
Students should add exact references used for:
- partial adjustment;
- target leverage;
- dynamic panel estimation.
At minimum review:
- Lemmon, Roberts, and Zender (2008);
- relevant partial-adjustment literature;
- Arellano–Bond / Blundell–Bond if GMM is used.
Arellano, M., & Bond, S. (1991).
Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations.
Review of Economic Studies, 58(2), 277–297.
Blundell, R., & Bond, S. (1998).
Initial conditions and moment restrictions in dynamic panel data models.
Journal of Econometrics, 87(1), 115–143.
Use only if the actual estimator corresponds to these frameworks.
PHẦN BN. INFERENCE REFERENCES
Cameron, A. C., Gelbach, J. B., & Miller, D. L. (2011).
Robust inference with multiway clustering.
Journal of Business & Economic Statistics, 29(2), 238–249.
Cameron, A. C., & Miller, D. L. (2015).
A practitioner's guide to cluster-robust inference.
Journal of Human Resources, 50(2), 317–372.
PHẦN BO. CHECKLIST CUỐI BƯỚC 4
Đã xác định endogeneity concern cụ thể.
Không viết reverse causality chung chung.
Đã nhận diện omitted global shocks là concern lớn.
Đã hiểu annual UCT chỉ có time variation.
Đã xếp hạng identification strategies.
Đã xây predetermined U.S. exposure nếu có data.
Đã xây predetermined China exposure nếu có data.
Đã tách U.S. và China exposure.
Đã ưu tiên firm-level exposure nếu có credible source.
Chỉ dùng Country × Year FE khi interaction còn firm-level variation.
Đã chạy placebo lead UCT.
Đã kiểm tra UCT persistence khi đọc placebo.
Nếu dùng event study: đã xác định event ex ante.
Nếu dùng event study: đã kiểm tra pre-trends.
Nếu dùng DiD: đã có treated/comparison variation.
Không gọi Post dummy đơn thuần là DiD.
Đã kiểm tra dynamic leverage.
Đã hiểu lagged DV + FE có dynamic bias.
Nếu dùng GMM: đã báo AR(1).
Nếu dùng GMM: đã báo AR(2).
Nếu dùng GMM: đã báo Hansen/Sargan.
Nếu dùng GMM: đã báo instrument count.
Nếu dùng GMM: đã kiểm tra instrument proliferation.
Không dùng GMM để “instrument UCT” một cách máy móc.
Đã đọc exact 2SLS/GMM của Bui et al. trước khi áp dụng.
Đã tạo IV Candidate Screening Table.
Không dùng GPR làm IV cho UCT một cách máy móc.
Không dùng EPU làm IV cho UCT một cách máy móc.
Không dùng VIX làm IV cho UCT một cách máy móc.
Nếu có IV: đã giải thích relevance.
Nếu có IV: đã giải thích exclusion restriction.
Nếu có IV: đã báo first-stage F.
Nếu có IV: đã báo weak-IV diagnostics.
Đã kiểm tra debt-demand mechanism.
Đã kiểm tra investment channel nếu có data.
Đã kiểm tra credit-supply channel nếu có data.
Đã kiểm tra financial constraints.
Đã kiểm tra cash/financial flexibility nếu phù hợp.
Đã kiểm tra ST debt / rollover channel.
Đã kiểm tra trade-diversion channel.
Đã kiểm tra supply-chain channel.
Đã hạn chế số heterogeneity tests theo theory.
Đã tính marginal effects cho interactions.
Đã hoàn thành Advanced Results Evidence Matrix.
Đã quyết định causal-language level.
Đã viết Draft Endogeneity & Advanced Analyses 1.800–2.800 từ.
PHẦN BP. CẤU TRÚC ĐỀ XUẤT CHO MANUSCRIPT
65. Section 5.1 — Identification Concerns
Khoảng 250–400 từ:
- annual UCT;
- omitted time shocks;
- why firm-level reverse causality is less central.
66. Section 5.2 — Exposure-Based Identification
Khoảng 400–600 từ:
- U.S. exposure;
- China exposure;
- FE;
- results.
67. Section 5.3 — Falsification and Dynamic Tests
Khoảng 300–500 từ:
- placebo leads;
- pre-trends/event if available;
- dynamic leverage.
68. Section 5.4 — Mechanisms
Khoảng 500–800 từ:
- investment/debt demand;
- credit supply;
- financial constraints;
- rollover;
- trade diversion;
- supply chains.
69. Section 5.5 — Additional Heterogeneity
Khoảng 300–500 từ:
- institutions;
- financial development;
- firm characteristics.
PHẦN BQ. KHI NÀO BƯỚC 4 ĐƯỢC COI LÀ MẠNH?
Một package thuyết phục có thể là:
Main result
UCT associated with leverage.
Stronger identification
với:
- Firm FE;
- Country × Year FE;
- Industry × Year FE.
Falsification
Future UCT không dự báo past leverage.
Mechanisms
- exposed firms cut investment;
- constrained firms cut debt more;
- ST debt response stronger.
Heterogeneity
U.S. và China exposure khác nhau.
Dynamic robustness
Leverage persistence được xử lý.
Package này thường thuyết phục hơn một 2SLS với weak/exclusion-problem instrument.
PHẦN BR. CHUYỂN SANG BƯỚC 5
Sau Bước 4, sinh viên đã biết:
- baseline finding;
- robustness;
- identification strength;
- mechanisms;
- heterogeneity.
Bước tiếp theo không phải tiếp tục chạy regression vô hạn.
Bước 5 sẽ chuyển sang:
XÂY DỰNG THEORY, LITERATURE REVIEW VÀ HYPOTHESES DỰA TRÊN KẾT QUẢ ĐÃ CÓ
Mục tiêu:
- tổ chức literature;
- xây theoretical channels;
- viết H1/H2/H3;
- xác định contribution;
- tránh viết theory không khớp empirical evidence.
BƯỚC 5. XÂY DỰNG THEORY, LITERATURE REVIEW VÀ HYPOTHESES
Bài mẫu: U.S.–China Tension and Corporate Capital Structure in ASEAN Countries
Mục tiêu đầu ra: Từ các bằng chứng đã có ở Bước 1–4, sinh viên phải xây được một phần Literature Review and Hypothesis Development có logic, có khoảng trống nghiên cứu rõ, và dẫn trực tiếp đến empirical design.
Nguyên tắc quan trọng
Bước 5 được thực hiện sau khi nhóm đã hiểu dữ liệu và kết quả, nhưng không được viết theory chỉ để khớp coefficient đã thấy.
Phải phân biệt:
- Ex ante theory: literature cho phép dự đoán điều gì trước khi nhìn kết quả?
- Ex post interpretation: kết quả thực tế phù hợp channel nào?
Nếu literature cho phép nhiều predictions khác nhau, phải trình bày competing channels một cách trung thực.
Không được viết:
“Because our coefficient is negative, theory predicts a negative effect.”
Phải viết:
“Theory provides several competing channels. On balance, we formulate H1 based on the channel we consider economically dominant, while later empirical results determine whether that prediction is supported.”
PHẦN A. THÀNH PHẨM CUỐI BƯỚC 5
Bước 5 chỉ hoàn thành khi nhóm có đủ:
- Research Story Map
- Literature Stream Matrix
- Theory Channel Matrix
- Research Gap Table
- Contribution Matrix
- Hypothesis Evidence Table
- Draft Literature Review, 1.800–2.500 từ
- Draft Hypothesis Development, 1.200–1.800 từ
- Final hypotheses H1–H4/H5
- Danh sách citation đã kiểm chứng
- Bridge paragraph sang empirical section
PHẦN B. INPUT BẮT BUỘC TỪ BƯỚC 1–4
1. Không bắt đầu Literature Review từ trang trắng
Trước khi viết, sinh viên phải lấy lại:
Từ Bước 1
- exact UCT definition;
- exact ASEAN sample;
- capital-structure measures;
- main controls;
- closest papers.
Từ Bước 2
- baseline sign;
- economic magnitude;
- short-term vs long-term pattern;
- closest literature comparison.
Từ Bước 3
- robustness pattern;
- global-shock sensitivity;
- COVID sensitivity;
- country sensitivity;
- exposure results.
Từ Bước 4
- strongest identification design;
- mechanisms;
- heterogeneity;
- causal-language level.
PHẦN C. RESEARCH STORY MAP
2. Sinh viên phải tóm bài trong 6 dòng trước khi viết
Điền bảng:
| Câu hỏi | Trả lời |
|---|---|
| Shock là gì? | U.S.–China Tension |
| Firms ở đâu? | ASEAN-X |
| Outcome là gì? | Corporate Capital Structure |
| Main mechanism là gì? | |
| Vì sao ASEAN đặc biệt? | |
| Contribution chính là gì? |
Nếu chưa điền được 6 dòng này
Chưa nên viết literature review.
PHẦN D. CẤU TRÚC LITERATURE REVIEW
3. Không tổ chức theo kiểu “Paper A nói…, Paper B nói…”
Literature Review nên được tổ chức theo 5 streams:
Stream 1
U.S.–China Tension: Measurement and Economic Consequences
Stream 2
Uncertainty / Geopolitical Risk and Corporate Financing
Stream 3
Capital Structure Theory
Stream 4
ASEAN Exposure to U.S.–China Tensions
Stream 5
Heterogeneity: Financial Constraints, Institutions and Exposure
PHẦN E. STREAM 1 — U.S.–CHINA TENSION
4. Core papers
Rogers, Sun, and Sun (2024)
Dùng cho:
- definition of UCT;
- textual construction;
- time-series properties;
- bilateral-tension concept.
Peng et al. (2025)
Dùng cho:
- financial-market relevance;
- time-varying effects;
- asymmetric effects.
Chai and Wang (2026)
Dùng cho:
- corporate strategic response;
- technological restrictions;
- supply-chain adaptation;
- overseas capital reallocation.
Không được nhầm
Chai and Wang:
không sử dụng Rogers UCT làm main relation measure.
PHẦN F. CÁCH VIẾT STREAM 1
5. Mục tiêu
Không cần mô tả UCT quá dài.
Phải dẫn đến câu hỏi:
Nếu tension đã ảnh hưởng asset prices, investment, supply chains và corporate strategic decisions, nó có ảnh hưởng financing policy hay không?
Draft mẫu
A growing literature examines the economic consequences of deteriorating U.S.–China relations. Rogers et al. (2024) develop a news-based U.S.–China Tension Index that captures the intensity of contentious bilateral relations across trade, technology, investment, and geopolitical dimensions. Recent evidence suggests that the index contains economically relevant information for financial markets and corporate decisions. Peng et al. (2025) document time-varying and asymmetric stock-market responses to U.S.–China tensions, while Chai and Wang (2026) show that worsening bilateral relations alter firms' overseas capital-allocation decisions through technological and supply-chain channels. Together, these studies indicate that U.S.–China tensions affect both market valuations and real corporate behavior. However, comparatively little is known about how such tensions influence firms' financing structures, particularly outside the United States and China.
PHẦN G. STREAM 2 — UNCERTAINTY / GEOPOLITICAL RISK AND CAPITAL STRUCTURE
6. Core papers
Chowdhury et al. (2025)
Main reference.
Finds:
Mechanisms:
- debt demand;
- credit supply;
- financial flexibility.
Vega-Gutiérrez et al. (2025)
Finds:
in their European setting.
Moderators include:
- institutional quality;
- financial development;
- diversification.
Bui et al. (2026)
Closest direct UCT–leverage evidence for Vietnam.
Finds:
- lower overall leverage;
- lower short-term leverage;
- stronger effects for constrained firms / selected conditions.
PHẦN H. WHY MIXED LITERATURE IS USEFUL
7. Không được coi trái dấu là vấn đề
Mixed literature giúp xây theory tốt hơn.
Có ít nhất hai broad channels:
Channel A — Risk / Flexibility Channel
Predicts:
Channel B — Financing Need / Agency Channel
Can predict:
Therefore
The sign is theoretically nontrivial.
This makes ASEAN setting more interesting.
PHẦN I. DRAFT STREAM 2
8. Đoạn mẫu
The broader uncertainty literature provides competing predictions for corporate financing. Heightened geopolitical risk can increase expected distress costs, reduce credit supply, and increase the value of maintaining financial flexibility, thereby reducing leverage. Consistent with this mechanism, Chowdhury et al. (2025) document a persistent negative relation between geopolitical risk and corporate leverage. The closest direct evidence for U.S.–China tensions is provided by Bui et al. (2026), who find lower overall and short-term leverage among Vietnamese listed firms during periods of elevated bilateral tension. However, uncertainty does not uniformly reduce debt financing. Vega-Gutiérrez et al. (2025) report a positive association between economic policy uncertainty and leverage in European firms and show that institutional and financial conditions shape this relationship. These findings suggest that the financing response to uncertainty depends on both the nature of the shock and the economic environment in which firms operate.
PHẦN J. STREAM 3 — CAPITAL STRUCTURE THEORY
9. Không viết lại cả giáo trình capital structure
Chỉ dùng theory cần thiết để giải thích UCT.
Core foundations:
- Modigliani and Miller;
- trade-off theory;
- pecking-order theory;
- agency theory;
- financial flexibility;
- debt maturity / rollover risk.
PHẦN K. TRADE-OFF THEORY
10. Logic
Firms balance:
against:
UCT may increase:
- cash-flow volatility;
- uncertainty of export demand;
- input-price risk;
- financing costs;
- default risk.
Thus:
Prediction
Negative leverage response.
PHẦN L. FINANCIAL FLEXIBILITY
11. Logic
When future opportunities and risks become uncertain, firms value:
- unused borrowing capacity;
- liquidity;
- lower refinancing commitments.
Thus:
This is especially relevant for:
- constrained firms;
- highly leveraged firms;
- firms dependent on short-term borrowing.
PHẦN M. CREDIT-SUPPLY CHANNEL
12. Logic
Banks and debt investors may respond to UCT by:
- increasing spreads;
- reducing credit limits;
- shortening maturities;
- increasing collateral requirements.
Then:
leading to:
Important
Observed leverage reflects both:
- firm demand;
- lender supply.
PHẦN N. PECKING-ORDER / FINANCING NEED CHANNEL
13. Competing logic
If UCT reduces internal cash flow:
firms may increase debt before issuing equity.
Thus:
possible.
Additionally, ASEAN firms benefiting from trade diversion may need debt to finance:
- capacity expansion;
- working capital;
- new factories;
- supply-chain relocation.
Thus:
for selected firms.
PHẦN O. THEORY CHANNEL MATRIX
14. Bảng bắt buộc
| Channel | UCT effect on intermediate outcome | Predicted leverage effect | Main reference |
|---|---|---|---|
| Distress cost | Risk ↑ | − | Trade-off theory |
| Financial flexibility | Desired debt capacity ↑ | − | Chowdhury et al. |
| Credit supply | Loan supply ↓ | − | Chowdhury et al. |
| Investment delay | Financing demand ↓ | − | Uncertainty-investment literature |
| Internal cash flow | Internal funds ↓ | + possible | Pecking order |
| Trade diversion | Growth opportunities ↑ | + possible | Jimoh et al. |
| Supply-chain disruption | Operating risk ↑ | − | Chai & Wang |
| Agency/institutional channel | Debt discipline/value changes | +/− | Vega-Gutiérrez et al. |
PHẦN P. STREAM 4 — WHY ASEAN?
15. Core paper
Jimoh et al. (2026)
Use for:
- ASEAN asymmetry;
- U.S. dependence;
- China dependence;
- trade diversion;
- heterogeneous effects.
Supporting:
- IMF ASEAN fragmentation evidence;
- official ASEAN sources.
PHẦN Q. ASEAN IS NOT A SINGLE ECONOMY
16. Key argument
ASEAN countries differ in:
- U.S. export dependence;
- China import dependence;
- FDI exposure;
- financial development;
- institutions;
- sector composition;
- capital-market depth.
Therefore:
can produce:
PHẦN R. DRAFT ASEAN SETTING
17. Đoạn mẫu
ASEAN provides a useful setting because its economies occupy different positions between the United States and China. Regional firms participate extensively in global value chains, often relying on Chinese intermediate inputs while exporting final products to U.S. markets. At the same time, U.S.–China tensions can create trade-diversion and production-relocation opportunities for selected ASEAN economies. Jimoh et al. (2026) show that the trade effects of bilateral tensions differ substantially across ASEAN members and depend on their pre-existing exposure to the two major economies. These differences imply that the financing consequences of U.S.–China tensions are unlikely to be homogeneous across the region.
PHẦN S. STREAM 5 — HETEROGENEITY
18. Main dimensions
Firm-Level
- financial constraints;
- cash holdings;
- prior leverage;
- short-term debt dependence;
- U.S./China exposure.
Country-Level
- institutional quality;
- financial development;
- trade exposure.
Industry-Level
- export intensity;
- China input dependence;
- tariff exposure.
PHẦN T. FINANCIAL CONSTRAINTS
19. Logic
Constrained firms have:
- fewer financing alternatives;
- weaker liquidity;
- higher external-finance costs.
Thus UCT shocks may cause stronger deleveraging.
Closest evidence
Bui et al. (2026)
Prediction
If the negative channel dominates:
larger among constrained firms.
PHẦN U. INSTITUTIONAL QUALITY
20. Logic
Stronger institutions can:
- protect creditors;
- reduce information asymmetry;
- improve enforcement;
- support financing access.
Thus they may cushion UCT effects.
But they may also enable faster financial-market repricing.
Reference
Vega-Gutiérrez et al. (2025)
Do not assume one sign without reasoning.
PHẦN V. FINANCIAL DEVELOPMENT
21. Logic
Deeper financial systems provide:
- more bank capacity;
- bond-market alternatives;
- equity financing;
- refinancing options.
Possible prediction:
negative UCT effect weaker in more developed systems.
PHẦN W. INFORMATION ENVIRONMENT
22. Wang, Wang, and Johnson
Use to support:
capital structure depends partly on geography and information environment.
In ASEAN:
cross-country information quality and financing access differ.
This can amplify heterogeneous UCT responses.
Do not make this a central hypothesis unless directly tested.
PHẦN X. LITERATURE STREAM MATRIX
23. Bảng bắt buộc
| Stream | Core papers | What we know | What remains unknown | Used in paper |
|---|---|---|---|---|
| UCT measurement/effects | Rogers; Peng; Chai & Wang | UCT affects markets/corporate decisions | Financing effects outside US/China | Intro + X |
| Uncertainty & leverage | Chowdhury; Vega-Gutiérrez; Bui | Mixed leverage responses | ASEAN multi-country effects | Theory + H1 |
| Capital structure theory | MM; Myers; etc. | Financing tradeoffs | Response to bilateral tension | Theory |
| ASEAN trade tension | Jimoh; IMF | ASEAN effects heterogeneous | Corporate financing consequences | Setting + H2 |
| Institutions/constraints | Vega-Gutiérrez; Bui; Wang | Moderators matter | Which ASEAN firms/countries react most? | H3/H4 |
PHẦN Y. RESEARCH GAP
24. Gap không phải “chưa ai làm đúng tiêu đề này”
Bad gap:
No study has examined U.S.–China tension and capital structure in ASEAN.
Even if technically true, this is weak.
Better gap has 3 layers
Gap 1 — Shock
Existing capital-structure literature mainly studies:
- EPU;
- GPR.
UCT is more bilateral and economically specific.
Gap 2 — Geography
Closest UCT–leverage evidence is Vietnam/single-country.
ASEAN responses are heterogeneous.
Gap 3 — Mechanism/Identification
Little evidence on whether:
- U.S. exposure;
- China exposure;
- trade diversion;
- financial constraints;
- debt maturity
shape the financing response.
PHẦN Z. RESEARCH GAP TABLE
25. Bảng bắt buộc
| Existing literature | What it establishes | Limitation | Current paper response |
|---|---|---|---|
| Chowdhury et al. | GPR affects leverage | Broad GPR; U.S. setting | Bilateral UCT; ASEAN |
| Vega-Gutiérrez et al. | EPU affects leverage | European EPU context | UCT and ASEAN |
| Bui et al. | UCT affects Vietnam leverage | Single country | Multi-country ASEAN |
| Peng et al. | UCT affects markets | Not capital structure | Corporate financing |
| Chai & Wang | Tension changes corporate strategy | Chinese firms; different index | ASEAN financing response |
| Jimoh et al. | ASEAN trade effects heterogeneous | Trade outcomes, not financing | Exposure × UCT financing design |
PHẦN AA. CONTRIBUTION
26. Contribution phải match actual analyses
Potential contributions:
Contribution 1 — Shock-specific
Use bilateral UCT rather than broad EPU/GPR.
Contribution 2 — Regional
Extend evidence to ASEAN multi-country setting.
Contribution 3 — Capital Structure Dimension
Study:
- total leverage;
- short-term leverage;
- long-term leverage;
- debt maturity.
Contribution 4 — Exposure
Distinguish:
- U.S. exposure;
- China exposure.
Contribution 5 — Mechanisms
Test:
- financial constraints;
- debt demand;
- credit supply;
- trade diversion;
- supply chains.
Contribution 6 — Identification
Use exposure-based designs with stronger FE.
Rule
Only claim contributions actually implemented.
PHẦN AB. CONTRIBUTION MATRIX
27. Bảng
| Contribution | Evidence in paper | Main comparator | Strong enough? |
|---|---|---|---|
| ASEAN multi-country | Country sample | Bui et al. | |
| Bilateral UCT | Rogers index | GPR/EPU papers | |
| Debt maturity | ST/LT/DM results | Bui/Chowdhury | |
| U.S./China exposure | Interaction tests | Jimoh | |
| Institutions | IQ interaction | Vega-Gutiérrez | |
| Stronger identification | Firm exposure + CY FE | Prior studies |
PHẦN AC. HYPOTHESIS DEVELOPMENT
28. Không cần quá nhiều hypotheses
Recommended:
- H1: main leverage effect
- H2: U.S./China exposure heterogeneity
- H3: financial constraints
- H4: debt maturity / short-term debt
Optional:
- H5: institutions / financial development
4 hypotheses thường đủ.
PHẦN AD. H1 — MAIN EFFECT
29. Negative directional H1
Nếu paper lựa chọn risk/flexibility channel làm dominant theory:
H1. Higher U.S.–China tension is associated with lower corporate leverage among ASEAN firms.
Supporting logic
causes/reflects:
- greater operating risk;
- higher expected distress cost;
- lower credit supply;
- higher value of flexibility.
Therefore:
PHẦN AE. H1 DEVELOPMENT — SAMPLE
30. Draft
U.S.–China tensions can increase uncertainty surrounding trade policy, supply chains, export demand, and firms' future cash flows. Under the trade-off framework, greater uncertainty raises expected distress costs and reduces the optimal use of debt. Firms may also preserve unused borrowing capacity when future financing needs become less predictable, while lenders may tighten credit standards in response to heightened risk. These mechanisms are consistent with recent evidence that geopolitical risk reduces corporate leverage (Chowdhury et al., 2025) and that U.S.–China tensions are associated with lower overall and short-term leverage among Vietnamese firms (Bui et al., 2026). Although other forms of uncertainty can generate different leverage responses, we expect the risk and financial-flexibility channels to dominate on average in ASEAN. We therefore propose:
H1: Higher U.S.–China tension is associated with lower corporate leverage among ASEAN firms.
PHẦN AF. IF BASELINE RESULT IS POSITIVE
31. Do not rewrite history silently
If empirical result turns out positive:
H1 can remain negative if that was the theory-derived hypothesis.
Then Results say:
H1 is not supported.
And discussion explains:
- trade diversion;
- financing need;
- agency/institutional effects.
Alternative
If before formal hypotheses are finalized literature genuinely suggests ambiguity:
Use non-directional H1:
H1: U.S.–China tension is significantly associated with corporate leverage among ASEAN firms.
But directional hypotheses are usually more theoretically informative.
PHẦN AG. H2 — TRADE EXPOSURE
32. Need separate U.S. and China exposure
A single combined exposure can hide opposite mechanisms.
Possible H2a/H2b.
H2a — China Exposure
If supply-chain vulnerability dominates:
H2a: The negative association between U.S.–China tension and corporate leverage is stronger for firms/economies with greater pre-existing exposure to China.
H2b — U.S. Exposure
Sign can be ambiguous.
High U.S. exposure may:
- suffer demand uncertainty;
- benefit from trade diversion.
If no clear theoretical dominance:
make H2b non-directional.
H2b: The relationship between U.S.–China tension and corporate leverage varies with firms' or countries' pre-existing exposure to the United States.
PHẦN AH. H2 DEVELOPMENT
33. Draft
The impact of U.S.–China tension should depend on firms' economic exposure to the two major economies. ASEAN firms that rely heavily on Chinese inputs, suppliers, or production networks may experience greater operating and supply-chain risk when bilateral tensions intensify. Such firms may face greater cash-flow uncertainty and weaker debt capacity, implying a stronger deleveraging response. At the same time, firms more connected to U.S. markets may either face greater export uncertainty or benefit from trade diversion away from China. Jimoh et al. (2026) demonstrate that ASEAN responses to U.S.–China trade tensions vary substantially with pre-existing trade dependence, supporting an exposure-based approach rather than a homogeneous regional effect.
PHẦN AI. H3 — FINANCIAL CONSTRAINTS
34. Hypothesis
H3: The negative association between U.S.–China tension and corporate leverage is stronger among financially constrained firms.
Logic
Constrained firms have:
- limited external financing options;
- weaker liquidity buffers;
- higher marginal financing costs.
UCT can amplify these frictions.
Supporting paper
Bui et al. (2026).
PHẦN AJ. H3 DEVELOPMENT
35. Draft
Financially constrained firms are likely to be particularly sensitive to geopolitical and trade-related shocks because they have limited access to external financing and fewer resources with which to absorb temporary disruptions. When U.S.–China tension raises uncertainty about sales, input costs, or financing conditions, constrained firms may be forced to reduce borrowing more aggressively than unconstrained firms. Consistent with this reasoning, Bui et al. (2026) report stronger UCT–leverage effects among financially constrained Vietnamese firms. We therefore expect the financing response to UCT to be more pronounced among constrained ASEAN firms.
PHẦN AK. H4 — SHORT-TERM DEBT / DEBT MATURITY
36. Why this matters
Short-term debt exposes firms to:
- rollover risk;
- refinancing risk;
- credit-renewal uncertainty.
UCT can make short-term funding particularly risky.
Possible H4a
H4a: Higher U.S.–China tension is associated with a stronger reduction in short-term leverage than in long-term leverage.
Possible H4b
If maturity increases:
H4b: Higher U.S.–China tension is associated with a longer debt-maturity structure as firms reduce rollover exposure.
Important
H4b is not guaranteed.
Credit supply can also shorten maturity.
If literature is ambiguous:
use H4a only.
PHẦN AL. H4 DEVELOPMENT
37. Draft
Debt maturity provides an additional margin through which firms can respond to geopolitical uncertainty. Short-term debt exposes firms to frequent refinancing and therefore becomes particularly costly when lenders reassess risk or credit conditions deteriorate. Firms may respond by reducing short-term borrowing or shifting toward longer maturities to limit rollover risk. The direct evidence from Vietnam indicates that U.S.–China tensions are associated with lower short-term leverage, suggesting that refinancing-sensitive debt may respond particularly strongly to bilateral tensions. We therefore expect the leverage response to be concentrated in short-maturity debt.
PHẦN AM. OPTIONAL H5 — INSTITUTIONS / FINANCIAL DEVELOPMENT
38. Use only if actually tested
Possible hypothesis
H5: The negative association between U.S.–China tension and corporate leverage is weaker in ASEAN countries with stronger institutional quality and deeper financial systems.
Logic
Better systems provide:
- creditor protection;
- information quality;
- alternative financing;
- financial intermediation capacity.
Supporting
Vega-Gutiérrez et al. (2025).
Caution
Their empirical sign may differ from current setting.
Use their moderation evidence, not blindly their coefficient direction.
PHẦN AN. HYPOTHESIS EVIDENCE TABLE
39. Bảng bắt buộc
| Hypothesis | Theory | Supporting paper | Opposing evidence | Empirical test |
|---|---|---|---|---|
| H1: UCT → leverage − | Trade-off/flexibility | Chowdhury; Bui | Vega-Gutiérrez | Baseline |
| H2a: China exposure strengthens effect | Supply-chain risk | Jimoh; Chai & Wang | Trade diversion | UCT×ChinaExposure |
| H2b: U.S. exposure moderates | Trade exposure | Jimoh | Competing signs | UCT×USExposure |
| H3: Constraints strengthen effect | Financing friction | Bui | UCT×Constraint | |
| H4: ST debt more sensitive | Rollover risk | Bui; debt literature | Lenders may shorten maturity | ST/LT/DM |
| H5: Strong institutions cushion | Institutional finance | Vega-Gutiérrez | Possible faster repricing | UCT×IQ |
PHẦN AO. DO NOT HIDE OPPOSING EVIDENCE
40. Every directional hypothesis should mention competing logic
Example H1:
Supporting negative
- trade-off;
- credit supply;
- financial flexibility.
Opposing positive
- financing need;
- trade diversion;
- pecking-order response;
- Vega-Gutiérrez.
Then explain why chosen dominant prediction.
This is much stronger than a one-sided literature review.
PHẦN AP. HOW TO USE EACH PROVIDED PAPER IN STEP 5
41. Rogers et al. (2024)
Use for:
- UCT definition;
- why bilateral tension is distinct.
Not for:
- capital-structure result.
42. Peng et al. (2025)
Use for:
- financial relevance;
- asymmetric/time-varying nature.
Not for:
- leverage prediction.
43. Chai and Wang (2026)
Use for:
- supply-chain;
- strategic reallocation;
- technology blockade.
Not for:
- Rogers UCT definition.
44. Jimoh et al. (2026)
Use for:
- Why ASEAN?;
- exposure heterogeneity;
- U.S. vs China dependence.
Not for:
- direct leverage effect.
45. Vega-Gutiérrez et al. (2025)
Use for:
- uncertainty–leverage literature;
- competing positive evidence;
- institutions;
- financial development.
46. Chowdhury et al. (2025)
Use for:
- main theoretical comparison;
- negative leverage channel;
- demand and supply mechanisms;
- forward leverage structure.
47. Wang, Wang, and Johnson
Use for:
- information environment;
- geography;
- financing heterogeneity.
Supporting role only.
PHẦN AQ. PAPER-TO-PARAGRAPH MAP
48. Bảng
| Paragraph | Main papers |
|---|---|
| UCT background | Rogers + Peng |
| Corporate response to tension | Chai & Wang |
| Uncertainty and leverage | Chowdhury + Vega-Gutiérrez + Bui |
| ASEAN setting | Jimoh |
| Capital-structure theory | MM + Myers + empirical CS literature |
| H1 | Chowdhury + Bui + Vega opposing |
| H2 | Jimoh + Chai & Wang |
| H3 | Bui + financing-constraint literature |
| H4 | Bui + debt maturity literature |
| H5 optional | Vega-Gutiérrez |
PHẦN AR. LITERATURE REVIEW OUTLINE
49. Recommended Section 2
2.1. U.S.–China Tension and Corporate Decisions
Approx:
450–600 words
2.2. Uncertainty and Corporate Capital Structure
Approx:
550–750 words
2.3. ASEAN as an Exposure Setting
Approx:
400–550 words
2.4. Hypothesis Development
Approx:
1.200–1.800 words
PHẦN AS. WHAT NOT TO INCLUDE
50. Avoid textbook history
Do not spend 500 words explaining:
- MM Proposition I;
- MM Proposition II;
- every capital-structure theory.
Only explain theory used to derive hypotheses.
51. Avoid citation dumping
Bad:
Studies have examined uncertainty (A, 2020; B, 2021; C, 2022; D, 2023).
Better:
Prior studies provide conflicting evidence on how uncertainty affects leverage. Chowdhury et al. document a negative effect through financing demand and supply, whereas Vega-Gutiérrez et al. find a positive relationship that depends on institutional conditions.
PHẦN AT. RESEARCH GAP PARAGRAPH
52. Draft mẫu
Despite growing evidence that U.S.–China tensions affect financial markets, investment, supply chains, and corporate strategic behavior, their implications for corporate financing remain less understood. Existing capital-structure studies focus predominantly on broad measures of economic-policy or geopolitical uncertainty, while the closest direct evidence on U.S.–China tensions is limited to a single-country Vietnamese setting. This leaves open whether the financing response generalizes across ASEAN, a region characterized by substantial differences in trade exposure, institutional quality, and financial development. Moreover, existing work provides limited evidence on whether U.S.– and China-specific exposure, debt maturity, and financial constraints shape firms' financing responses. We address these gaps by examining multiple dimensions of capital structure across ASEAN firms and by exploiting heterogeneity in pre-existing economic exposure to the two major economies.
PHẦN AU. CONTRIBUTION PARAGRAPH
53. Draft mẫu
This study contributes to the literature in three ways. First, we extend research on uncertainty and corporate financing by focusing on a bilateral U.S.–China tension measure rather than a broad domestic or geopolitical uncertainty index. Second, we move beyond single-country evidence by examining firms across ASEAN, where exposure to the United States and China varies substantially across economies and industries. Third, we examine multiple financing margins and exposure-based channels, including short- versus long-term debt, debt maturity, financial constraints, and pre-existing U.S. and China exposure. These analyses help distinguish a general uncertainty effect from economically meaningful exposure to bilateral tensions.
Rule
Nếu paper không chạy một contribution:
delete it.
PHẦN AV. BRIDGE FROM THEORY TO EMPIRICS
54. Cuối Section 2 phải dẫn tự nhiên sang Methodology
Draft
Taken together, the literature suggests that U.S.–China tension can affect corporate leverage through several competing channels and that the net effect is likely to depend on firms' financing constraints and economic exposure to the two countries. We therefore examine the average relationship between UCT and corporate capital structure and subsequently exploit heterogeneity in trade exposure, financing conditions, and debt maturity to distinguish among these mechanisms. The next section describes the data, variable construction, and empirical specifications used to test these predictions.
PHẦN AW. HARKING CHECK
55. Bảng bắt buộc
For each hypothesis:
| Hypothesis | Derived from literature before result? | Result known when finalized? | Theory changed after result? | How disclosed? |
|---|---|---|---|---|
| H1 | ||||
| H2 | ||||
| H3 | ||||
| H4 |
Teaching rule
If a hypothesis was created after exploratory analysis:
do not pretend it was pre-registered.
Use language such as:
- exploratory prediction;
- additional analysis;
- ex post mechanism analysis
where appropriate.
PHẦN AX. CLAIM–CITATION CHECK
56. Every important literature statement needs a source
Bảng:
| Claim | Citation | Verified in original paper? |
|---|---|---|
| UCT is news-based | Rogers | |
| UCT financial effects are asymmetric | Peng | |
| Tension changes overseas allocation | Chai & Wang | |
| GPR reduces leverage | Chowdhury | |
| EPU can increase leverage | Vega-Gutiérrez | |
| ASEAN responses differ by exposure | Jimoh | |
| UCT lowers Vietnam leverage | Bui |
PHẦN AY. LITERATURE EVIDENCE MATRIX
57. Sinh viên bắt buộc hoàn thành
| Paper | Exact section/page | Claim used | Theory/channel | Hypothesis supported | Sentence draft |
|---|---|---|---|---|---|
| Rogers | |||||
| Peng | |||||
| Chai & Wang | |||||
| Jimoh | |||||
| Vega-Gutiérrez | |||||
| Chowdhury | |||||
| Bui | |||||
| Wang et al. |
PHẦN AZ. EXCEL FILE CHO BƯỚC 5
58. File đề xuất
05_THEORY_LITERATURE_EVIDENCE.xlsx
Sheet 1 — STORY
| Question | Answer |
|---|
Sheet 2 — LITERATURE_STREAMS
| Stream | Paper | Main result | Use | Gap |
|---|
Sheet 3 — THEORY_CHANNELS
| Channel | Intermediate effect | Leverage prediction | Supporting paper | Opposing logic |
|---|
Sheet 4 — GAP
| Existing paper | Contribution | Limitation | Our response |
|---|
Sheet 5 — CONTRIBUTION
| Contribution | Evidence | Comparator | Keep/Delete |
|---|
Sheet 6 — HYPOTHESES
| H | Statement | Theory | Evidence | Opposing evidence | Test |
|---|
Sheet 7 — CITATION_CHECK
| Claim | Citation | Verified | Page/section |
|---|
Sheet 8 — WRITING
| Paragraph | Main message | Papers | Draft |
|---|
PHẦN BA. CORE REFERENCES
59. UCT
Rogers, J. H., Sun, B., & Sun, T. (2024).
U.S.–China Tension.
SSRN Working Paper.
DOI: 10.2139/ssrn.4815838
60. UCT Financial Effects
Peng, C., Deng, H., Xie, J., & Liu, X. (2025).
US-China tension and stock market performance in US and China: New insights from time-varying quantile causality method.
Finance Research Letters, 85, 107888.
DOI: 10.1016/j.frl.2025.107888
61. Corporate Strategic Response
Chai, H., & Wang, Y. (2026).
China–US relations and overseas capital allocation of Chinese firms.
Finance Research Letters, 93, 109621.
DOI: 10.1016/j.frl.2026.109621
62. ASEAN
Jimoh, S. O., Goh, S. K., & Tham, S. Y. (2026).
Between giants: The US-China trade war and ASEAN’s evolving role in global trade.
International Economics and Economic Policy, 23(2), 1–24.
DOI: 10.1007/s10368-026-00730-x
63. GPR and Capital Structure
Chowdhury, M. S. R., Aram, M., Javadi, S., & Nejadmalayeri, A. (2025).
Geopolitical risk and corporate capital structure.
Journal of Corporate Finance, 93, 102796.
DOI: 10.1016/j.jcorpfin.2025.102796
64. EPU and Capital Structure
Vega-Gutiérrez, P. L., López-Iturriaga, F. J., & Rodríguez-Sanz, J. A. (2025).
Economic policy uncertainty and capital structure in Europe: An agency approach.
The European Journal of Finance, 31(1), 53–75.
DOI: 10.1080/1351847X.2024.2364044
65. Closest UCT-Leverage Paper
Bui Tien Thinh, Le Thi My Nuong, Ha Thi Thuy, & Dao Nhat Minh. (2026).
Corporate leverage and U.S.-China tensions: Evidence from Vietnam.
Cogent Economics & Finance, 14(1), 2613610.
DOI: 10.1080/23322039.2026.2613610
66. Geography / Information Environment
Wang, X., Wang, J., & Johnson, L.
Geography and capital structure.
Canadian Journal of Administrative Sciences, 35(1), 107–122.
DOI: 10.1002/cjas.1383
Verify final bibliographic year according to the target journal style.
PHẦN BB. CAPITAL STRUCTURE FOUNDATIONS
67.
Modigliani, F., & Miller, M. H. (1958).
The cost of capital, corporation finance and the theory of investment.
American Economic Review, 48(3), 261–297.
Myers, S. C. (1984).
The capital structure puzzle.
The Journal of Finance, 39(3), 575–592.
Myers, S. C., & Majluf, N. S. (1984).
Corporate financing and investment decisions when firms have information that investors do not have.
Journal of Financial Economics, 13(2), 187–221.
Jensen, M. C., & Meckling, W. H. (1976).
Theory of the firm: Managerial behavior, agency costs and ownership structure.
Journal of Financial Economics, 3(4), 305–360.
PHẦN BC. EMPIRICAL CAPITAL STRUCTURE
68.
Rajan, R. G., & Zingales, L. (1995).
What do we know about capital structure? Some evidence from international data.
The Journal of Finance, 50(5), 1421–1460.
Frank, M. Z., & Goyal, V. K. (2009).
Capital structure decisions: Which factors are reliably important?
Financial Management, 38(1), 1–37.
Lemmon, M. L., Roberts, M. R., & Zender, J. F. (2008).
Back to the beginning: Persistence and the cross-section of corporate capital structure.
The Journal of Finance, 63(4), 1575–1608.
PHẦN BD. SINH VIÊN PHẢI TÌM THÊM
69. Minimum additional literature
UCT / Trade Tension
At least:
2 additional papers
Uncertainty / GPR → Capital Structure
At least:
3 additional papers
ASEAN Capital Structure
At least:
3 papers
Debt Maturity / Rollover Risk
At least:
2 papers
Financial Constraints
At least:
2 papers
Institutions / Financial Development
At least:
2 papers
Total target for theory/literature section
Approximately:
25–35 high-relevance references
Not 60 loosely related papers.
PHẦN BE. QUALITY FILTER FOR REFERENCES
70. Priority
- Top/strong finance/accounting/economics journals
- Strong field journals
- Recent direct studies
- Foundational classics
- Official institutional reports for ASEAN context
Avoid
- random working papers when published version exists;
- citation from secondary blogs;
- papers only sharing keywords but no conceptual role.
PHẦN BF. WRITING CHECKLIST
71. Literature Review
Không viết theo kiểu paper-by-paper list.
Đã chia literature thành streams.
Đã giải thích UCT khác EPU/GPR.
Đã có direct UCT corporate evidence.
Đã có uncertainty–capital structure literature.
Đã trình bày evidence trái dấu.
Đã giải thích ASEAN exposure heterogeneity.
Đã xác định research gap.
Gap không chỉ là “chưa ai làm”.
Đã xác định contribution thật sự.
Không claim contribution chưa chạy.
PHẦN BG. HYPOTHESIS CHECKLIST
72.
H1 có theory rõ.
H1 có supporting evidence.
H1 có opposing logic được thừa nhận.
H2 dùng exposure logic.
U.S. và China exposure được phân biệt.
H3 dựa trên financing constraints.
H4 dựa trên rollover/debt maturity.
H5 chỉ dùng nếu actually tested.
Không tạo hypothesis chỉ vì result significant.
Exploratory analyses được gọi đúng là exploratory.
PHẦN BH. FINAL HYPOTHESES TEMPLATE
73. Recommended set
H1
Higher U.S.–China tension is associated with lower corporate leverage among ASEAN firms.
H2a
The negative association between U.S.–China tension and corporate leverage is stronger for firms or economies with greater pre-existing exposure to China.
H2b
The relationship between U.S.–China tension and corporate leverage varies with pre-existing exposure to the United States.
H3
The negative association between U.S.–China tension and corporate leverage is stronger among financially constrained firms.
H4
The negative association between U.S.–China tension and debt financing is stronger for short-term debt than for long-term debt.
Optional H5
The negative association between U.S.–China tension and corporate leverage is weaker in ASEAN economies with stronger institutions and deeper financial systems.
PHẦN BI. FINAL OUTPUT STRUCTURE
74. Draft manuscript section
2. Literature Review and Hypothesis Development
2.1. U.S.–China Tension and Corporate Decisions
450–600 words.
2.2. Uncertainty and Corporate Capital Structure
550–750 words.
2.3. ASEAN Exposure and Regional Heterogeneity
400–550 words.
2.4. Hypothesis Development
1.200–1.800 words.
Total
Approximately:
2.600–3.700 words
depending journal style.
PHẦN BJ. KHI NÀO ĐƯỢC CHUYỂN SANG BƯỚC 6?
Bước 5 hoàn thành khi:
- research gap rõ;
- contribution rõ;
- literature không phải list;
- competing theories được trình bày;
- H1–H4 có references;
- hypotheses match empirical tests;
- không HARKing giả tạo;
- mọi citation được kiểm chứng;
- theory dẫn tự nhiên sang model.
Sau đó chuyển sang:
BƯỚC 6. VIẾT HOÀN CHỈNH RESULTS & DISCUSSION
Ở Bước 6, sinh viên sẽ ghép:
- baseline results;
- robustness;
- advanced identification;
- mechanisms;
- heterogeneity;
- theory/literature
thành một Section Results & Discussion hoàn chỉnh, thay vì các bảng kết quả rời rạc.
BƯỚC 6. VIẾT HOÀN CHỈNH RESULTS & DISCUSSION
Bài mẫu: U.S.–China Tension and Corporate Capital Structure in ASEAN Countries
Mục tiêu đầu ra: Từ toàn bộ evidence đã xây ở Bước 2–5, sinh viên phải viết được một phần Results and Discussion hoàn chỉnh, có logic học thuật, không lặp bảng, không kể coefficient máy móc, và kết nối trực tiếp với theory, hypotheses và contribution.
Nguyên tắc của Bước 6
Results & Discussion không phải:
“Table 3 shows… Table 4 shows… Table 5 shows…”
mà phải trả lời một chuỗi câu hỏi:
- Main finding là gì?
- Magnitude có đáng kể về kinh tế không?
- Finding có robust không?
- Identification mạnh đến đâu?
- Channel nào giải thích kết quả?
- Kết quả khác nhau theo firm/country nào?
- Kết quả phù hợp hay mâu thuẫn literature nào?
- Điều này bổ sung gì cho hiểu biết về ASEAN?
- Có được dùng causal language không?
Bước 6 là bước viết, không phải tiếp tục chạy thêm regression vô hạn.
PHẦN A. THÀNH PHẨM CUỐI BƯỚC 6
Bước 6 chỉ hoàn thành khi nhóm có:
- Results Storyboard
- Table-to-Claim Matrix
- Hypothesis Results Table
- Economic Significance Summary
- Robustness Summary
- Identification Strength Summary
- Mechanism Summary
- Heterogeneity Summary
- Literature Discussion Matrix
- Full Results & Discussion draft, khoảng 3.500–5.500 từ
- Danh sách claims cần giảm mức độ mạnh
- Danh sách tables/figures đưa main text và appendix
PHẦN B. TRƯỚC KHI VIẾT: TÓM BÀI TRONG 8 CÂU
1. Research Storyboard
Sinh viên phải điền trước:
| Câu hỏi | Câu trả lời của paper |
|---|---|
| 1. Baseline UCT effect là gì? | |
| 2. Economic magnitude là bao nhiêu? | |
| 3. ST và LT debt phản ứng khác nhau không? | |
| 4. Main result robust đến đâu? | |
| 5. Strongest identification evidence là gì? | |
| 6. Main mechanism là gì? | |
| 7. Heterogeneity quan trọng nhất là gì? | |
| 8. Contribution đối với ASEAN literature là gì? |
Quy tắc
Nếu chưa trả lời được 8 câu này bằng ngôn ngữ đơn giản:
chưa bắt đầu viết Results & Discussion.
PHẦN C. CẤU TRÚC RESULTS & DISCUSSION ĐỀ XUẤT
2. Section 4. Empirical Results and Discussion
Recommended structure:
4.1. Descriptive Statistics and Correlations
4.2. Baseline Results
4.3. Economic Significance and Debt Composition
4.4. Robustness Tests
4.5. Identification and Endogeneity Analyses
4.6. Mechanisms
4.7. Heterogeneity and ASEAN Exposure
4.8. Discussion of Findings
Tùy journal có thể tách:
Results và Discussion.
Nhưng logic vẫn phải giữ.
PHẦN D. TABLE ORDER
3. Không để thứ tự bảng theo thứ tự chạy Stata
Main-text order nên phản ánh research story.
Table 1
Descriptive Statistics
Table 2
Correlation Matrix / VIF if needed
Table 3
Baseline UCT and Capital Structure
Table 4
Alternative Capital-Structure Measures
Table 5
Global Controls / Robustness
Table 6
Exposure-Based Identification
Table 7
Mechanisms
Table 8
Heterogeneity
Appendix
- many alternative VCEs;
- leave-one-country-out details;
- alternative UCT scaling;
- additional subperiods;
- extended GMM diagnostics;
- minor controls.
Rule
Main text tables answer:
main economic questions.
Appendix tables answer:
technical reassurance.
PHẦN E. TABLE-TO-CLAIM MATRIX
4. Bảng bắt buộc
| Table | Main question | Main coefficient | Main claim | Supporting citation | Main/Appendix |
|---|---|---|---|---|---|
| Table 1 | Sample looks plausible? | N/A | Main | ||
| Table 3 | Does UCT affect leverage? | UCT | Chowdhury; Bui | Main | |
| Table 4 | Which debt component responds? | UCT | Bui; debt literature | Main | |
| Table 5 | Is result robust? | UCT | Main | ||
| Table 6 | Does exposure matter? | UCT×Exposure | Jimoh | Main | |
| Table 7 | Why does it happen? | UCT / interaction | Chowdhury; Chai-Wang | Main | |
| Table 8 | For whom is effect stronger? | interaction | Bui; Vega | Main |
PHẦN F. 4.1 DESCRIPTIVE STATISTICS
5. Mục tiêu
Descriptive statistics không phải phần để đọc từng mean.
Phải trả lời:
- sample size có hợp lý không?
- leverage distributions có plausible không?
- UCT có enough variation không?
- country representation có lệch không?
- correlations có warning gì không?
6. Variables cần report
Tối thiểu:
- Book Leverage;
- ST Leverage;
- LT Leverage;
- Debt Maturity nếu dùng;
- UCT;
- Tangibility;
- Profitability;
- Size;
- Market-to-Book;
- GDP Growth;
- Inflation.
Có thể thêm:
- Cash;
- Exposure;
- SA/WW;
- Institutional Quality
nếu main paper dùng.
PHẦN G. DESCRIPTIVE TABLE
7. Standard table
| Variable | N | Mean | SD | P25 | Median | P75 | Min | Max |
|---|---|---|---|---|---|---|---|---|
Winsorized variables
Min/Max sau winsorization.
UCT
Không winsorize baseline.
Do đó extreme UCT values cần được nhìn riêng.
PHẦN H. CÁCH VIẾT DESCRIPTIVES
8. Không viết
Mean Size is 7.2. Mean Tangibility is 0.31. Mean Profitability is 0.05…
Viết tốt hơn
Table 1 reports summary statistics for the variables used in the empirical analysis. The sample displays substantial cross-sectional variation in corporate leverage and firm characteristics. The UCT index also varies considerably over time, providing the aggregate tension variation used in the baseline specifications. Importantly, the relative magnitudes of short- and long-term debt indicate that ASEAN firms differ meaningfully in their reliance on refinancing-sensitive financing, motivating our separate analysis of debt maturity components.
Nếu country concentration lớn
The sample is not evenly distributed across ASEAN economies, with [countries] contributing the largest number of firm-year observations. We therefore assess country sensitivity in later analyses.
PHẦN I. CORRELATIONS
9. Correlation Matrix
Dùng để:
- show simple association;
- identify extreme correlation among controls;
- inspect UCT vs global uncertainty measures.
Không dùng để kết luận hypothesis.
10. Special correlation checks
Report or inspect:
Why?
Annual UCT direct effect may overlap global shocks.
PHẦN J. 4.2 BASELINE RESULTS
11. Baseline Results phải viết theo 4 lớp
Layer 1
Statistical direction.
Layer 2
Economic magnitude.
Layer 3
Debt composition.
Layer 4
Literature/theory interpretation.
Không chỉ đọc coefficient.
PHẦN K. BASELINE PARAGRAPH 1 — TABLE SETUP
12. Draft mẫu
Table 3 presents the baseline estimates of the relationship between U.S.–China tension and the capital-structure choices of ASEAN firms. The dependent variables capture alternative dimensions of debt financing, including [insert exact outcomes]. All regressions include the firm-level and macroeconomic controls described in Section 3 and firm fixed effects. Because the annual UCT index is common to all firms in a given year, the direct-effect specifications do not include a full set of year fixed effects, which would absorb the main explanatory variable.
Must customize
- Table number;
- actual DVs;
- FE;
- VCE.
PHẦN L. BASELINE PARAGRAPH 2 — MAIN UCT RESULT
13. Nếu H1 được support
The coefficient on UCT is negative and statistically significant in the main leverage specification. In Column (1), the estimated coefficient is [β], with a [standard error/t-statistic] of [ ], indicating that periods of heightened U.S.–China tension are associated with lower corporate leverage among ASEAN firms. The result supports H1 and is consistent with the view that greater bilateral tension increases financing uncertainty and encourages firms to preserve financial flexibility.
Không dùng “supports H1” nếu H1 không directional hoặc chưa formalized.
14. Nếu H1 không support
Contrary to H1, the coefficient on UCT is positive and statistically significant. This result indicates that ASEAN firms increase their reliance on debt during periods of heightened bilateral tension. The finding suggests that financing-need or trade-diversion channels may dominate the financial-flexibility channel in the average ASEAN firm.
Sau đó phải so với Vega-Gutiérrez và ASEAN opportunity logic.
15. Nếu insignificant
The coefficient on UCT is statistically indistinguishable from zero in the baseline specification. We therefore do not find evidence of a uniform average relationship between bilateral tension and leverage across ASEAN firms. Given the substantial variation in ASEAN countries' exposure to the United States and China documented by Jimoh et al. (2026), this result motivates the exposure-based tests reported below.
PHẦN M. 4.3 ECONOMIC SIGNIFICANCE
16. Statistical significance không đủ
Phải tính:
Relative effect:
17. Economic Significance Table
| DV | β | SD(UCT) | 1-SD effect | Mean DV | % of mean |
|---|---|---|---|---|---|
Optional
IQR effect.
PHẦN N. ECONOMIC SIGNIFICANCE WRITING
18. Draft
The estimated relationship is economically meaningful. A one-standard-deviation increase in UCT is associated with a [X]-percentage-point change in [DV], equivalent to approximately [Y]% of its sample mean. This magnitude indicates that the estimated financing response is not merely statistically detectable but economically relevant for firms' capital structures.
If effect small
Do not exaggerate.
Although statistically significant, the magnitude is modest relative to the mean leverage ratio.
PHẦN O. DEBT COMPOSITION DISCUSSION
19. One of the paper's strongest potential contributions
Do not merely report:
- TDB;
- SDB;
- LDB.
Tell the financing story.
20. Pattern 1 — ST debt drives effect
If:
stronger than:
write:
The decomposition indicates that the leverage response is concentrated in short-term borrowing. This pattern is consistent with firms reducing refinancing-sensitive obligations when bilateral tension increases uncertainty about future cash flows and credit conditions.
Link
- Bui et al. (2026);
- rollover-risk logic.
21. Pattern 2 — LT debt drives effect
The response is concentrated in long-term borrowing, suggesting that elevated U.S.–China tension primarily affects access to or demand for long-horizon external financing rather than immediate short-term refinancing.
22. Pattern 3 — Debt maturity rises
If:
while ST debt falls:
Firms appear to extend debt maturity and reduce rollover exposure.
23. Pattern 4 — Debt maturity falls
Possible credit-supply interpretation:
Lenders may shorten maturities during periods of heightened uncertainty.
Do not assume maturity always increases.
PHẦN P. FORMAL DIFFERENCE TEST
24. If claiming ST effect stronger than LT
Need test:
Do not infer from
- ST significant;
- LT insignificant.
Manuscript
If formal difference is not tested:
write:
The coefficient is numerically larger for short-term debt.
Not:
The effect is significantly stronger.
PHẦN Q. 4.4 ROBUSTNESS SECTION
25. Do not narrate 15 tests one by one
Group tests into families.
Family A
Alternative measures.
Family B
Global controls.
Family C
Time/event sensitivity.
Family D
Country/sample sensitivity.
Family E
Alternative inference.
PHẦN R. ROBUSTNESS PARAGRAPH — ALTERNATIVE DVs
26. Draft
We first examine whether the baseline result depends on the definition of capital structure. Re-estimating the model using [book leverage, market leverage, short-term leverage, long-term leverage, debt maturity] yields a broadly similar pattern. The coefficient on UCT remains [direction] for [list main measures], indicating that the baseline result is not driven by a single leverage construction.
If mixed
The evidence is strongest for [measures] and weaker for [measures], suggesting that UCT affects specific financing margins rather than all dimensions of capital structure uniformly.
PHẦN S. ROBUSTNESS PARAGRAPH — GLOBAL CONTROLS
27. Critical because no Year FE in annual direct model
Draft
Because annual UCT varies primarily over time, an important concern is that it may proxy for broader global shocks. We therefore augment the baseline specification with global EPU, geopolitical risk, financial-market volatility, and U.S. monetary conditions. The UCT coefficient remains [sign/magnitude] after these controls, suggesting that the estimated relationship contains information beyond general global uncertainty.
If coefficient weakens
The reduction in magnitude indicates that part of the baseline association overlaps with broader global uncertainty conditions.
This is acceptable.
PHẦN T. ROBUSTNESS PARAGRAPH — COVID / EVENT YEARS
28. Draft
We also assess whether the results are driven by unusually turbulent periods. Excluding the COVID-19 years and, separately, the 2018–2019 trade-war escalation period produces [similar/different] estimates. The coefficient also remains [stable/unstable] when the highest-UCT year is excluded. These tests indicate that [the baseline association is not attributable to a single extreme episode / a substantial portion of the effect is concentrated in major tension episodes].
PHẦN U. ROBUSTNESS PARAGRAPH — COUNTRY SENSITIVITY
29. Draft
The number of listed firms differs considerably across ASEAN economies, raising the possibility that the regional result is driven by one large country sample. We therefore exclude each country sequentially. The UCT coefficient retains the same sign in [X of Y] estimations and ranges from [minimum] to [maximum], suggesting that [no single economy drives the result / the regional estimate is particularly sensitive to Country X].
PHẦN V. ROBUSTNESS PARAGRAPH — VCE
30. Draft
We further examine the sensitivity of statistical inference to alternative treatments of serial and common-year dependence. The coefficient estimates remain economically similar under [methods], although statistical precision varies across inference procedures. Given that the identifying variation in annual UCT is much closer to the number of time periods than to the number of firm-year observations, we interpret p-values cautiously and place greater emphasis on coefficient stability and exposure-based evidence.
This is a very important sentence for this paper.
PHẦN W. ROBUSTNESS SUMMARY
31. Table
| Test family | Sign stable? | Magnitude stable? | Precision stable? | Interpretation |
|---|---|---|---|---|
| Alt leverage | ||||
| Global controls | ||||
| COVID/events | ||||
| Country LOO | ||||
| VCE |
Final wording
Choose:
- robust;
- broadly robust;
- moderately robust;
- sensitive.
Avoid:
“All robustness tests confirm our hypothesis”
unless literally true.
PHẦN X. 4.5 IDENTIFICATION AND ENDOGENEITY
32. Start by stating the actual concern
Draft
The main identification concern is that annual UCT is an aggregate time-series measure and may therefore be correlated with omitted global shocks that also affect ASEAN firms' financing decisions. Conventional firm-level reverse causality is less central because the leverage decision of an individual ASEAN firm is unlikely to influence aggregate U.S.–China bilateral tension.
This should be near the beginning of identification subsection.
PHẦN Y. EXPOSURE-BASED IDENTIFICATION
33. This should be the centerpiece if available
Country-level exposure
Firm-level exposure preferred
with:
if firm-level variation exists.
PHẦN Z. WRITING EXPOSURE RESULTS
34. Draft
To strengthen identification, we exploit differences in pre-existing economic exposure to the United States and China. Because the exposure measures are determined before the major escalation in bilateral tensions, they are less likely to reflect firms' endogenous responses to contemporaneous UCT. The interaction between UCT and [China/U.S.] exposure is [sign] and statistically [significant/insignificant], indicating that firms with greater prior exposure experience [stronger/weaker] financing responses when bilateral tensions increase.
PHẦN AA. IF CHINA AND U.S. EXPOSURES DIFFER
35. This is economically valuable
Example:
but:
Interpretation
The asymmetric interaction coefficients suggest that the same bilateral tension shock creates different financing consequences depending on the side of the trade relationship to which ASEAN firms are more exposed. Greater China exposure appears to amplify supply-chain and operating-risk channels, while U.S. exposure may capture trade-diversion or market-expansion opportunities.
Link to
Jimoh et al. (2026).
PHẦN AB. PLACEBO / PRE-TREND
36. Writing
We perform falsification tests using future UCT values and, where applicable, event-time leads. Future tension does not significantly predict prior leverage, and the pre-event interaction coefficients are economically small, reducing concerns that the main findings reflect differential pre-existing trends.
If placebo fails
Do not hide.
The placebo estimates indicate some pre-existing association, limiting a strong causal interpretation.
PHẦN AC. DYNAMIC CAPITAL STRUCTURE
37. Role in Results
Dynamic model should answer:
Does UCT matter after accounting for leverage persistence?
Not:
“GMM proves causality.”
38. Writing
Because capital structures adjust gradually, we estimate dynamic specifications that include lagged leverage. The coefficient on lagged leverage is [ρ], confirming substantial persistence in financing policy. The estimated UCT coefficient remains [direction], suggesting that the baseline relationship is not solely attributable to persistent leverage differences across firms.
If System GMM
Report:
- AR(1);
- AR(2);
- Hansen/Sargan;
- instrument count;
- groups.
PHẦN AD. 2SLS RESULTS
39. Only if credible IV exists
Results paragraph must contain
- instrument;
- exclusion logic;
- first-stage coefficient;
- first-stage F/weak-IV statistic;
- second-stage UCT coefficient;
- comparison with baseline;
- limitation.
Never write only
“The IV coefficient remains significant.”
PHẦN AE. 4.6 MECHANISMS
40. Mechanism section answers
Why does UCT change capital structure?
Recommended order:
- Debt demand / investment
- Credit supply
- Financial constraints
- Rollover risk
- Trade diversion / supply-chain adjustment
PHẦN AF. DEBT-DEMAND MECHANISM
41. Possible evidence
If:
and:
then:
lower financing demand is one plausible channel.
Writing
We first examine a debt-demand channel. UCT is negatively associated with corporate investment, suggesting that firms scale back capital expenditures when bilateral tensions rise. Lower investment reduces external financing needs and is therefore consistent with the observed decline in leverage.
Do not call formal mediation unless actually implemented.
PHẦN AG. CREDIT-SUPPLY MECHANISM
42. Possible evidence
- bank lending declines;
- borrowing costs rise;
- debt issuance falls;
- ST debt access contracts.
Writing
We next examine whether external credit conditions contribute to the leverage response. During periods of higher UCT, [credit-growth/borrowing-cost/debt-issuance proxy] changes in a direction consistent with tighter financing conditions. This evidence suggests that reduced credit supply complements the debt-demand mechanism.
Important
If only aggregate credit growth:
say:
consistent with tighter credit conditions
not:
proves bank supply contraction.
PHẦN AH. FINANCIAL CONSTRAINTS
43. Writing interaction results
The UCT effect is stronger among financially constrained firms. The interaction coefficient between UCT and [constraint measure] is [ ], and the marginal effect of UCT becomes more negative as financing constraints increase. This pattern is consistent with the view that firms with fewer financing alternatives are more vulnerable to bilateral tension shocks.
Link
Bui et al. (2026).
PHẦN AI. ROLLOVER MECHANISM
44. If ST debt response strong
The concentration of the effect in short-term borrowing further supports a refinancing channel. Short-maturity debt must be rolled over frequently and therefore exposes firms more directly to changes in lender risk appetite and financing conditions.
PHẦN AJ. TRADE-DIVERSION CHANNEL
45. Do not assume all UCT channels are negative
If high U.S. exposure/beneficiary sectors show positive financing response:
Some ASEAN firms appear to respond to U.S.–China tension by increasing external financing. This result is consistent with trade-diversion and production-relocation opportunities that raise investment and working-capital needs.
Link
Jimoh et al.; ASEAN fragmentation evidence.
PHẦN AK. SUPPLY-CHAIN CHANNEL
46. If China exposure strengthens negative effect
The stronger deleveraging response among China-exposed firms is consistent with heightened supply-chain and technological-risk concerns. Chai and Wang (2026) show that deteriorating bilateral relations induce firms to adjust overseas capital allocation in response to technological restrictions and supply-chain security concerns.
Remember
Chai & Wang use a different tension index.
Do not imply they use Rogers UCT.
PHẦN AL. MECHANISM SUMMARY TABLE
47. Bảng
| Mechanism | Evidence | Direction | Strength |
|---|---|---|---|
| Investment/debt demand | Strong/Moderate/Suggestive | ||
| Credit supply | |||
| Financial constraints | |||
| Rollover | |||
| Trade diversion | |||
| Supply chain |
Strength labels
Use:
- Strong
- Moderate
- Suggestive
- Not supported
Avoid calling every mechanism “confirmed”.
PHẦN AM. 4.7 HETEROGENEITY
48. Heterogeneity is not a catalogue
Only discuss dimensions connected to theory:
- U.S. exposure;
- China exposure;
- constraints;
- cash;
- institutional quality;
- financial development;
- industry exposure.
PHẦN AN. MARGINAL EFFECTS
49. For continuous interactions
Report:
at:
- P25;
- P50;
- P75.
Table
| Moderator level | Marginal effect UCT | SE | p |
|---|---|---|---|
| P25 | |||
| P50 | |||
| P75 |
PHẦN AO. WRITE HETEROGENEITY ECONOMICALLY
50. Bad
Interaction is −0.003***.
Better
The marginal effect of UCT becomes increasingly negative as China exposure rises. At the 25th percentile of exposure the estimated effect is [ ], compared with [ ] at the 75th percentile, indicating that firms more integrated with China adjust debt more strongly during periods of heightened bilateral tension.
PHẦN AP. INSTITUTIONAL QUALITY
51. Writing
If stronger institutions cushion:
The negative UCT–leverage relation is weaker in countries with stronger institutional quality, consistent with the view that better creditor protection and information environments help firms preserve financing access during external shocks.
If stronger institutions amplify:
The result may reflect faster financial-market transmission or more market-based repricing.
Link
Vega-Gutiérrez et al.
PHẦN AQ. FINANCIAL DEVELOPMENT
52. Writing
Firms in deeper financial systems exhibit [weaker/stronger] leverage responses to UCT. This suggests that the availability of alternative financing channels [buffers/amplifies] the transmission of bilateral tension to corporate financing.
PHẦN AR. 4.8 DISCUSSION
53. Discussion must answer “So what?”
Not repeat coefficients.
Recommended 4-paragraph structure:
Paragraph 1
What main finding means for uncertainty-capital structure literature.
Paragraph 2
Why ASEAN matters.
Paragraph 3
What mechanisms imply.
Paragraph 4
What findings do not prove / boundaries.
PHẦN AS. DISCUSSION PARAGRAPH 1 — LITERATURE
54. Draft
The findings contribute to the growing literature on uncertainty and corporate financing by showing that a bilateral geopolitical and trade-related shock is associated with meaningful changes in firms' capital structures. The negative leverage response is consistent with Chowdhury et al. (2025), who document lower leverage under heightened geopolitical risk, and with Bui et al. (2026), who report lower overall and short-term leverage in Vietnam during periods of elevated U.S.–China tension. At the same time, the evidence differs from Vega-Gutiérrez et al. (2025), whose positive EPU–leverage relation highlights that financing responses depend on the nature of uncertainty and the institutional setting.
Adapt if actual result differs.
PHẦN AT. DISCUSSION PARAGRAPH 2 — ASEAN
55. Draft
The ASEAN evidence also shows why a regional approach adds information beyond a single-country design. ASEAN economies differ substantially in their dependence on U.S. demand, Chinese inputs, financial development, and institutional environments. Consistent with Jimoh et al. (2026), our exposure-based results indicate that the same bilateral tension shock does not translate uniformly across the region. This heterogeneity is economically important because it implies that U.S.–China tensions can represent both a source of financing risk and a source of trade-diversion opportunity for ASEAN firms.
PHẦN AU. DISCUSSION PARAGRAPH 3 — MECHANISMS
56. Draft
The mechanism analyses suggest that the financing response operates through multiple channels. [Lower investment] is consistent with a reduction in debt demand, while [credit-condition evidence] indicates that external financing supply may also tighten. The stronger response among financially constrained and short-term-debt-dependent firms further points to financing frictions and rollover risk. At the same time, the behavior of firms with high U.S. exposure suggests that trade-diversion opportunities can partly offset the negative risk channel.
PHẦN AV. DISCUSSION PARAGRAPH 4 — BOUNDARIES
57. Draft
These results should nevertheless be interpreted with appropriate caution. Annual UCT is an aggregate time-series measure, making it difficult to absorb all global time shocks while estimating its direct effect. Our exposure-based specifications, falsification tests, and robustness analyses reduce this concern but do not necessarily eliminate every source of endogeneity. Accordingly, we interpret the evidence as showing a robust and economically meaningful financing response to U.S.–China tension, while reserving stronger causal language for specifications with credible exposure-based identification.
PHẦN AW. HYPOTHESIS RESULTS TABLE
58. Bảng bắt buộc
| Hypothesis | Empirical test | Result | Supported? | Evidence strength |
|---|---|---|---|---|
| H1 | UCT → Leverage | Yes/No/Mixed | ||
| H2a | UCT×China Exposure | |||
| H2b | UCT×U.S. Exposure | |||
| H3 | UCT×Constraints | |||
| H4 | ST vs LT debt | |||
| H5 optional | UCT×IQ/FD |
Use
Discussion should explicitly state:
- supported;
- not supported;
- partially supported.
Not pretend all hypotheses succeed.
PHẦN AX. RESULTS CLAIM LADDER
59. Level 1 — Descriptive
UCT is negatively correlated with leverage.
60. Level 2 — Conditional association
UCT is negatively associated with leverage after controls and firm FE.
61. Level 3 — Robust association
The negative association remains across alternative measures and samples.
62. Level 4 — Identification-consistent effect
Exposure-based and falsification tests are consistent with a financing effect of UCT.
63. Level 5 — Causal
Use only if genuinely supported.
Rule
Every paragraph must use language matching the evidence level.
PHẦN AY. OVERCLAIM CHECK
64. Search manuscript for these words
- cause
- causal
- impact
- effect
- leads to
- results in
- drives
- determines
For each use:
ask:
Does the corresponding empirical design justify this verb?
Safer alternatives
- associated with;
- related to;
- consistent with;
- suggests;
- appears to;
- is accompanied by.
PHẦN AZ. DO NOT DISCUSS P-VALUE ALONE
65. Every main result paragraph should contain at least 3 of 5
- sign;
- magnitude;
- statistical precision;
- economic interpretation;
- literature comparison.
Not just:
significant at 1%.
PHẦN BA. DO NOT WRITE EVERY CONTROL
66. Controls are supporting evidence
Mention only:
- if sign confirms model plausibility;
- if economically interesting;
- if unexpected and requires explanation.
Do not spend 600 words on controls.
PHẦN BB. DO NOT REPEAT METHOD IN RESULTS
67. Results can remind, not re-explain
Bad:
Firm fixed effects control for all time-invariant firm characteristics…
again for every table.
Better:
All specifications retain the baseline firm fixed effects and controls.
Detailed justification belongs Methodology.
PHẦN BC. DO NOT REPEAT THEORY VERBATIM
68. Discussion should use theory to interpret evidence
Not reproduce Section 2.
Theory
Higher distress costs predict lower leverage.
Results discussion
The observed decline in leverage, particularly among constrained firms, is consistent with the distress-cost and financial-flexibility channels developed in H1.
PHẦN BD. CONTRADICTORY RESULTS
69. Do not hide contradictions
Example:
- baseline UCT negative;
- country exposure interaction positive;
- investment increases in some subsample.
Possible coherent story
Average risk channel negative, but opportunity channel positive for trade-diversion beneficiaries.
This can make paper more interesting.
PHẦN BE. INSIGNIFICANT ADVANCED TEST
70. Do not delete automatically
If mechanism insignificant:
We do not find statistically detectable evidence for this channel.
Then retain if:
- theoretically important;
- pre-specified;
- helps rule out an explanation.
PHẦN BF. MULTIPLE HETEROGENEITY RESULTS
71. Prioritize
Main text:
- U.S./China exposure;
- constraints;
- maturity.
Appendix:
- numerous secondary splits;
- additional industries;
- minor country groups.
Goal
Reader should remember 2–3 heterogeneity findings, not 15.
PHẦN BG. FIGURES
72. Useful figures
Figure 1
UCT over time.
Figure 2
Average leverage over time.
Figure 3
Marginal effect of UCT by China exposure.
Figure 4
Event-time coefficient plot if event design exists.
Figure 5
Leave-one-country-out coefficient plot.
Keep only figures that clarify an argument.
PHẦN BH. FIGURE 1 DISCUSSION
73. UCT time series
Do not use chart merely decoratively.
Use to identify:
- major spikes;
- 2018–2019;
- COVID;
- later escalation.
Writing
Figure 1 illustrates substantial variation in bilateral tension over the sample period, with notable spikes around [verified events]. This temporal concentration motivates the event-period and extreme-year robustness tests.
PHẦN BI. TABLE NOTES
74. Every regression table note should state
- dependent variable;
- UCT definition;
- controls;
- FE;
- VCE;
- winsorization;
- significance symbols;
- sample restriction if nonstandard.
Do not make readers search Methodology for every detail.
PHẦN BJ. MANUSCRIPT RESULTS OUTLINE
75. Suggested word allocation
4.1 Descriptives
300–450 words
4.2 Baseline
700–900 words
4.3 Robustness
600–900 words
4.4 Identification
600–900 words
4.5 Mechanisms
700–1.000 words
4.6 Heterogeneity
500–800 words
4.7 Discussion
600–900 words
Total
Approximately:
4.000–5.800 words
Can be shorter for journals with tight limits.
PHẦN BK. COMPLETE WRITING SKELETON
76. Opening Results paragraph
Table 1 presents descriptive statistics for the variables used in the empirical analysis. [Brief sample observations.] Table 3 then reports the baseline estimates of the relation between U.S.–China tension and corporate capital structure.
77. Main result
The coefficient on UCT is [ ], indicating [ ]. A one-standard-deviation increase in UCT corresponds to [economic effect].
78. Debt composition
The effect is concentrated in [ST/LT] debt, which suggests [economic mechanism].
79. Comparison with literature
This evidence is consistent with [Chowdhury/Bui] but differs from [Vega], suggesting [why].
80. Robustness
The result remains [broadly stable/sensitive] across [main tests].
81. Identification
Exposure-based specifications show [ ], providing stronger evidence that the relationship is tied to economic exposure rather than common time shocks.
82. Mechanisms
Evidence from [investment/credit/constraints/maturity] is consistent with [channels].
83. ASEAN heterogeneity
U.S. and China exposure produce [same/different] responses, consistent with Jimoh et al.
84. Closing Discussion
Taken together, the results indicate [one-sentence paper message].
PHẦN BL. ONE-SENTENCE PAPER MESSAGE
85. Students must create one
Examples only:
Negative average effect
U.S.–China tensions are associated with lower corporate leverage in ASEAN, particularly through short-term debt and among firms with greater China exposure and tighter financing constraints.
Mixed effect
U.S.–China tensions generate heterogeneous financing responses across ASEAN, with risk-driven deleveraging among China-exposed firms but stronger borrowing among firms positioned to benefit from trade diversion.
Positive average effect
ASEAN firms increase leverage during periods of heightened U.S.–China tension, consistent with external financing needs generated by trade diversion and strategic supply-chain reallocation.
Use only the version supported by actual results.
PHẦN BM. RESULTS EVIDENCE MATRIX
86. Bảng bắt buộc
| Manuscript sentence/claim | Table/Column | β / statistic | Economic magnitude | Citation | Causal language level |
|---|---|---|---|---|---|
Rule
Every major claim in Results should be traceable to:
- regression evidence;
- literature;
- correct language strength.
PHẦN BN. DISCUSSION EVIDENCE MATRIX
87. Bảng
| Discussion claim | Empirical evidence | Theory | Literature comparator | Alternative explanation |
|---|---|---|---|---|
| UCT lowers leverage | Trade-off | Chowdhury | Financing need | |
| ST debt more sensitive | Rollover | Bui | Credit maturity supply | |
| China exposure stronger | Supply chain | Jimoh/Chai | Industry composition | |
| Constraints amplify | Financing friction | Bui | Firm size | |
| Institutions cushion | Financing access | Vega | Faster repricing |
PHẦN BO. EXCEL FILE CHO BƯỚC 6
88. File đề xuất
06_RESULTS_DISCUSSION_EVIDENCE.xlsx
Sheet 1 — STORYBOARD
| Question | Answer | Evidence |
|---|
Sheet 2 — TABLE_CLAIMS
| Table | Column | Coefficient | Main claim | Citation |
|---|
Sheet 3 — HYPOTHESES
| Hypothesis | Test | Result | Supported? |
|---|
Sheet 4 — ECONOMIC_EFFECT
| DV | β | SD X | Mean Y | Effect | % Mean |
|---|
Sheet 5 — ROBUSTNESS
| Test | β | p | Direction | Interpretation |
|---|
Sheet 6 — IDENTIFICATION
| Design | Main coefficient | FE | VCE | Strength |
|---|
Sheet 7 — MECHANISMS
| Mechanism | Evidence | Result | Strength |
|---|
Sheet 8 — HETEROGENEITY
| Moderator | Interaction | Marginal effect | Interpretation |
|---|
Sheet 9 — DISCUSSION
| Claim | Data evidence | Theory | Paper comparison |
|---|
Sheet 10 — WRITING
| Section | Paragraph | Main message | Draft text |
|---|
PHẦN BP. PAPERS BẮT BUỘC KHI VIẾT DISCUSSION
89. Chowdhury et al. (2025)
Use for:
- negative geopolitical-risk leverage relation;
- debt demand;
- credit supply;
- financing flexibility.
90. Bui et al. (2026)
Use for:
- closest UCT–leverage result;
- Vietnam comparison;
- short-term leverage;
- financial constraints.
91. Vega-Gutiérrez et al. (2025)
Use for:
- competing positive uncertainty–leverage evidence;
- institutions;
- financial development.
92. Jimoh et al. (2026)
Use for:
- ASEAN heterogeneity;
- U.S./China exposure;
- trade-diversion logic.
93. Chai and Wang (2026)
Use for:
- supply-chain/technology mechanism;
- strategic corporate response.
94. Peng et al. (2025)
Use for:
- time-varying/asymmetric UCT financial effects;
- motivation for nonlinear/event sensitivity.
PHẦN BQ. RESULTS & DISCUSSION SAMPLE — NEGATIVE MAIN RESULT
95. Integrated sample passage
Our baseline estimates indicate that heightened U.S.–China tension is associated with lower corporate leverage among ASEAN firms. The coefficient on UCT is negative across the principal leverage specifications and statistically significant for [actual measures]. The magnitude is economically meaningful: a one-standard-deviation increase in UCT corresponds to a [X]-percentage-point decline in [main leverage], or approximately [Y]% of its sample mean. The result therefore supports H1 and is consistent with the view that bilateral tension increases uncertainty surrounding firms' cash flows and financing conditions.
The debt decomposition provides further insight. The negative association is concentrated in [short-term/long-term] debt. If short-term debt responds more strongly, the evidence suggests that firms reduce refinancing-sensitive borrowing and seek to limit rollover exposure during periods of elevated tension. This pattern is closely related to Bui et al. (2026), who document lower overall and short-term leverage among Vietnamese firms exposed to U.S.–China tensions.
Our findings are also broadly consistent with Chowdhury et al. (2025), who show that geopolitical risk reduces corporate leverage through contractions in both debt demand and credit supply. The present setting differs, however, because UCT captures a bilateral U.S.–China shock and the sample consists of firms across ASEAN economies that maintain heterogeneous economic links with both countries. This distinction is important because bilateral tension may create not only risk but also trade-diversion opportunities.
The robustness tests indicate that the baseline result is [broadly/moderately] stable. The coefficient retains its direction under alternative book- and market-based leverage measures, after controlling for broader global uncertainty, and when major event periods and individual ASEAN countries are excluded. Statistical precision varies under alternative inference procedures, which is unsurprising given that annual UCT provides limited independent time-series variation. We therefore emphasize the stability of the estimated magnitudes and the exposure-based tests rather than relying exclusively on conventional firm-level p-values.
The exposure analyses strengthen the economic interpretation. Firms/economies with greater pre-existing China exposure exhibit a more negative response to UCT, whereas the response associated with U.S. exposure is [weaker/more positive]. This asymmetry is consistent with Jimoh et al. (2026), who document heterogeneous ASEAN trade responses to U.S.–China tensions, and suggests that supply-chain vulnerability and trade-diversion opportunities operate simultaneously across the region.
Mechanism tests further indicate that [investment declines / credit conditions tighten / financially constrained firms react more strongly]. These patterns are consistent with both a reduction in financing demand and tighter external financing conditions. The stronger response among [constrained/high-ST-debt] firms also supports the financial-flexibility and rollover-risk channels. At the same time, the [positive/weaker] response among firms positioned to benefit from trade diversion demonstrates that UCT should not be interpreted as a purely negative regional shock.
Taken together, the evidence suggests that U.S.–China tensions have economically meaningful financing consequences for ASEAN firms, but those consequences depend on the maturity of debt, firms' financing capacity, and their pre-existing exposure to the two major economies. This regional heterogeneity helps reconcile our negative average result with evidence of positive uncertainty–leverage relations in other settings, such as Vega-Gutiérrez et al. (2025).
PHẦN BR. SAMPLE — MIXED / HETEROGENEOUS RESULT
96.
The empirical evidence does not support a uniform financing response across ASEAN. The average coefficient on UCT is [weak/mixed], but the exposure analyses reveal substantial heterogeneity. Firms with greater China exposure reduce leverage as bilateral tension increases, whereas firms with greater U.S. exposure exhibit [a smaller decline/an increase] in debt financing. This pattern suggests that a regional average masks two competing channels: supply-chain and operating-risk effects for firms dependent on China, and trade-diversion or expansion opportunities for firms positioned to serve U.S. markets.
This interpretation is consistent with Jimoh et al. (2026), who show that ASEAN countries respond asymmetrically to U.S.–China trade tensions depending on their pre-existing trade relationships. It also helps reconcile the mixed uncertainty–capital structure literature. While Chowdhury et al. (2025) document a negative leverage response to geopolitical risk, Vega-Gutiérrez et al. (2025) find a positive response to EPU in Europe. Our results suggest that both patterns can arise within ASEAN depending on economic exposure.
PHẦN BS. WHAT RESULTS BELONG IN APPENDIX?
97. Move to appendix when
- alternative VCE adds little new;
- many leave-one-country-out models;
- numerous alternative UCT transformations;
- minor control variations;
- secondary industry splits;
- detailed GMM diagnostics;
- extra placebo tests.
Keep in main text when
result changes economic interpretation.
PHẦN BT. TABLE REDUNDANCY CHECK
98. For every table ask
If this table disappeared, would the paper's conclusion change?
If:
No
consider appendix.
PHẦN BU. DISCUSSION QUALITY CHECK
99. Each major result should answer
- What did we find?
- Is it economically meaningful?
- Why might it happen?
- Which prior paper does it support/conflict with?
- What is specifically new in ASEAN?
- What alternative explanation remains?
PHẦN BV. CITATION QUALITY CHECK
100. Do not cite paper only because same keyword appears
Each citation needs a role:
- measurement;
- theory;
- empirical comparison;
- mechanism;
- ASEAN context;
- methodology.
Avoid citation stacking
Five citations at the end of every sentence without explaining relevance.
PHẦN BW. NUMERIC CONSISTENCY CHECK
101. Before finalizing
Check all:
- β;
- SE;
- p;
- N;
- firms;
- countries;
- years;
- economic significance;
- table number;
- column number.
Common error
Text says:
−0.006***
but table after rerun says:
−0.004**.
Must update prose.
PHẦN BX. SIGN LANGUAGE CHECK
102. Be precise
If DV:
higher means more long-term debt.
Then:
negative UCT coefficient means:
shorter maturity
not:
lower leverage.
Every outcome needs its own economic interpretation.
PHẦN BY. SCALE LANGUAGE CHECK
103. Percentage vs percentage point
If leverage ratio changes:
from:
to:
change is:
1.5 percentage points
and:
5% relative to mean.
Do not confuse the two.
PHẦN BZ. HYPOTHESIS WORDING CHECK
104. Results terminology
Use:
- H1 is supported;
- H1 is not supported;
- H1 receives partial support.
Avoid:
H1 is proven.
PHẦN CA. FINAL RESULTS CHECKLIST
105.
Table order follows research story.
Descriptive section is concise.
Baseline result clearly stated.
UCT coefficient reported accurately.
Economic significance calculated.
ST/LT/maturity pattern interpreted.
Formal coefficient-difference claim only if tested.
Control coefficients not over-discussed.
Robustness tests grouped by concern.
Global-shock concern discussed.
COVID/event sensitivity discussed.
Country sensitivity discussed.
VCE limitations acknowledged.
Exposure identification highlighted.
U.S. and China exposure separated.
Placebo/pre-trend result reported if available.
Dynamic result interpreted correctly.
GMM not presented as magic causal proof.
IV only discussed if instrument credible.
Debt-demand mechanism discussed.
Credit-supply mechanism discussed carefully.
Constraints mechanism discussed.
Rollover channel discussed.
Trade-diversion channel considered.
Supply-chain channel considered.
Heterogeneity limited to theory-relevant tests.
Marginal effects used for continuous interactions.
Literature comparisons are balanced.
Chowdhury compared appropriately.
Bui compared appropriately.
Vega opposing evidence included.
Jimoh used for ASEAN asymmetry.
Chai & Wang not misused as Rogers UCT source.
Peng used for time-varying UCT relevance.
Causal language matches identification strength.
Every major claim maps to a table/column.
Every number in prose matches final table.
Appendix receives secondary robustness.
Main text contains only economically important tables.
Final one-sentence paper message has been written.
PHẦN CB. FINAL DELIVERABLE
Bước 6 hoàn thành khi nhóm nộp:
- Results Storyboard
- Table-to-Claim Matrix
- Hypothesis Results Table
- Economic Significance Summary
- Robustness Summary
- Identification Strength Summary
- Mechanism Summary
- Heterogeneity Summary
- Results Evidence Matrix
- Discussion Evidence Matrix
- Final Results & Discussion draft, khoảng 3.500–5.500 từ
PHẦN CC. CHUYỂN SANG BƯỚC 7
Sau Bước 6, gần như toàn bộ “ruột” của paper đã hoàn thành:
- Methodology;
- Baseline Results;
- Robustness;
- Identification;
- Mechanisms;
- Heterogeneity;
- Theory;
- Literature;
- Hypotheses;
- Results & Discussion.
Bước tiếp theo sẽ là:
BƯỚC 7. VIẾT INTRODUCTION VÀ XÂY DỰNG CONTRIBUTION STORY
Ở Bước 7, sinh viên sẽ không viết Introduction từ đầu theo cảm tính mà lấy trực tiếp:
- main finding;
- economic significance;
- ASEAN motivation;
- research gap;
- identification strength;
- contributions
đã được xác định ở Bước 1–6 để tạo Introduction hoàn chỉnh.
BƯỚC 7. VIẾT INTRODUCTION VÀ XÂY DỰNG CONTRIBUTION STORY
Bài mẫu: U.S.–China Tension and Corporate Capital Structure in ASEAN Countries
Mục tiêu đầu ra: Từ toàn bộ evidence đã hoàn thiện ở Bước 1–6, sinh viên phải viết được một phần Introduction hoàn chỉnh, có research story rõ, không chung chung, không lặp literature review, và làm nổi bật contribution thực sự của paper.
Nguyên tắc của Bước 7
Introduction phải trả lời nhanh:
- Vấn đề kinh tế quan trọng là gì?
- Vì sao U.S.–China tension đáng nghiên cứu?
- Vì sao capital structure là outcome quan trọng?
- Vì sao ASEAN là setting đặc biệt?
- Literature hiện biết gì?
- Chưa biết gì?
- Paper này làm gì?
- Main findings là gì?
- Contribution là gì?
Introduction không phải:
“U.S.–China relations are important. Capital structure is important. ASEAN is important.”
mà phải tạo thành một câu chuyện nghiên cứu liền mạch.
PHẦN A. THÀNH PHẨM CUỐI BƯỚC 7
Bước 7 chỉ hoàn thành khi nhóm có:
- Introduction Storyboard
- Hook Options Table
- Problem Statement
- Why ASEAN Paragraph
- Research Gap Paragraph
- What We Do Paragraph
- Main Findings Paragraph
- Contribution Paragraph
- Full Introduction draft, khoảng 1.500–2.200 từ
- Final one-sentence paper message
- Introduction Claim–Evidence Matrix
PHẦN B. INPUT BẮT BUỘC TỪ BƯỚC 1–6
1. Không viết Introduction trước khi có các câu trả lời sau
Từ Bước 1
- UCT là gì?
- sample ASEAN-X là gì?
- DVs là gì?
- main methodology là gì?
Từ Bước 2
- baseline sign là gì?
- magnitude bao nhiêu?
- ST vs LT debt khác nhau không?
Từ Bước 3
- main result robust đến đâu?
- COVID/global-shock concern thế nào?
Từ Bước 4
- strongest identification là gì?
- mechanism nào mạnh nhất?
- heterogeneity nào đáng nhớ nhất?
Từ Bước 5
- research gap là gì?
- contribution thật sự là gì?
- hypotheses là gì?
Từ Bước 6
- one-sentence paper message là gì?
PHẦN C. INTRODUCTION STORYBOARD
2. Bảng bắt buộc
| Introduction element | Nội dung paper |
|---|---|
| Hook | |
| Economic problem | |
| Why capital structure? | |
| Why ASEAN? | |
| What literature knows | |
| Research gap | |
| What we do | |
| Main finding | |
| Strongest mechanism | |
| Strongest heterogeneity | |
| Strongest identification | |
| Contribution 1 | |
| Contribution 2 | |
| Contribution 3 |
Rule
Nếu một ô chưa có câu trả lời:
không được viết Introduction bằng prose trước.
PHẦN D. STRUCTURE OF INTRODUCTION
3. Recommended structure
Paragraph 1 — Hook + Economic Context
Paragraph 2 — Why This Matters for Corporate Financing
Paragraph 3 — Why ASEAN?
Paragraph 4 — What Prior Literature Knows
Paragraph 5 — Research Gap
Paragraph 6 — What We Do
Paragraph 7 — Main Findings
Paragraph 8 — Contributions
Paragraph 9 — Paper Roadmap
Tổng:
8–9 paragraphs
thường đủ.
PHẦN E. PARAGRAPH 1 — HOOK
4. Hook không nên là câu quá chung
Bad:
U.S.–China relations have become increasingly important in recent years.
Better:
U.S.–China tensions have moved beyond bilateral diplomacy to reshape trade, technology, supply chains, investment, and financial-market conditions across third countries.
Core citations
- Rogers et al. (2024)
- Peng et al. (2025)
- Chai and Wang (2026)
- Jimoh et al. (2026)
PHẦN F. HOOK OPTIONS
5. Option A — Global spillover hook
U.S.–China tensions have become an important source of global economic and financial uncertainty, affecting not only bilateral trade but also supply-chain configurations, investment decisions, and financial markets in third countries.
Best if paper emphasizes:
- regional spillovers;
- ASEAN.
6. Option B — Corporate-finance hook
Firms increasingly make financing decisions in an environment shaped by geopolitical and trade fragmentation. U.S.–China tensions can alter expected cash flows, investment opportunities, refinancing conditions, and access to external capital, making corporate capital structure a key margin of adjustment.
Best if paper emphasizes:
- financing mechanisms.
7. Option C — ASEAN hook
ASEAN sits at the center of the economic reallocation generated by U.S.–China tensions. Its firms may simultaneously face supply-chain disruption from China and trade-diversion opportunities linked to U.S. demand, making the region an unusually informative setting in which to study corporate financing responses.
Best if regional contribution is strongest.
PHẦN G. PARAGRAPH 1 — FULL DRAFT
8. Draft mẫu
U.S.–China tensions have evolved from a bilateral political issue into an important source of global economic and financial uncertainty. Changes in bilateral relations now affect trade policy, technology restrictions, investment screening, supply-chain configuration, and financial markets. Rogers et al. (2024) develop a news-based U.S.–China Tension Index that captures the intensity of these tensions, while subsequent studies show that bilateral tensions have meaningful effects on stock markets and corporate strategic decisions (Peng et al., 2025; Chai & Wang, 2026). These developments suggest that U.S.–China tensions can influence not only operating and investment decisions but also firms' financing policies.
PHẦN H. PARAGRAPH 2 — WHY CAPITAL STRUCTURE?
9. Transition from macro shock to corporate finance
Question:
Why should a tension shock affect debt financing?
Channels
- distress risk;
- cash-flow uncertainty;
- credit supply;
- financing need;
- financial flexibility;
- rollover risk.
10. Draft mẫu
Corporate capital structure provides a particularly important channel through which such tensions may be transmitted to the firm level. Greater uncertainty can raise expected distress costs, reduce lenders' willingness to supply credit, and increase the value of preserving financial flexibility, implying lower leverage. At the same time, firms facing weaker internal cash flows or new trade-diversion opportunities may increase borrowing to finance working capital or expansion. The net effect is therefore theoretically ambiguous and depends on both the nature of the shock and the economic environment in which firms operate.
PHẦN I. PARAGRAPH 3 — WHY ASEAN?
11. Đây là paragraph rất quan trọng
Không chỉ viết:
ASEAN is economically important.
Phải viết:
ASEAN provides identifying/economic variation.
Core logic
ASEAN firms differ in:
- China input exposure;
- U.S. export exposure;
- trade dependence;
- financial development;
- institutional quality;
- industry structure.
12. Draft mẫu
ASEAN offers a particularly informative setting for examining this question because its economies are deeply connected to both China and the United States but in markedly different ways. Some economies rely heavily on Chinese intermediate inputs and regional supply chains, whereas others are strongly exposed to U.S. export demand or benefit from production relocation and trade diversion. Jimoh et al. (2026) show that ASEAN countries respond asymmetrically to U.S.–China trade tensions depending on their pre-existing trade exposure. This heterogeneity implies that a common bilateral tension shock may generate very different financing responses across ASEAN firms.
PHẦN J. PARAGRAPH 4 — WHAT LITERATURE KNOWS
13. Do not list papers
Synthesize 3 literatures:
Literature 1
UCT and financial/corporate outcomes.
Literature 2
Uncertainty/GPR and capital structure.
Literature 3
ASEAN trade exposure.
14. Draft mẫu
Existing evidence points to economically meaningful but heterogeneous financing responses to uncertainty. Chowdhury et al. (2025) show that geopolitical risk reduces corporate leverage through both debt-demand and credit-supply channels. The closest direct evidence on U.S.–China tensions is provided by Bui et al. (2026), who document lower overall and short-term leverage among Vietnamese listed firms. However, Vega-Gutiérrez et al. (2025) find that economic policy uncertainty is positively associated with leverage in European firms and emphasize the moderating role of institutional and financial conditions. Together, these studies indicate that uncertainty does not have a uniform effect on capital structure.
PHẦN K. PARAGRAPH 5 — RESEARCH GAP
15. Gap phải có cấu trúc 3 tầng
Gap 1 — Shock-specific
UCT khác GPR/EPU.
Gap 2 — Geography
Existing direct UCT-leverage evidence mainly Vietnam.
Gap 3 — Mechanism/heterogeneity
Little evidence on:
- ASEAN multi-country response;
- U.S. vs China exposure;
- ST vs LT debt;
- debt maturity;
- financial constraints;
- trade diversion.
16. Draft gap paragraph
Despite these advances, three gaps remain. First, most capital-structure studies examine broad measures of economic-policy or geopolitical uncertainty rather than bilateral U.S.–China tension. Second, the closest direct UCT–leverage evidence comes from a single-country Vietnamese setting, leaving open whether the relationship generalizes across ASEAN economies with very different trade and financial structures. Third, existing studies provide limited evidence on whether pre-existing exposure to the United States and China, financing constraints, and debt maturity shape firms' responses to bilateral tension. These gaps are particularly important in ASEAN, where the same external shock can create both supply-chain risks and trade-diversion opportunities.
PHẦN L. PARAGRAPH 6 — WHAT WE DO
17. Introduction must state exactly what paper does
Include:
- sample;
- data;
- X;
- Y;
- timing;
- main empirical strategy;
- stronger exposure design.
Do not include every control.
18. Draft mẫu
We address these questions using a panel of publicly listed firms from [ASEAN-X countries] over [sample period]. We measure bilateral tension using the U.S.–China Tension Index of Rogers et al. (2024) and examine multiple dimensions of capital structure, including total leverage, short-term leverage, long-term leverage, and debt maturity. Our baseline specifications relate UCT in year $t$ to firms' financing outcomes in year $t+1$ and include firm fixed effects and a comprehensive set of firm-level and macroeconomic controls. Because annual UCT varies primarily over time, we complement the direct specification with exposure-based designs that interact UCT with predetermined U.S. and China exposure and allow stronger time fixed effects.
PHẦN M. PARAGRAPH 7 — MAIN FINDINGS
19. Introduction should report results
Do not hide results until Section 4.
Need include:
- main sign;
- economic magnitude;
- debt composition;
- strongest robustness/identification;
- strongest heterogeneity.
Keep concise.
PHẦN N. MAIN FINDINGS TEMPLATE — NEGATIVE RESULT
20. Draft
We obtain four main findings. First, higher U.S.–China tension is associated with lower corporate leverage among ASEAN firms. A one-standard-deviation increase in UCT is associated with a [X]-percentage-point decline in [main leverage], equivalent to approximately [Y]% of its sample mean. Second, the decline is concentrated in [short-term/long-term] debt, suggesting that [rollover/refinancing/long-term financing] is an important adjustment margin. Third, the results remain [broadly] robust to alternative capital-structure measures, global uncertainty controls, event-period exclusions, country sensitivity tests, and alternative inference procedures. Fourth, the effect is stronger among [China-exposed / financially constrained / high short-term-debt] firms, while firms with greater [U.S. exposure] exhibit [different response], highlighting substantial heterogeneity within ASEAN.
PHẦN O. MAIN FINDINGS TEMPLATE — MIXED RESULT
21. Draft
The results reveal substantial heterogeneity rather than a uniform ASEAN-wide response. Although the average UCT coefficient is [weak/mixed], firms with greater China exposure reduce leverage as bilateral tensions rise, whereas firms with greater U.S. exposure exhibit [a weaker decline/an increase] in debt financing. This asymmetry is consistent with the coexistence of supply-chain risk and trade-diversion opportunities across ASEAN. The findings remain robust to [main robustness tests] and are strongest among [financially constrained / short-term-debt-dependent] firms.
PHẦN P. PARAGRAPH 8 — CONTRIBUTIONS
22. Contributions should be 3 or 4, not 8
Recommended:
Contribution 1
To uncertainty/capital structure literature.
Contribution 2
To UCT literature.
Contribution 3
To ASEAN literature.
Contribution 4
To financing mechanism / identification.
PHẦN Q. CONTRIBUTION 1 — SHOCK
23. Draft
First, we extend the uncertainty–capital structure literature by examining a bilateral U.S.–China tension measure rather than a broad domestic policy or geopolitical-risk index.
Comparator
- Chowdhury;
- Vega-Gutiérrez.
PHẦN R. CONTRIBUTION 2 — ASEAN
24. Draft
Second, we extend the emerging UCT–leverage literature beyond a single-country setting and show that financing responses differ systematically across ASEAN economies and firms with different pre-existing exposure to the United States and China.
Comparator
- Bui et al.;
- Jimoh et al.
PHẦN S. CONTRIBUTION 3 — DEBT COMPOSITION
25. Draft
Third, we examine multiple financing margins rather than a single leverage ratio, allowing us to distinguish total debt from short-term and long-term borrowing and to identify the role of debt maturity and rollover exposure.
PHẦN T. CONTRIBUTION 4 — IDENTIFICATION / MECHANISM
26. Draft
Finally, our exposure-based analyses help separate a common global tension shock from economically meaningful differences in firms' and countries' exposure, while mechanism tests provide evidence on financing constraints, credit conditions, trade diversion, and supply-chain risk.
Only keep if these analyses are actually in paper.
PHẦN U. FULL CONTRIBUTION PARAGRAPH
27. Draft mẫu
This study contributes to the literature in four ways. First, we extend research on uncertainty and corporate financing by focusing on bilateral U.S.–China tension rather than broad domestic or geopolitical uncertainty measures. Second, we move beyond the existing single-country evidence and examine a multi-country ASEAN setting characterized by substantial differences in trade exposure, institutions, and financial development. Third, we study multiple financing margins, including short- and long-term debt and debt maturity, which allows us to identify whether firms adjust overall leverage or primarily refinancing-sensitive borrowing. Fourth, our exposure-based and mechanism analyses help distinguish general uncertainty effects from economically meaningful U.S. and China exposure and provide evidence on the channels through which bilateral tension affects corporate financing.
PHẦN V. CONTRIBUTION TEST
28. For every contribution ask
If this analysis disappeared from the paper, could we still claim the contribution?
If yes:
contribution is probably too vague.
PHẦN W. CONTRIBUTION MATRIX
29. Bảng
| Contribution | Prior paper comparator | New evidence | Table supporting it | Keep? |
|---|---|---|---|---|
| Bilateral UCT | GPR/EPU papers | UCT evidence | ||
| ASEAN multi-country | Bui Vietnam | ASEAN-X | ||
| Debt composition | Bui/Chowdhury | ST/LT/DM | ||
| Exposure | Jimoh | UCT×US/China Exposure | ||
| Mechanism | Chowdhury | Constraints/investment/credit |
PHẦN X. PARAGRAPH 9 — ROADMAP
30. Keep it short
The remainder of the paper proceeds as follows. Section 2 reviews the related literature and develops the hypotheses. Section 3 describes the data and empirical methodology. Section 4 presents the main empirical results and robustness analyses. Section 5 reports identification, mechanism, and heterogeneity tests. Section 6 concludes.
Adjust section numbers.
PHẦN Y. FULL INTRODUCTION SKELETON
31. Paragraph-by-paragraph
P1
UCT global importance.
P2
Why financing / capital structure?
P3
Why ASEAN?
P4
Prior literature.
P5
Gap.
P6
What we do.
P7
Main findings.
P8
Contributions.
P9
Roadmap.
PHẦN Z. INTRODUCTION WORD ALLOCATION
32. Recommended
| Paragraph | Words |
|---|---|
| Hook | 150–220 |
| Financing relevance | 150–220 |
| Why ASEAN | 180–250 |
| Prior literature | 180–250 |
| Gap | 180–250 |
| What we do | 180–250 |
| Findings | 220–320 |
| Contributions | 250–350 |
| Roadmap | 50–80 |
Total:
approximately:
1.500–2.100 words
PHẦN AA. INTRODUCTION CLAIM–EVIDENCE MATRIX
33. Bảng bắt buộc
| Introduction claim | Evidence/source | Verified? | Paragraph |
|---|---|---|---|
| UCT affects markets | Peng | P1 | |
| Bilateral tension changes corporate decisions | Chai & Wang | P1 | |
| GPR affects leverage | Chowdhury | P4 | |
| UCT affects Vietnam leverage | Bui | P4 | |
| EPU can increase leverage | Vega | P4 | |
| ASEAN response is heterogeneous | Jimoh | P3/P5 | |
| Main paper result | Our Table X | P7 | |
| Exposure result | Our Table X | P7 | |
| Main contribution | Our design | P8 |
PHẦN AB. DO NOT OVERLOAD INTRODUCTION WITH METHOD
34. Include only enough method to establish credibility
Good:
We use firm fixed effects and exposure-based designs.
Bad:
We winsorize variables at 1% and 99%, cluster standard errors by…, construct market leverage as…
Those details belong Methodology.
PHẦN AC. DO NOT WRITE LITERATURE REVIEW TWICE
35. Introduction literature = compressed
Introduction:
3–6 key studies
Literature Review:
more detailed streams.
Rule
If P4 becomes 600 words:
too long.
PHẦN AD. DO NOT HIDE THE CLOSEST PAPER
36. Bui et al. (2026) must appear clearly
Bad strategy:
avoid citing closest direct paper to make novelty look bigger.
Correct:
Closest evidence exists for Vietnam, and we explain exactly how ASEAN multi-country design goes beyond it.
This increases credibility.
PHẦN AE. DIFFERENTIATE FROM BUI ET AL. (2026)
37. Potential differences
Only use actual differences:
- ASEAN multi-country sample;
- exposure heterogeneity;
- U.S. vs China exposure;
- debt maturity;
- institutions/financial development;
- stronger exposure FE;
- regional trade-diversion mechanism.
Do not claim
First paper ever on UCT and capital structure.
if Bui already exists.
PHẦN AF. DIFFERENTIATE FROM CHOWDHURY ET AL. (2025)
38.
Chowdhury:
- broad GPR;
- U.S. firms;
- capital structure.
Current:
- bilateral UCT;
- ASEAN firms;
- exposure asymmetry.
Draft
Unlike broad geopolitical-risk measures, UCT captures a specific bilateral relationship whose economic consequences can differ depending on firms' links to the United States and China.
PHẦN AG. DIFFERENTIATE FROM JIMOH ET AL. (2026)
39.
Jimoh:
- trade outcomes;
- ASEAN;
- exposure.
Current:
- firm capital structure;
- corporate finance;
- debt composition.
Draft
We bring this exposure logic from international trade into corporate finance by examining whether heterogeneous trade dependence translates into heterogeneous financing responses.
PHẦN AH. PAPER TITLE CHECK
40. Current working title
U.S.–China Tension and Corporate Capital Structure in ASEAN Countries
Potential stronger titles later:
Option 1
U.S.–China Tension and Corporate Capital Structure: Evidence from ASEAN
Option 2
Between Two Giants: U.S.–China Tension and Corporate Financing in ASEAN
Option 3
U.S.–China Tension, Trade Exposure, and Corporate Capital Structure in ASEAN
Option 4
Geopolitical Tension and Corporate Financing: Evidence from U.S.–China Exposure in ASEAN
Do not finalize title until results/contribution are settled.
PHẦN AI. ONE-SENTENCE PAPER MESSAGE
41. Must be consistent with actual result
Negative result
U.S.–China tensions are associated with lower leverage among ASEAN firms, particularly through short-term debt and among firms with greater China exposure and tighter financing constraints.
Heterogeneous result
U.S.–China tensions generate asymmetric financing responses across ASEAN, with deleveraging among China-exposed firms and stronger borrowing among firms positioned to benefit from U.S.-oriented trade diversion.
Positive result
ASEAN firms increase leverage during periods of heightened U.S.–China tension, particularly where trade-diversion opportunities raise external financing needs.
PHẦN AJ. ABSTRACT TEST
42. Introduction must contain everything needed later for abstract
By end of Step 7, student should know:
- question;
- sample;
- X;
- Y;
- method;
- main finding;
- magnitude;
- heterogeneity;
- contribution.
But abstract is written later.
PHẦN AK. OPENING SENTENCE TEST
43. Opening sentence should not require 5 citations
It should be readable and direct.
Good
U.S.–China tensions increasingly shape the economic environment in which firms make investment and financing decisions.
Better than
In the era of globalization and economic integration, geopolitical uncertainty has become an increasingly important issue affecting economies around the world.
Avoid generic filler.
PHẦN AL. FIRST PAGE TEST
44. By the end of page 1, reader should know
- UCT matters;
- capital structure question;
- ASEAN setting;
- key literature conflict;
- why paper is needed.
If gap first appears on page 3:
Introduction is too slow.
PHẦN AM. MAIN FINDING TEST
45. Main result should appear before contribution paragraph
Do not write:
“We make three contributions…”
before telling the reader what you find.
Results justify contributions.
PHẦN AN. ECONOMIC MAGNITUDE IN INTRODUCTION
46. Include one magnitude if meaningful
Example:
A one-standard-deviation increase in UCT is associated with a X-percentage-point reduction in leverage, equivalent to Y% of the sample mean.
Do not include 5 coefficients.
One key magnitude is enough.
PHẦN AO. ROBUSTNESS IN INTRODUCTION
47. One sentence only
Example:
The findings remain robust to alternative leverage measures, global uncertainty controls, event-period exclusions, and country-sensitivity tests.
Do not list 15 robustness checks.
PHẦN AP. IDENTIFICATION IN INTRODUCTION
48. Mention strongest design
Example:
Exposure-based specifications using predetermined U.S. and China exposure and stronger time fixed effects yield similar conclusions.
If firm-level exposure with Country×Year FE
say so:
Our strongest specifications compare firms with different pre-existing exposure within the same country-year.
This is compelling.
PHẦN AQ. MECHANISM IN INTRODUCTION
49. Include only strongest 1–2 mechanisms
Examples:
- financial constraints;
- rollover risk;
- investment/debt demand;
- supply-chain exposure.
Do not report every exploratory channel.
PHẦN AR. HETEROGENEITY IN INTRODUCTION
50. Include only economically central heterogeneity
For this paper, likely:
- China exposure;
- U.S. exposure;
- financial constraints.
Not every industry split.
PHẦN AS. CONTRIBUTION LANGUAGE
51. Avoid exaggerated phrases
Bad:
- groundbreaking;
- pioneering;
- unprecedented;
- first-ever.
Use only if rigorously verified.
Better
- extends;
- provides new evidence;
- complements;
- shows;
- highlights;
- documents.
PHẦN AT. RESEARCH GAP LANGUAGE
52. Avoid
“There is a lack of studies.”
Better:
“Existing studies establish X, but leave unresolved Y.”
This makes gap substantive.
PHẦN AU. INTRODUCTION EVIDENCE FLOW
53. Ideal flow
Every paragraph must push this chain forward.
PHẦN AV. SAMPLE FULL INTRODUCTION — NEGATIVE RESULT TEMPLATE
54. Draft
U.S.–China tensions increasingly shape the economic environment in which firms make investment and financing decisions. Bilateral tensions now extend beyond tariffs to encompass technology restrictions, investment screening, geopolitical disputes, and supply-chain reconfiguration. Rogers et al. (2024) develop a news-based U.S.–China Tension Index that captures the intensity of these developments. Subsequent research shows that such tensions affect financial markets and corporate strategic behavior. Peng et al. (2025) document time-varying and asymmetric stock-market responses to UCT, while Chai and Wang (2026) show that deteriorating China–U.S. relations influence firms' overseas capital allocation through technological and supply-chain channels. These findings raise a natural corporate-finance question: do U.S.–China tensions also alter firms' financing structures?
Capital structure is a particularly important margin of adjustment because geopolitical tension can affect both firms' demand for financing and lenders' willingness to supply debt. Greater uncertainty can increase expected distress costs, weaken future cash-flow visibility, tighten credit conditions, and increase the value of financial flexibility, all of which can reduce leverage. However, the prediction is not unambiguous. Firms whose internal cash flows decline may rely more heavily on external debt, and firms that benefit from trade diversion may need additional financing to expand production capacity or working capital. The financing response to U.S.–China tension may therefore depend on firms' economic exposure and the institutional environment in which they operate.
ASEAN provides an especially informative setting for studying these competing channels. Its economies are deeply integrated with both China and the United States but occupy markedly different positions in regional and global value chains. Some firms depend heavily on Chinese suppliers and intermediate inputs, while others are strongly exposed to U.S. demand or benefit from production relocation away from China. Jimoh et al. (2026) show that ASEAN countries respond asymmetrically to U.S.–China trade tensions depending on their pre-existing trade exposure. The same bilateral tension shock may therefore generate financing pressure for some firms while creating expansion opportunities for others.
Existing research provides mixed evidence on the relationship between uncertainty and corporate financing. Chowdhury et al. (2025) show that geopolitical risk reduces corporate leverage through contractions in both debt demand and credit supply. The closest direct evidence on U.S.–China tensions is provided by Bui et al. (2026), who document lower overall and short-term leverage among Vietnamese listed firms. In contrast, Vega-Gutiérrez et al. (2025) report a positive association between economic policy uncertainty and leverage in European firms and show that institutional and financial conditions moderate the relationship. These results indicate that the financing response to uncertainty depends on both the type of shock and the surrounding economic environment.
Despite these advances, important questions remain unresolved. Most capital-structure studies focus on broad domestic policy or geopolitical-risk measures rather than bilateral U.S.–China tension. The direct UCT–leverage evidence is also concentrated in a single-country Vietnamese setting, leaving open whether the relationship generalizes across the diverse economies of ASEAN. More importantly, existing work provides limited evidence on whether firms' pre-existing exposure to the United States and China, debt maturity, and financing constraints shape the response. These gaps are economically important because the same U.S.–China tension can represent both supply-chain risk and trade-diversion opportunity within ASEAN.
We address these questions using a panel of publicly listed firms from [ASEAN-X countries] over [sample period]. We measure bilateral tension using the U.S.–China Tension Index of Rogers et al. (2024) and examine multiple dimensions of corporate capital structure, including total, short-term, and long-term leverage and debt maturity. Our baseline specifications relate UCT in year $t$ to financing outcomes in year $t+1$ and include firm fixed effects and firm- and country-level controls. Because annual UCT varies primarily over time, we complement the direct specification with exposure-based analyses that interact UCT with predetermined U.S. and China exposure and permit stronger time fixed effects.
We obtain several findings. First, higher U.S.–China tension is associated with lower corporate leverage among ASEAN firms. A one-standard-deviation increase in UCT is associated with a [X]-percentage-point reduction in [main leverage], corresponding to approximately [Y]% of its sample mean. Second, the reduction is concentrated in [short-term debt], indicating that refinancing-sensitive borrowing is an important margin of adjustment. Third, the results are [broadly] robust to alternative capital-structure measures, broader global uncertainty controls, event-period exclusions, country sensitivity, and alternative inference procedures. Fourth, exposure-based analyses show that [China exposure amplifies the negative leverage response, whereas U.S. exposure produces a weaker or different response]. The effect is also stronger among [financially constrained / high-short-term-debt] firms.
This study contributes to the literature in four ways. First, we extend research on uncertainty and corporate financing by focusing on a bilateral U.S.–China tension measure rather than broad economic-policy or geopolitical-risk indices. Second, we move beyond the existing single-country evidence and provide a multi-country ASEAN perspective, where firms differ substantially in their connections to the two major economies. Third, we examine several margins of financing adjustment, including short- and long-term debt and debt maturity, which allows us to identify whether firms respond through overall deleveraging or changes in refinancing exposure. Finally, our exposure-based and mechanism analyses help distinguish common global uncertainty from economically meaningful differences in U.S. and China exposure and provide evidence on the channels through which bilateral tensions affect corporate financing.
The remainder of the paper is organized as follows. Section 2 reviews the related literature and develops the hypotheses. Section 3 describes the data and empirical methodology. Section 4 presents the main results and robustness tests. Section 5 reports identification, mechanism, and heterogeneity analyses. Section 6 concludes.
PHẦN AW. IF RESULTS ARE MIXED
55. Replace findings and contribution story
Do not force negative H1 story.
Better story
The central finding is not a uniform leverage decline but asymmetric financing responses across ASEAN exposure.
Then title may become:
U.S.–China Tension, Trade Exposure, and Corporate Capital Structure in ASEAN
This can be stronger than a weak average effect.
PHẦN AX. INTRODUCTION REVISION CHECK
56. First revision
Delete:
- generic geopolitical background;
- textbook capital structure;
- repeated definitions;
- long lists of countries.
57. Second revision
Highlight:
- gap;
- findings;
- contribution.
If those three cannot be located in 30 seconds:
rewrite.
58. Third revision
Check every citation.
Each claim must match original source.
PHẦN AY. TITLE–INTRODUCTION CONSISTENCY
59. Title must match strongest result
If strongest evidence is exposure heterogeneity:
title should mention:
Trade Exposure
If strongest evidence is debt maturity:
could mention:
Debt Structure
If average leverage result is weak:
do not title paper as if direct average effect is the main contribution.
PHẦN AZ. ABSTRACT PREPARATION TABLE
60. Fill now, write abstract later
| Element | Answer |
|---|---|
| Research question | |
| Sample | |
| Period | |
| Main X | |
| Main Y | |
| Main method | |
| Main coefficient | |
| Economic magnitude | |
| Main heterogeneity | |
| Main contribution |
This will be used in a later step.
PHẦN BA. INTRODUCTION EXCEL FILE
61. File đề xuất
07_INTRODUCTION_EVIDENCE.xlsx
Sheet 1 — STORYBOARD
| Element | Content |
|---|
Sheet 2 — HOOK
| Hook option | Evidence | Keep? |
|---|
Sheet 3 — GAP
| Prior literature | Known | Unknown | Our response |
|---|
Sheet 4 — FINDINGS
| Finding | Table | Magnitude | Keep in intro? |
|---|
Sheet 5 — CONTRIBUTIONS
| Contribution | Comparator | Evidence | Keep? |
|---|
Sheet 6 — CLAIMS
| Intro sentence | Source | Verified? |
|---|
Sheet 7 — DRAFT
| Paragraph | Main message | Draft |
|---|
PHẦN BB. PAPERS BẮT BUỘC TRONG INTRODUCTION
62.
Rogers et al. (2024)
→ UCT.
Peng et al. (2025)
→ financial-market relevance.
Chai and Wang (2026)
→ corporate strategic/supply-chain response.
Jimoh et al. (2026)
→ ASEAN heterogeneity.
Chowdhury et al. (2025)
→ GPR–capital structure.
Vega-Gutiérrez et al. (2025)
→ competing uncertainty–leverage evidence.
Bui et al. (2026)
→ closest direct UCT–leverage evidence.
Not every one must appear in first paragraph.
PHẦN BC. INTRODUCTION CHECKLIST
63.
Opening sentence is specific.
P1 establishes UCT relevance.
P2 explains why financing matters.
P3 explains why ASEAN is special.
ASEAN is not described as homogeneous.
P4 synthesizes literature rather than listing papers.
Chowdhury et al. included.
Vega-Gutiérrez opposing evidence included.
Bui et al. closest paper included.
Research gap has at least 2–3 substantive layers.
Gap is not merely “no prior study”.
What-we-do paragraph states sample, X, Y and design.
Main findings appear before contributions.
Economic magnitude included if meaningful.
Strongest exposure result included.
Strongest mechanism included.
Robustness summarized in one sentence.
Identification strength stated accurately.
Contribution 1 has comparator.
Contribution 2 has comparator.
Contribution 3 has actual evidence.
No unsupported “first study” claim.
No generic filler.
No methodology overload.
No duplicated literature review.
Roadmap is short.
Title matches actual strongest result.
Every factual citation checked.
Every empirical number matches final table.
Introduction length around 1.500–2.200 words.
One-sentence paper message finalized.
PHẦN BD. FINAL DELIVERABLE
Bước 7 hoàn thành khi nhóm nộp:
- Introduction Storyboard
- Research Gap Table
- Contribution Matrix
- Introduction Claim–Evidence Matrix
- Full Introduction draft
- One-sentence paper message
- Working title
- Abstract Preparation Table
PHẦN BE. CHUYỂN SANG BƯỚC 8
Sau Bước 7, paper đã có:
- Methodology;
- Results;
- Robustness;
- Identification;
- Mechanisms;
- Heterogeneity;
- Literature Review;
- Hypotheses;
- Introduction.
Bước tiếp theo:
BƯỚC 8. VIẾT CONCLUSION, IMPLICATIONS VÀ LIMITATIONS
Bước 8 sẽ hướng dẫn sinh viên:
- tóm main findings mà không copy Results;
- viết theoretical implications;
- viết managerial/financial implications;
- viết ASEAN policy implications;
- viết limitations trung thực nhưng không tự phá paper;
- viết future research;
- khóa contribution cuối cùng.
BƯỚC 8. VIẾT CONCLUSION, IMPLICATIONS VÀ LIMITATIONS
Bài mẫu: U.S.–China Tension and Corporate Capital Structure in ASEAN Countries
Mục tiêu đầu ra: Viết hoàn chỉnh phần Conclusion của paper theo cách ngắn, rõ, nhất quán với toàn bộ evidence từ Bước 1–7 và không lặp lại Results.
Nguyên tắc của Bước 8
Conclusion phải trả lời:
- Paper nghiên cứu câu hỏi gì?
- Main findings là gì?
- Findings bổ sung gì cho theory/literature?
- Findings có ý nghĩa gì với firms, lenders và policymakers?
- Findings có giới hạn ở đâu?
- Future research nên đi theo hướng nào?
Conclusion không phải:
“Table 3 shows…, Table 4 shows…, Table 5 shows…”
và cũng không phải:
“This study has several limitations.”
rồi liệt kê chung chung.
Conclusion phải là phần chốt câu chuyện nghiên cứu.
PHẦN A. THÀNH PHẨM CUỐI BƯỚC 8
Bước 8 chỉ hoàn thành khi nhóm có:
- Conclusion Storyboard
- Main Finding Summary
- Theoretical Implications Table
- Managerial Implications Table
- Policy Implications Table
- Limitations Table
- Future Research Table
- Final Contribution Lock
- Full Conclusion draft, khoảng 900–1.400 từ
- One-paragraph short conclusion, 250–350 từ để dùng nếu journal yêu cầu ngắn
PHẦN B. CONCLUSION STORYBOARD
1. Trước khi viết prose, điền bảng
| Câu hỏi | Câu trả lời của paper |
|---|---|
| Research question | |
| Main UCT effect | |
| Economic magnitude | |
| ST vs LT debt finding | |
| Strongest robustness finding | |
| Strongest identification evidence | |
| Main mechanism | |
| Main heterogeneity | |
| Main theoretical implication | |
| Main managerial implication | |
| Main ASEAN policy implication | |
| Main limitation | |
| Best future-research direction |
Rule
Nếu một dòng chưa rõ:
không được viết Conclusion dài.
PHẦN C. STRUCTURE OF CONCLUSION
2. Recommended structure
Paragraph 1 — Research Question + What We Do
Paragraph 2 — Main Findings
Paragraph 3 — Theoretical Implications
Paragraph 4 — Managerial / Financing Implications
Paragraph 5 — ASEAN Policy Implications
Paragraph 6 — Limitations
Paragraph 7 — Future Research + Closing Sentence
Tổng:
6–7 paragraphs
thường đủ.
PHẦN D. PARAGRAPH 1 — RESEARCH QUESTION
3. Không mở Conclusion bằng cách lặp Introduction
Bad:
U.S.–China tension has become increasingly important in recent years.
Better:
This study examines whether and how U.S.–China tension affects the capital-structure decisions of publicly listed firms in ASEAN.
Tiếp theo phải nói ngắn
- sample;
- UCT;
- outcomes;
- key design.
4. Draft mẫu
This study examines whether and how U.S.–China tension affects the capital-structure decisions of publicly listed firms in ASEAN. Using the U.S.–China Tension Index developed by Rogers et al. (2024), we relate bilateral tension to subsequent firm leverage and examine multiple financing margins, including total, short-term, and long-term debt and debt maturity. We complement the baseline specifications with robustness, exposure-based, mechanism, and heterogeneity analyses to assess whether the observed financing responses reflect economically meaningful differences across firms and ASEAN economies.
PHẦN E. PARAGRAPH 2 — MAIN FINDINGS
5. Conclusion should summarize, not repeat coefficients
Need mention:
- direction;
- economic importance;
- debt composition;
- heterogeneity;
- identification strength.
Usually no need to repeat exact coefficient.
PHẦN F. MAIN FINDINGS TEMPLATE — NEGATIVE RESULT
6. Draft
The results indicate that higher U.S.–China tension is associated with lower corporate leverage on average. The reduction is economically meaningful and is concentrated in [short-term/long-term] debt, suggesting that firms adjust [refinancing-sensitive / longer-horizon] financing in response to heightened bilateral tension. The findings remain broadly robust across alternative capital-structure measures, broader uncertainty controls, event-period restrictions, country-sensitivity tests, and alternative inference procedures. Exposure-based analyses further show that the financing response is stronger among [China-exposed / financially constrained / high-short-term-debt] firms, while [U.S.-exposed firms display a weaker or different response].
PHẦN G. MAIN FINDINGS TEMPLATE — MIXED/HETEROGENEOUS RESULT
7. Draft
The evidence does not support a uniform financing response across ASEAN. Instead, the effects of U.S.–China tension depend importantly on firms' and countries' pre-existing exposure to the two major economies. Firms with greater China exposure exhibit [deleveraging / a stronger negative response], whereas firms more exposed to U.S. markets show [a smaller decline / an increase in borrowing]. These patterns indicate that supply-chain risk and trade-diversion opportunities operate simultaneously within ASEAN and that a regional average can conceal economically important heterogeneity.
PHẦN H. MAIN FINDINGS SUMMARY TABLE
8. Bảng bắt buộc
| Finding | Direction | Strength | Main evidence | Keep in Conclusion? |
|---|---|---|---|---|
| Average UCT effect | Strong/Moderate/Suggestive | Yes | ||
| ST debt effect | ||||
| LT debt effect | ||||
| Debt maturity effect | ||||
| China exposure | ||||
| U.S. exposure | ||||
| Constraints | ||||
| Credit supply | ||||
| Trade diversion |
Rule
Conclusion chỉ giữ:
3–5 findings quan trọng nhất.
PHẦN I. THEORETICAL IMPLICATIONS
9. Không viết
This study contributes to trade-off theory.
quá chung.
Phải nói paper bổ sung cái gì cho theory.
PHẦN J. THEORETICAL IMPLICATION 1 — UNCERTAINTY AND LEVERAGE
10. Nếu average effect âm
The findings support the view that uncertainty-related increases in expected distress costs and the value of financial flexibility can reduce firms' reliance on debt.
Link
- trade-off theory;
- financial flexibility;
- Chowdhury et al. (2025).
11. Nếu effect dương
The findings suggest that external-financing needs and trade-diversion opportunities can dominate the risk-reduction channel in selected settings.
Link
- pecking order;
- financing need;
- Vega-Gutiérrez et al. (2025).
PHẦN K. THEORETICAL IMPLICATION 2 — CAPITAL STRUCTURE IS MULTI-DIMENSIONAL
12. Nếu ST/LT khác nhau
Paper adds:
uncertainty affects not only leverage level but also debt composition.
Draft
The distinction between short- and long-term debt shows that capital structure should not be treated as a single leverage ratio when analyzing geopolitical shocks. Bilateral tension can affect refinancing-sensitive debt differently from long-horizon borrowing, making debt maturity an economically important adjustment margin.
PHẦN L. THEORETICAL IMPLICATION 3 — EXPOSURE MATTERS
13. Core ASEAN insight
Same macro shock:
does not imply same firm response.
Response depends on:
Draft
The exposure-based findings also extend the uncertainty literature by showing that aggregate geopolitical shocks are transmitted through firms' underlying economic linkages. The same UCT shock can have different financing consequences depending on firms' exposure to China and the United States.
Link
Jimoh et al. (2026).
PHẦN M. THEORETICAL IMPLICATION 4 — COMPETING CHANNELS
14. Important if results are heterogeneous
Paper can reconcile prior mixed literature:
- Chowdhury: negative leverage response;
- Vega-Gutiérrez: positive response.
Draft
The heterogeneous ASEAN evidence helps reconcile mixed findings in the broader uncertainty–capital structure literature. Risk, distress-cost, and credit-supply channels may dominate for vulnerable firms, while trade-diversion and financing-need channels may dominate for firms positioned to benefit from economic reallocation.
This is stronger than saying:
our results agree with prior studies.
PHẦN N. THEORETICAL IMPLICATIONS TABLE
15. Bảng
| Evidence | Theory/literature implication | Main comparator |
|---|---|---|
| UCT lowers leverage | Risk/flexibility channel | Chowdhury |
| ST debt more sensitive | Rollover channel | Bui/debt maturity literature |
| China exposure strengthens effect | Exposure transmission | Jimoh; Chai-Wang |
| U.S. exposure differs | Trade-diversion channel | Jimoh |
| Institutions moderate | Institutional finance | Vega-Gutiérrez |
| Constraints amplify | Financing frictions | Bui |
PHẦN O. MANAGERIAL IMPLICATIONS
16. Target audience
- CFOs;
- treasurers;
- corporate finance managers;
- risk managers.
Do not write vague advice
Bad:
Firms should pay attention to geopolitical risks.
Better:
Firms with high China supply-chain dependence and short-term debt exposure may need to preserve liquidity and diversify refinancing sources before tension spikes.
PHẦN P. MANAGERIAL IMPLICATION 1 — DEBT MATURITY
17. If ST debt is most sensitive
Possible implication:
- reduce excessive rollover dependence;
- diversify maturities;
- stagger refinancing dates;
- secure committed credit facilities.
Draft
For corporate treasurers, the stronger response in short-term debt suggests that refinancing structure is a key vulnerability during periods of bilateral tension. Firms with concentrated short-term maturities may benefit from staggering refinancing dates, maintaining committed credit lines, and preserving liquidity buffers.
Avoid
telling all firms to reduce short-term debt.
Context matters.
PHẦN Q. MANAGERIAL IMPLICATION 2 — TRADE EXPOSURE
18. Firms should map financing risk to operating exposure
Relevant exposures:
- China suppliers;
- U.S. customers;
- imported intermediate goods;
- export dependence.
Draft
Financing policy should also be coordinated with trade and supply-chain exposure. A firm highly dependent on Chinese inputs may face a different financing risk from one positioned to capture U.S.-oriented trade diversion, even when both operate in the same ASEAN economy.
PHẦN R. MANAGERIAL IMPLICATION 3 — FINANCIAL FLEXIBILITY
19. If constrained firms react more strongly
Implications:
- maintain liquidity;
- diversify funding sources;
- avoid excessive maturity concentration;
- secure financing before periods of elevated tension.
Do not make normative claim stronger than evidence.
PHẦN S. MANAGERIAL IMPLICATIONS TABLE
20. Bảng
| Finding | Managerial risk | Possible response | Evidence strength |
|---|---|---|---|
| ST debt sensitive | Rollover risk | Maturity diversification | |
| China exposure stronger | Supply-chain financing risk | Funding contingency | |
| Constraints amplify | Limited finance access | Liquidity buffer / funding diversification | |
| U.S. exposure positive | Expansion funding need | Capacity financing planning |
PHẦN T. LENDER / INVESTOR IMPLICATIONS
21. Potential users
- banks;
- bond investors;
- credit analysts;
- rating agencies.
Draft
Lenders and credit analysts may also need to incorporate firms' bilateral trade and supply-chain exposure into credit assessment. Country-level geopolitical indicators alone may be insufficient because firms within the same economy can face very different U.S.–China exposure and refinancing risks.
PHẦN U. ASEAN POLICY IMPLICATIONS
22. Target audience
- financial regulators;
- central banks;
- ministries of finance;
- trade/industry agencies;
- ASEAN policymakers.
Policy implication must follow findings.
PHẦN V. POLICY IMPLICATION 1 — CREDIT RESILIENCE
23. If constraints/credit supply matter
Possible implication:
- monitor refinancing stress;
- strengthen credit access;
- avoid abrupt credit contraction;
- support viable firms, not blanket credit expansion.
Draft
Where bilateral tension tightens financing conditions, policymakers may need to monitor refinancing pressure among externally exposed and financially constrained firms. The objective should not be indiscriminate credit support but preservation of financing access for otherwise viable firms facing temporary geopolitical disruptions.
PHẦN W. POLICY IMPLICATION 2 — FINANCIAL DEVELOPMENT
24. If deeper financial systems cushion effect
Policy relevance:
- corporate bond markets;
- diversified banking systems;
- capital markets;
- alternative funding channels.
Draft
The cross-country results also suggest that deeper and more diversified financial systems can influence how external geopolitical shocks are transmitted to corporate financing. Policies that broaden firms' access to bank, bond, and equity financing may therefore improve resilience to external tension shocks.
PHẦN X. POLICY IMPLICATION 3 — TRADE DIVERSIFICATION
25. If exposure matters
ASEAN firms/economies may benefit from:
- diversified trade partners;
- supply-chain diversification;
- regional sourcing;
- FDI opportunities.
Draft
The asymmetric effects associated with U.S. and China exposure also highlight the importance of trade and supply-chain diversification. Policies that reduce excessive dependence on a single market while facilitating firms' participation in regional value chains may limit financing vulnerability while preserving the potential benefits of trade diversion.
PHẦN Y. POLICY IMPLICATION 4 — ASEAN COORDINATION
26. Use carefully
If paper finds strong cross-country heterogeneity:
one-size-fits-all ASEAN response is inappropriate.
Draft
Because the financing consequences of U.S.–China tensions differ across ASEAN economies, regional policy responses should recognize heterogeneity in trade structure, financial depth, and exposure rather than assume a uniform ASEAN shock.
PHẦN Z. POLICY IMPLICATIONS TABLE
27. Bảng
| Empirical finding | Policy concern | Policy implication | Level |
|---|---|---|---|
| Constraints amplify | Financing access | Monitor vulnerable firms | Country |
| ST debt sensitive | Refinancing risk | Financial stability monitoring | Central bank |
| Exposure heterogeneity | Uneven shock | Targeted rather than uniform policy | ASEAN |
| Financial development cushions | Limited alternatives | Deepen capital markets | Country |
| Trade diversion | Expansion needs | Facilitate viable investment finance | Country/ASEAN |
PHẦN AA. LIMITATIONS
28. Limitations phải trung thực nhưng không tự phá paper
Bad:
This study suffers from severe endogeneity and therefore the results may be unreliable.
Better:
The aggregate time-series nature of annual UCT limits the extent to which common global shocks can be absorbed in direct-effect specifications. We mitigate this concern through global controls and exposure-based designs, but stronger firm-level exposure data would permit even tighter identification.
Structure:
Limitation → What we did → What remains → Future solution
PHẦN AB. LIMITATION 1 — UCT TIME-SERIES VARIATION
29. Main limitation of this paper
Annual UCT:
has limited independent time variation.
Consequence
Cannot include:
in main direct-effect model.
Write carefully
First, annual UCT is an aggregate time-series measure, which limits the ability of the direct specification to absorb all common global shocks using year fixed effects. We address this concern through global controls, event-period sensitivity tests, and exposure-based specifications, but future research with more granular firm-level tension exposure could provide stronger identification.
PHẦN AC. LIMITATION 2 — ASEAN DATA COVERAGE
30. If final sample is ASEAN-6
Do not write:
ASEAN firms generally…
without qualification.
Draft
Second, firm-level data availability is uneven across ASEAN. Our final sample covers [countries], meaning that the results are most directly applicable to the listed-firm sectors of these economies rather than to every ASEAN member or private firm.
PHẦN AD. LIMITATION 3 — PUBLICLY LISTED FIRMS
31. Listed firms differ from SMEs/private firms
They are:
- larger;
- more transparent;
- better financed.
This is important in ASEAN
Because SMEs dominate many economies.
Draft
Third, the sample consists of publicly listed firms, which generally have greater access to external finance than private firms and SMEs. The financing effects of geopolitical tension may therefore be stronger among smaller or privately held firms that face tighter credit constraints.
PHẦN AE. LIMITATION 4 — EXPOSURE MEASUREMENT
32. If using country-level exposure
Country exposure is coarse.
Draft
Fourth, country-level trade exposure cannot capture differences in firm-specific customer, supplier, or geographic-sales relationships. More granular supply-chain and segment-level data would allow future studies to distinguish direct firm exposure from the broader national trade environment.
PHẦN AF. LIMITATION 5 — CREDIT SUPPLY VS DEMAND
33. If mechanism uses leverage only
Observed debt reflects:
- financing demand;
- financing supply.
Draft
Finally, changes in observed leverage reflect both firms' demand for debt and lenders' willingness to supply financing. Although our mechanism tests provide evidence consistent with both channels, matched bank–firm or loan-level data would allow a cleaner separation of credit demand from credit supply.
PHẦN AG. LIMITATIONS TABLE
34. Bảng bắt buộc
| Limitation | Why it matters | What paper does about it | What remains |
|---|---|---|---|
| Annual UCT time variation | Omitted global shocks | Controls/exposure | Firm-level exposure |
| ASEAN coverage | External validity | Report ASEAN-X | More countries/data |
| Listed firms | Generalizability | Clear sample | SMEs/private firms |
| Country exposure | Measurement error | Predetermined exposure | Firm-level trade links |
| Supply vs demand | Mechanism ambiguity | Mechanism tests | Bank–firm data |
PHẦN AH. LIMITATIONS TO AVOID
35. Do not write trivial limitations
Avoid:
- “The sample size is limited” if sample is huge;
- “Other variables may matter”;
- “Future studies can use other methods”;
- “This study only uses secondary data”;
- “Time and resources were limited.”
These are not useful scholarly limitations.
PHẦN AI. FUTURE RESEARCH
36. Future research should solve limitations or extend mechanisms
Not random suggestions.
Recommended directions:
- firm-level U.S./China exposure;
- bank–firm lending data;
- private firms/SMEs;
- sectoral heterogeneity;
- monthly/quarterly UCT;
- event-level identification;
- cross-region comparison;
- debt pricing and covenants.
PHẦN AJ. FUTURE RESEARCH 1 — FIRM-LEVEL EXPOSURE
37.
Future research could combine UCT with firm-level customer, supplier, and geographic-segment data to identify how specific trade and supply-chain linkages transmit bilateral tensions into financing decisions.
PHẦN AK. FUTURE RESEARCH 2 — BANK–FIRM LENDING
38.
Matched bank–firm lending data would allow researchers to distinguish changes in firms' credit demand from changes in lenders' credit supply and to examine loan pricing, maturity, collateral, and commitment terms.
This connects well to the user's separate Tourism uncertainty and bank lending project logic, but do not mention that project in the paper.
PHẦN AL. FUTURE RESEARCH 3 — PRIVATE FIRMS AND SMEs
39.
Extending the analysis to private firms and SMEs would be particularly valuable in ASEAN, where smaller firms often rely more heavily on bank financing and may be more vulnerable to geopolitical disruptions.
PHẦN AM. FUTURE RESEARCH 4 — OTHER REGIONS
40.
Compare ASEAN with:
- South Asia;
- Latin America;
- Europe;
- Africa.
Question
Does ASEAN's dual integration with U.S. and China make response unique?
PHẦN AN. FUTURE RESEARCH 5 — OTHER CORPORATE OUTCOMES
41. Related outcomes
- investment;
- cash holdings;
- payout;
- earnings management;
- supply-chain reallocation;
- default risk;
- cost of debt.
But
Do not list ten outcomes in Conclusion.
Choose 1–2 tied to mechanisms.
PHẦN AO. FUTURE RESEARCH TABLE
42. Bảng
| Limitation / open question | Future data | Future method | Potential contribution |
|---|---|---|---|
| Firm exposure | Segment/supply-chain data | Exposure FE | Stronger identification |
| Credit supply | Bank–firm loans | Borrower FE | Separate supply/demand |
| SMEs | Private-firm data | Panel/credit data | External validity |
| Event timing | Monthly UCT | Event study | Dynamic response |
| Regional comparison | Multi-region panel | Exposure design | Generalizability |
PHẦN AP. FINAL CONTRIBUTION LOCK
43. Conclusion must not introduce new contribution
Take contributions from Step 7 and lock them.
Bảng
| Contribution | Claimed in Introduction? | Supported by Results? | Repeated in Conclusion? | Final keep? |
|---|---|---|---|---|
| UCT vs broad uncertainty | ||||
| ASEAN multi-country | ||||
| Debt composition | ||||
| Exposure heterogeneity | ||||
| Mechanism |
Rule
If not supported by results:
remove from Introduction and Conclusion.
PHẦN AQ. FULL CONCLUSION TEMPLATE — NEGATIVE RESULT
44. Draft mẫu
This study examines whether U.S.–China tension affects the capital-structure decisions of publicly listed firms in ASEAN. Using the U.S.–China Tension Index of Rogers et al. (2024), we analyze multiple dimensions of corporate financing, including total, short-term, and long-term leverage and debt maturity. We complement the baseline analysis with robustness, exposure-based, mechanism, and heterogeneity tests to assess whether financing responses differ across firms and ASEAN economies.
The results indicate that higher U.S.–China tension is associated with lower corporate leverage on average. The effect is economically meaningful and is concentrated in [short-term/long-term] debt, suggesting that [refinancing-sensitive borrowing / long-term financing] is an important adjustment margin. The findings remain broadly robust to alternative capital-structure measures, global uncertainty controls, event-period exclusions, country-sensitivity tests, and alternative inference procedures. Exposure-based analyses further show that the effect is stronger among [China-exposed / financially constrained / high-short-term-debt] firms, whereas firms with greater [U.S.] exposure exhibit [a weaker or different response].
These findings contribute to the uncertainty and capital-structure literature in several respects. First, they show that a bilateral geopolitical and trade-related shock can influence corporate financing beyond the effects captured by broad policy or geopolitical uncertainty indices. Second, the distinction between short- and long-term debt indicates that geopolitical tension affects not only the level of leverage but also the composition and maturity of debt. Third, the heterogeneous response across U.S. and China exposure demonstrates that the transmission of aggregate tension depends on firms' underlying economic linkages. This exposure-based perspective helps reconcile previous evidence of both negative and positive uncertainty–leverage relationships.
The results also have implications for corporate financial management. Firms with high exposure to bilateral trade and supply-chain risks may need to integrate geopolitical risk into liquidity and debt-maturity planning. The stronger response among [constrained/high-short-term-debt] firms suggests that financing flexibility and refinancing capacity are especially important when bilateral tensions intensify. Corporate treasurers may therefore benefit from diversified funding sources, staggered debt maturities, and financing contingency plans that reflect firms' underlying trade exposure.
From a policy perspective, the findings suggest that U.S.–China tensions do not constitute a uniform shock across ASEAN. Policymakers should therefore avoid one-size-fits-all responses and pay particular attention to firms and sectors with concentrated trade exposure or refinancing needs. Deeper and more diversified domestic financial systems may help firms absorb external shocks by providing alternative sources of financing, while policies that support supply-chain and market diversification can reduce vulnerability to bilateral tensions without preventing firms from capturing trade-diversion opportunities.
Several limitations should be acknowledged. Annual UCT is an aggregate time-series measure, limiting the extent to which the direct-effect specification can absorb all common global shocks through year fixed effects. We mitigate this concern using global controls, event-period sensitivity tests, and exposure-based specifications, but more granular firm-level exposure measures would permit stronger identification. In addition, firm-level coverage is uneven across ASEAN and the sample is restricted to publicly listed firms, which may be less financially constrained than private firms and SMEs. Finally, changes in leverage reflect both firms' demand for debt and lenders' willingness to supply credit, and the available data do not fully separate these two channels.
Future research could combine UCT with firm-level customer, supplier, and geographic-segment data to identify more precisely how bilateral tensions are transmitted through trade and supply chains. Matched bank–firm lending data would also allow researchers to distinguish debt demand from credit supply and examine loan pricing and maturity responses. Extending the analysis to private firms and SMEs would further clarify whether the financing consequences of U.S.–China tension are stronger among firms with more limited access to external capital. Overall, the evidence highlights corporate financing as an important but heterogeneous channel through which U.S.–China tensions are transmitted to ASEAN economies.
PHẦN AR. FULL CONCLUSION TEMPLATE — HETEROGENEOUS RESULT
45.
This study examines how U.S.–China tension affects corporate financing across ASEAN. Rather than documenting a uniform regional response, the results reveal substantial heterogeneity linked to firms' and countries' pre-existing exposure to the United States and China. Firms with greater China exposure exhibit [a stronger decline in leverage], while firms positioned to benefit from U.S.-oriented trade diversion show [a weaker decline/an increase in borrowing]. These asymmetric effects are particularly visible in [short-term debt / constrained firms / selected sectors].
The findings suggest that geopolitical tension should not be treated as a uniformly negative corporate-finance shock. In ASEAN, the same bilateral event can increase operating and supply-chain risk for some firms while creating expansion and financing opportunities for others. This helps explain why previous studies report both negative and positive uncertainty–leverage relations and highlights the importance of exposure-based rather than purely aggregate measures.
For firms and policymakers, the central implication is that geopolitical-financing risk depends on underlying economic linkages. Corporate financing strategies and policy responses should therefore account for differences in trade dependence, supply-chain structure, refinancing exposure, and access to alternative financing. [Continue with limitations/future research as above.]
PHẦN AS. SHORT CONCLUSION VERSION
46. 250–350 words
Use if journal requires concise conclusion.
Template
This study examines the relationship between U.S.–China tension and corporate capital structure among ASEAN firms. Using the U.S.–China Tension Index of Rogers et al. (2024), we find that [main finding]. The effect is concentrated in [debt component] and is stronger among [main exposure/constraint group]. The results remain [broadly/moderately] robust across alternative leverage measures, global controls, event-period tests, and country-sensitivity analyses. Exposure-based specifications further indicate that [main exposure result].
These findings extend the uncertainty–capital structure literature by showing that bilateral geopolitical tension can affect both the level and composition of debt financing and that the response depends on firms' economic exposure to the United States and China. The ASEAN setting is particularly informative because the region combines supply-chain vulnerability with trade-diversion opportunities.
The results imply that firms should incorporate geopolitical exposure into liquidity, refinancing, and debt-maturity planning, while policymakers should recognize that financing effects vary substantially across countries and sectors. The study is nevertheless subject to limitations associated with the aggregate time-series nature of UCT, uneven ASEAN firm coverage, and the difficulty of separating debt demand from credit supply. Future research using firm-level supply-chain exposure and matched bank–firm lending data could provide stronger identification of these channels.
PHẦN AT. LIMITATION LANGUAGE LADDER
47. Too weak
This study has some limitations.
Better
Several limitations qualify the interpretation of the findings.
Best
The aggregate time-series nature of UCT limits the extent to which common global shocks can be absorbed in direct-effect specifications; although our exposure-based tests mitigate this concern, more granular firm-level exposure data would allow stronger identification.
Be specific.
PHẦN AU. IMPLICATION LANGUAGE CHECK
48. Avoid prescriptive overreach
Bad:
ASEAN governments should provide subsidized loans whenever UCT rises.
Evidence likely does not support this.
Better:
Policymakers may need to monitor refinancing pressure among highly exposed and constrained firms during episodes of elevated bilateral tension.
PHẦN AV. POLICY CLAIM CHECK
49. For every policy implication ask
- Is it directly linked to an empirical finding?
- Does paper identify mechanism strongly enough?
- Is recommendation proportional to evidence?
- Is it relevant to ASEAN?
- Is it implementable at the stated level?
PHẦN AW. LIMITATION–FUTURE RESEARCH PAIRING
50. Every limitation should have a matching future direction
| Limitation | Future research |
|---|---|
| Aggregate UCT | Firm-level/monthly exposure |
| Country exposure | Firm supply-chain data |
| Listed firms | SMEs/private firms |
| Supply vs demand | Bank–firm lending |
| ASEAN-X only | Wider ASEAN/other regions |
This makes the ending constructive.
PHẦN AX. CONCLUSION VS INTRODUCTION CONSISTENCY
51. Check 4 items
Introduction says contribution 1
Conclusion must support it.
Introduction says main finding X
Conclusion must repeat X accurately.
Introduction says mechanism Y
Conclusion must not say mechanism Z unless evidence changed.
Introduction says ASEAN-X
Conclusion must not generalize to all firms worldwide.
PHẦN AY. TITLE–CONCLUSION CONSISTENCY
52. Final title should reflect conclusion
If conclusion says:
exposure heterogeneity is the main contribution
then title should consider:
U.S.–China Tension, Trade Exposure, and Corporate Capital Structure in ASEAN
If conclusion says:
short-term debt is main finding
possible title:
U.S.–China Tension and Corporate Debt Structure in ASEAN
PHẦN AZ. CONCLUSION EVIDENCE MATRIX
53. Bảng bắt buộc
| Conclusion statement | Evidence table | Literature support | Claim level |
|---|---|---|---|
| Main UCT effect | Chowdhury/Bui | Association/stronger | |
| ST debt response | Bui/debt literature | ||
| Exposure heterogeneity | Jimoh | ||
| Constraints | Bui | ||
| Institution effect | Vega | ||
| Supply-chain channel | Chai & Wang |
PHẦN BA. REFERENCES IN CONCLUSION
54. Keep citations limited
Conclusion usually does not need many references.
Use only where needed for:
- comparison;
- theory implication;
- policy context.
Do not re-cite entire literature review.
PHẦN BB. FINAL WRITING CHECKLIST
55.
Conclusion opens with research question, not generic context.
Sample/method described in 1–2 sentences only.
Main finding stated clearly.
Economic meaning stated without repeating full table.
ST/LT/maturity result included if important.
Main exposure result included.
Main mechanism included.
Theoretical implications are specific.
Mixed prior literature is acknowledged where relevant.
Managerial implications follow actual findings.
Lender implications are evidence-based.
ASEAN policy implications are not generic.
No blanket policy recommendation unsupported by data.
Limitations are specific.
Limitations do not invalidate the whole study unnecessarily.
Every limitation has a mitigation already used where applicable.
Every limitation points to future research.
Annual UCT limitation is disclosed.
ASEAN-X coverage limitation is disclosed if relevant.
Listed-firm limitation is disclosed.
Country vs firm exposure limitation is disclosed if relevant.
Credit demand vs supply limitation is disclosed if relevant.
Future research is directly linked to limitations.
Conclusion does not introduce new empirical results.
Conclusion does not introduce new contributions.
Contribution wording matches Introduction.
Title matches final story.
Causal language matches identification.
Full Conclusion is around 900–1.400 words.
Short Conclusion version is available if needed.
PHẦN BC. FINAL DELIVERABLE
Bước 8 hoàn thành khi nhóm nộp:
- Conclusion Storyboard
- Main Finding Summary
- Theoretical Implications Table
- Managerial Implications Table
- Policy Implications Table
- Limitations Table
- Future Research Table
- Final Contribution Lock
- Conclusion Evidence Matrix
- Full Conclusion draft
- Short Conclusion version
PHẦN BD. CHUYỂN SANG BƯỚC 9
Sau Bước 8, toàn bộ nội dung học thuật chính của paper đã có:
- Introduction;
- Literature Review;
- Hypotheses;
- Methodology;
- Results;
- Robustness;
- Identification;
- Mechanisms;
- Heterogeneity;
- Discussion;
- Conclusion;
- Implications;
- Limitations.
Bước tiếp theo:
BƯỚC 9. ABSTRACT, TITLE, KEYWORDS VÀ FINAL STORY CHECK
Ở Bước 9, sinh viên sẽ:
- khóa title;
- viết abstract từ evidence đã có;
- chọn keywords;
- kiểm tra consistency giữa title–abstract–introduction–results–conclusion;
- kiểm tra contribution story lần cuối;
- loại các claims không có evidence.
BƯỚC 9. ABSTRACT, TITLE, KEYWORDS VÀ FINAL STORY CHECK
Bài mẫu: U.S.–China Tension and Corporate Capital Structure in ASEAN Countries
Mục tiêu đầu ra: Khóa toàn bộ research story của paper trước bước hoàn thiện cuối cùng, bao gồm:
- final title;
- abstract;
- keywords;
- main contribution;
- one-sentence paper message;
- consistency giữa Introduction, Results và Conclusion;
- kiểm tra mọi claim, số liệu, citation, table và appendix.
Nguyên tắc của Bước 9
Abstract không phải bản tóm tắt từng section.
Title không phải mô tả tất cả biến trong paper.
Keywords không phải các từ đã xuất hiện trong title.
Bước 9 phải trả lời:
- Paper này thực sự nói điều gì?
- Evidence mạnh nhất là gì?
- Contribution nào còn giữ sau toàn bộ empirical analysis?
- Title có phản ánh đúng contribution đó không?
- Abstract có đúng với final tables không?
- Introduction và Conclusion có nói cùng một câu chuyện không?
PHẦN A. THÀNH PHẨM CUỐI BƯỚC 9
Bước 9 chỉ hoàn thành khi nhóm có:
- Final Storyboard
- Final Title
- Title Alternatives Table
- Final Abstract
- Short Abstract Version
- Keywords
- Contribution Consistency Table
- Claim Consistency Table
- Numeric Consistency Table
- Citation Audit
- Table/Figure/Appendix Audit
- Final Story Check
- Danh sách lỗi phải sửa trước Bước 10
PHẦN B. FINAL STORYBOARD
1. Trước khi viết abstract, phải điền 10 dòng
| Element | Final answer |
|---|---|
| Research question | |
| Sample | |
| Sample period | |
| Main X | |
| Main Y | |
| Main design | |
| Main finding | |
| Economic magnitude | |
| Strongest heterogeneity | |
| Main contribution |
Rule
Abstract chỉ được viết sau khi 10 dòng này thống nhất với final regression tables.
PHẦN C. ONE-SENTENCE PAPER MESSAGE
2. Đây là câu quan trọng nhất của Bước 9
Phải viết được paper trong một câu.
Nếu average effect âm
U.S.–China tensions are associated with lower corporate leverage among ASEAN firms, particularly through short-term debt and among firms with greater China exposure and tighter financing constraints.
Nếu heterogeneous
U.S.–China tensions generate asymmetric financing responses across ASEAN, with deleveraging among China-exposed firms and stronger borrowing among firms positioned to benefit from U.S.-oriented trade diversion.
Nếu average effect dương
ASEAN firms increase leverage during periods of heightened U.S.–China tension, particularly where trade-diversion opportunities raise external financing needs.
Rule
Title, Abstract, Introduction và Conclusion phải xoay quanh cùng một câu.
PHẦN D. TITLE
3. Title phải phản ánh strongest evidence
Không cần đưa:
- every control;
- every mechanism;
- every country;
- every estimator.
Title nên chứa tối đa:
- main X;
- main Y;
- setting;
- strongest differentiator nếu thật sự quan trọng.
PHẦN E. TITLE OPTIONS
4. Neutral / baseline title
U.S.–China Tension and Corporate Capital Structure: Evidence from ASEAN
Dùng khi
Main contribution là:
- direct UCT–capital structure relation;
- ASEAN multi-country evidence.
5. Exposure-focused title
U.S.–China Tension, Trade Exposure, and Corporate Capital Structure in ASEAN
Dùng khi
Strongest evidence là:
và U.S./China exposure là contribution quan trọng.
6. Financing-focused title
U.S.–China Tension and Corporate Financing in ASEAN
Dùng khi
Paper có nhiều outcomes:
- leverage;
- maturity;
- debt composition.
Shorter and broader.
7. Debt-structure title
U.S.–China Tension and Corporate Debt Structure in ASEAN
Dùng khi
Strongest result là:
- ST debt;
- LT debt;
- debt maturity.
8. More thematic title
Between Two Giants: U.S.–China Tension and Corporate Financing in ASEAN
Dùng khi
Journal style chấp nhận title có hook.
Caution
Subtitle/hook should not become journalistic.
PHẦN F. TITLE DECISION TABLE
9. Bảng bắt buộc
| Title option | Matches main finding? | Matches contribution? | Too broad? | Too narrow? | Final score |
|---|---|---|---|---|---|
| U.S.–China Tension and Corporate Capital Structure: Evidence from ASEAN | |||||
| U.S.–China Tension, Trade Exposure, and Corporate Capital Structure in ASEAN | |||||
| U.S.–China Tension and Corporate Financing in ASEAN | |||||
| U.S.–China Tension and Corporate Debt Structure in ASEAN | |||||
| Between Two Giants: U.S.–China Tension and Corporate Financing in ASEAN |
Rule
Không chọn title vì “nghe hay”.
Chọn title phản ánh strongest evidence.
PHẦN G. TITLE WORDING RULES
10. Use “ASEAN” carefully
Nếu sample chỉ là:
title:
Evidence from ASEAN
vẫn có thể dùng nếu manuscript nói rõ sample.
Nhưng avoid:
“Evidence from All ASEAN Countries”
nếu không đủ.
11. Use “Effect” carefully
Title:
The Effect of U.S.–China Tension on Corporate Capital Structure
implies stronger causality.
If identification is not causal:
prefer:
U.S.–China Tension and Corporate Capital Structure
PHẦN H. ABSTRACT STRUCTURE
12. Recommended 6-sentence structure
Sentence 1 — Research Question / Context
Sentence 2 — Data and Sample
Sentence 3 — Method / Identification
Sentence 4 — Main Finding + Economic Magnitude
Sentence 5 — Heterogeneity / Mechanism
Sentence 6 — Contribution / Implication
This is usually enough.
PHẦN I. ABSTRACT — WHAT NOT TO DO
13. Avoid background overload
Bad:
In recent years, geopolitical tensions have become increasingly important…
Abstract has little space.
Start with research question.
14. Avoid literature review
No need:
Prior studies A, B, C find…
Usually no citations in abstract.
15. Avoid methods laundry list
Bad:
We use FE, OLS, 2SLS, GMM, DK, clustering, winsorization…
Better:
We use firm-level panel regressions and exposure-based specifications.
PHẦN J. ABSTRACT — NEGATIVE RESULT TEMPLATE
16. Full template
We examine whether U.S.–China tension affects the capital-structure decisions of publicly listed firms in ASEAN. Using a panel of [N firms / N firm-year observations] from [ASEAN-X countries] over [years] and the U.S.–China Tension Index of Rogers et al. (2024), we analyze total, short-term, and long-term leverage and debt maturity. Our empirical design combines firm-level panel regressions with exposure-based specifications that exploit predetermined differences in firms' or countries' exposure to the United States and China. We find that higher U.S.–China tension is associated with lower corporate leverage, with a one-standard-deviation increase in UCT corresponding to a [X]-percentage-point decline in [main leverage]. The effect is concentrated in [short-term debt] and is stronger among [China-exposed / financially constrained] firms, while [U.S.-exposed firms show a different response]. These findings show that bilateral geopolitical tension affects both the level and composition of corporate debt financing and that its transmission across ASEAN depends critically on firms' underlying economic exposure.
PHẦN K. ABSTRACT — HETEROGENEOUS RESULT TEMPLATE
17.
We examine how U.S.–China tension affects corporate financing across ASEAN. Using a panel of [firms/observations/countries] over [period] and the U.S.–China Tension Index of Rogers et al. (2024), we study multiple dimensions of corporate capital structure. Rather than finding a uniform regional response, we document substantial heterogeneity linked to pre-existing economic exposure. Firms with greater China exposure [reduce leverage], whereas firms with greater U.S. exposure [show a weaker decline/increase in borrowing], consistent with the coexistence of supply-chain risk and trade-diversion channels. The results are robust to alternative capital-structure measures, broader global uncertainty controls, event-period exclusions, and country-sensitivity tests. The findings highlight exposure as a key mechanism through which U.S.–China tensions are transmitted to corporate financing in ASEAN.
PHẦN L. ABSTRACT — POSITIVE RESULT TEMPLATE
18.
We examine whether U.S.–China tension influences the capital-structure decisions of ASEAN firms. Using [sample] and the U.S.–China Tension Index of Rogers et al. (2024), we find that higher bilateral tension is associated with greater corporate leverage. The positive response is concentrated in [debt component] and is strongest among firms with [U.S. exposure / trade-diversion potential]. Exposure-based and robustness analyses indicate that the result is not driven solely by general global uncertainty. The evidence is consistent with the view that trade diversion and strategic supply-chain reallocation can increase external financing needs for selected ASEAN firms. Our findings therefore show that geopolitical tension can create financing opportunities as well as financing risks in third-country economies.
PHẦN M. ECONOMIC MAGNITUDE IN ABSTRACT
19. Include only one magnitude
Best format:
A one-standard-deviation increase in UCT is associated with a X-percentage-point decline in leverage.
Optional:
equivalent to Y% of the sample mean.
Do not include
- five coefficients;
- p-values;
- t-statistics.
PHẦN N. ABSTRACT METHOD LANGUAGE
20. If identification is not fully causal
Use:
We use panel regressions and exposure-based specifications.
Not:
We identify the causal effect…
unless truly justified.
PHẦN O. ABSTRACT LENGTH
21. Prepare 2 versions
Version A — Standard
180–250 words
Version B — Short
120–150 words
Some journals set strict limits.
PHẦN P. SHORT ABSTRACT TEMPLATE
22.
This study examines the relationship between U.S.–China tension and corporate capital structure in ASEAN. Using [sample] and the U.S.–China Tension Index of Rogers et al. (2024), we find that [main finding]. The effect is concentrated in [debt component] and is stronger among [main exposure/constraint group]. Exposure-based analyses indicate that firms' pre-existing links to the United States and China shape the financing response to bilateral tension. The findings show that geopolitical shocks have heterogeneous effects on corporate debt financing across ASEAN and highlight the importance of trade and supply-chain exposure in understanding firms' financing decisions.
PHẦN Q. KEYWORDS
23. Keywords should add discoverability
Recommended pool:
- U.S.–China tension
- Corporate capital structure
- Corporate leverage
- Debt maturity
- ASEAN
- Geopolitical risk
- Trade exposure
- Financial constraints
- Supply chains
- Trade diversion
PHẦN R. KEYWORD SELECTION RULE
24. Choose 4–6
Recommended baseline:
U.S.–China tension; Corporate capital structure; ASEAN; Trade exposure; Debt maturity; Geopolitical risk
If title already contains many of these
Use keywords to add adjacent search terms:
- Corporate leverage
- Financial constraints
- Supply-chain exposure
- Trade diversion
PHẦN S. KEYWORDS TABLE
25.
| Keyword | In title? | Search relevance | Keep? |
|---|---|---|---|
| U.S.–China tension | High | ||
| Corporate capital structure | High | ||
| ASEAN | High | ||
| Corporate leverage | High | ||
| Trade exposure | High | ||
| Debt maturity | Medium/High | ||
| Financial constraints | Medium | ||
| Geopolitical risk | High | ||
| Supply chains | Medium |
PHẦN T. CONTRIBUTION CONSISTENCY CHECK
26. Contributions must be identical across
- Introduction;
- Abstract;
- Conclusion.
Bảng
| Contribution | Introduction | Abstract | Conclusion | Supported by Results? | Keep? |
|---|---|---|---|---|---|
| Bilateral UCT | |||||
| ASEAN multi-country | |||||
| Debt composition | |||||
| Exposure heterogeneity | |||||
| Mechanism |
Rule
If a contribution appears only in Introduction:
either add evidence/discussion elsewhere or remove it.
PHẦN U. CLAIM CONSISTENCY CHECK
27. Major claims
| Claim | Intro | Results | Conclusion | Evidence level | Consistent? |
|---|---|---|---|---|---|
| UCT lowers leverage | |||||
| ST debt more sensitive | |||||
| China exposure amplifies | |||||
| U.S. exposure differs | |||||
| Constraints amplify | |||||
| Credit supply channel |
PHẦN V. CAUSAL LANGUAGE AUDIT
28. Search the manuscript for
- affect
- effect
- impact
- cause
- causal
- lead to
- drive
- determine
For each occurrence
Classify:
| Phrase | Section | Evidence level | Keep/Soften |
|---|---|---|---|
Levels
- correlation
- conditional association
- robust association
- stronger identification
- causal
PHẦN W. NUMERIC CONSISTENCY AUDIT
29. Every final number must match output
Check:
- sample N;
- firms;
- countries;
- years;
- β;
- SE;
- p;
- economic significance;
- SD UCT;
- mean DV;
- interaction effects;
- marginal effects.
Table
| Number/claim | Intro | Abstract | Results | Conclusion | Final table | Match? |
|---|---|---|---|---|---|---|
| N observations | ||||||
| N firms | ||||||
| N countries | ||||||
| β main | ||||||
| 1-SD effect |
PHẦN X. SAMPLE LABEL AUDIT
30. Check every use of “ASEAN”
If actual final sample is:
the paper should clarify this early.
Search for phrases
- ASEAN countries
- ASEAN firms
- all ASEAN
- ASEAN economies
Replace unsupported claims
Bad:
firms in all ASEAN countries
Better:
firms from six ASEAN economies.
PHẦN Y. UCT SOURCE AUDIT
31. Critical
Check every paragraph:
Main UCT source must be:
Rogers, Sun, and Sun (2024)
Do not attribute Rogers UCT to
- Chai & Wang;
- Peng;
- Bui;
- PolicyUncertainty authors generally.
Peng is an application.
Chai & Wang use a different bilateral-relations tension measure.
PHẦN Z. VARIABLE NAME AUDIT
32. Code names must not leak into manuscript without definition
Search:
- TDB
- TDM
- LDB
- LDM
- DM
- MTB
- TUI/UCT inconsistencies.
Manuscript names
Use consistently:
- Book Leverage;
- Market Leverage;
- Short-Term Leverage;
- Long-Term Leverage;
- Debt Maturity.
PHẦN AA. FORMULA AUDIT
33. Check formulas across
- Methodology text;
- variable table;
- code;
- appendix.
Example
If:
it must not become:
somewhere else.
PHẦN AB. TIMING AUDIT
34. Check all descriptions
If model is:
do not write elsewhere:
contemporaneous UCT and leverage.
Table
| Section | Timing stated | Correct? |
|---|---|---|
| Abstract | ||
| Introduction | ||
| Methodology | ||
| Results | ||
| Conclusion |
PHẦN AC. FE AUDIT
35. Annual UCT direct model
Check manuscript never says:
Year Fixed Effects = Yes
if annual UCT is one common value per year and main coefficient is estimated.
Exposure design
Check FE accurately described.
Possible:
- Firm FE;
- Year FE;
- Industry×Year FE;
- Country×Year FE only with firm-level exposure variation.
PHẦN AD. VCE AUDIT
36. Check consistency between
- Methodology;
- table notes;
- code;
- robustness discussion.
Do not call:
“clustered at the country level”
in one place and:
“firm-clustered standard errors”
elsewhere unless models truly differ.
PHẦN AE. WINSORIZATION AUDIT
37. Check
Main rule:
- continuous DVs;
- continuous firm controls;
- 1st–99th percentiles.
Do not winsorize:
- UCT;
- macro variables
unless actual code differs.
PHẦN AF. SAMPLE SCREEN AUDIT
38.
Confirm paper consistently states:
- financial firms excluded;
- utilities excluded;
- AT > 0;
- SALE ≥ 0;
- missing data rule;
- unbalanced panel.
No silent changes across sections.
PHẦN AG. CITATION AUDIT
39. Every citation must be verified
Create:
| Citation | Claim | Original paper checked? | DOI verified? | Correct year? |
|---|---|---|---|---|
| Rogers et al. | UCT | |||
| Peng et al. | UCT financial effects | |||
| Chai & Wang | Corporate strategic response | |||
| Jimoh et al. | ASEAN heterogeneity | |||
| Chowdhury et al. | GPR leverage | |||
| Vega-Gutiérrez et al. | EPU leverage | |||
| Bui et al. | Vietnam UCT leverage | |||
| Wang et al. | Geography/capital structure |
PHẦN AH. DUPLICATE PAPER AUDIT
40. If same paper exists in two PDFs
Do not create two references.
Example:
Chai & Wang duplicate files should produce:
one bibliography entry.
PHẦN AI. REFERENCE YEAR AUDIT
41. Online-first vs volume year
Example:
Wang, Wang & Johnson may have:
- online publication year;
- final volume year.
Use target journal citation policy consistently.
Do not alternate years across manuscript.
PHẦN AJ. TABLE AUDIT
42. For each table check
| Item | Pass? |
|---|---|
| Table number correct | |
| Title informative | |
| DV named clearly | |
| UCT definition clear | |
| Controls reported | |
| FE reported | |
| VCE reported | |
| N reported | |
| Firms reported if needed | |
| Countries reported if needed | |
| Significance notes correct | |
| Variables defined |
PHẦN AK. FIGURE AUDIT
43. For each figure ask
- Does it make a point?
- Is point discussed in text?
- Is source given?
- Are axis labels clear?
- Does figure use same variable definition as tables?
Keep only useful figures.
PHẦN AL. APPENDIX AUDIT
44. Appendix should contain support, not hidden contradictions
Potential appendix:
- detailed country selection;
- alternative UCT transformations;
- VCE variants;
- leave-one-country-out;
- event robustness;
- GMM diagnostics;
- additional heterogeneity;
- variable details.
If a robustness result materially changes conclusion
it belongs in main discussion, not hidden in appendix.
PHẦN AM. TABLE/APPENDIX DECISION MATRIX
45.
| Analysis | Main text | Appendix | Why |
|---|---|---|---|
| Baseline | Yes | No | Core result |
| ST/LT debt | Yes | Maybe | Core mechanism |
| Exposure | Yes | No | Main contribution |
| Global controls | Yes/summary | Full detail | Key concern |
| Leave-one-country | Summary | Full detail | Robustness |
| Alternative VCE | Summary | Full detail | Technical |
| GMM diagnostics | Summary | Full detail | Technical |
| Minor heterogeneity | No | Yes | Secondary |
PHẦN AN. ABSTRACT CLAIM AUDIT
46. Every sentence in abstract must be backed by paper
Example
Abstract says:
“The effect is stronger among financially constrained firms.”
Need:
- interaction/subsample evidence;
- statistically/economically meaningful difference.
Not just:
- one subsample significant;
- another insignificant.
PHẦN AO. TITLE–ABSTRACT CONSISTENCY
47. If title says “Trade Exposure”
Abstract must include:
- exposure design;
- exposure result.
If exposure is only an appendix robustness:
do not put it in title.
PHẦN AP. TITLE–CONTRIBUTION CONSISTENCY
48. If title says “Corporate Capital Structure”
Paper should examine more than:
- one narrow debt ratio
ideally.
If only debt maturity:
use:
Debt Maturity
in title.
PHẦN AQ. KEYWORD–TITLE CONSISTENCY
49. Keywords need not duplicate title exactly
Example title:
U.S.–China Tension and Corporate Capital Structure: Evidence from ASEAN
Possible keywords:
- Geopolitical risk
- Corporate leverage
- Debt maturity
- Trade exposure
- Supply chains
PHẦN AR. JEL CODES — OPTIONAL
50. If journal requests JEL classification
Potential categories to investigate:
- G32 — Financing Policy; Financial Risk and Risk Management; Capital and Ownership Structure
- F50/F51 — International relations / international conflicts
- F14 — Empirical Studies of Trade
- G15 — International Financial Markets
Rule
Verify exact current JEL descriptions before submission.
Do not add random codes.
PHẦN AS. FINAL STORY CHECK
51. Five-part consistency test
Question 1
What is the paper about?
Title answers.
Question 2
What do we find?
Abstract answers.
Question 3
Why is it new?
Introduction answers.
Question 4
What evidence supports it?
Results answer.
Question 5
Why does it matter?
Conclusion answers.
If answers differ:
paper is not yet coherent.
PHẦN AT. STORY CONSISTENCY MATRIX
52.
| Story element | Title | Abstract | Introduction | Results | Conclusion |
|---|---|---|---|---|---|
| Main X | |||||
| Main Y | |||||
| ASEAN | |||||
| Main result | |||||
| Exposure | |||||
| Mechanism | |||||
| Contribution |
PHẦN AU. HYPOTHESIS CONSISTENCY CHECK
53.
| H | Intro/Theory | Tested in Results? | Conclusion status | Consistent? |
|---|---|---|---|---|
| H1 | ||||
| H2a | ||||
| H2b | ||||
| H3 | ||||
| H4 | ||||
| H5 |
If a hypothesis was never tested
Remove or move to future research.
PHẦN AV. CONTRIBUTION STRENGTH TEST
54. Rate each contribution
Level A — Core
Directly supported by main tables.
Level B — Supporting
Supported by additional analyses.
Level C — Aspirational
Sounds good but weak evidence.
Rule
Only Level A/B contributions remain in final manuscript.
Delete Level C.
PHẦN AW. CONTRIBUTION RANKING
55.
| Contribution | Evidence level | Rank |
|---|---|---|
| ASEAN multi-country | ||
| Bilateral UCT | ||
| Debt composition | ||
| U.S./China exposure | ||
| Mechanism | ||
| Institutional heterogeneity |
Final Introduction should emphasize top:
2–4 contributions.
PHẦN AX. FINAL ABSTRACT QUALITY TEST
56. Abstract must contain
Research question
Sample
Time period
Main variable
Main outcome
Empirical design
Main finding
Economic magnitude if useful
Main heterogeneity/mechanism
Contribution
Abstract must not contain
Literature review
Table numbers
p-values
Long methodology details
Unverified causal language
Claims absent from Results
PHẦN AY. ABSTRACT WORD-BY-WORD CHECK
57. Delete words that add little
Common removable phrases:
- “It is worth noting that”
- “Interestingly”
- “Importantly” used repeatedly
- “In today's rapidly changing world”
- “To the best of our knowledge” unless necessary
- “This novel study”
- “This pioneering paper”
Make abstract dense.
PHẦN AZ. TITLE QUALITY TEST
58. A good title should pass all
Main X identifiable
Main Y identifiable
Setting identifiable
No unsupported causal wording
Not too long
No unnecessary subtitle
Matches strongest result
Searchable terms included
Target:
approximately 10–16 words where possible.
PHẦN BA. KEYWORD QUALITY TEST
59.
4–6 keywords
No overly generic words such as “economics”
Includes searchable finance term
Includes geopolitical/trade term
Includes regional term if useful
Adds terms not fully captured by title
PHẦN BB. FINAL ABSTRACT DRAFT WORKSHEET
60.
Sentence 1 — Question
We examine ____________________________.
Sentence 2 — Data
Using _________________________________.
Sentence 3 — Method
Our empirical design ___________________.
Sentence 4 — Main result
We find _______________________________.
Sentence 5 — Mechanism/heterogeneity
The effect is __________________________.
Sentence 6 — Contribution
These findings ________________________.
Fill this worksheet before polishing language.
PHẦN BC. SHORT ABSTRACT WORKSHEET
61. 120–150 words
Must preserve:
- question;
- sample/data;
- main finding;
- main heterogeneity;
- contribution.
Delete:
- robustness detail;
- secondary mechanisms;
- long method description.
PHẦN BD. FINAL TITLE WORKSHEET
62.
Main X
Main Y
Region
Strongest differentiator
Generate:
- neutral title;
- exposure title;
- financing title;
- debt-structure title.
Then score using Title Decision Table.
PHẦN BE. FINAL STORY PARAGRAPH
63. Before finalizing paper, write 150–200 words without citations
Prompt:
What does this paper teach us that we did not know before?
This paragraph is internal
not necessarily manuscript text.
It must include
- main effect;
- ASEAN;
- exposure;
- mechanism;
- contribution.
If this paragraph is confusing:
paper story is still too complicated.
PHẦN BF. SAMPLE FINAL STORY PARAGRAPH
64. Negative-result version
This paper shows that U.S.–China tension is associated with lower corporate leverage in ASEAN, but the response is not uniform. The financing adjustment is concentrated in short-term debt and is stronger among firms with greater exposure to China and tighter financing constraints, indicating that refinancing risk, financial flexibility, and supply-chain vulnerability are important transmission channels. Firms positioned to benefit from U.S.-oriented trade diversion may respond differently, showing that geopolitical tension can create both financing risks and financing opportunities within the same region. By combining a bilateral tension measure with ASEAN exposure heterogeneity and multiple debt-structure outcomes, the study extends both the uncertainty–capital structure literature and the emerging literature on third-country responses to U.S.–China tensions.
PHẦN BG. FINAL REFERENCE CHECK
65. Before Step 10
Check:
- every in-text citation appears in reference list;
- every reference-list item is cited;
- author spelling;
- year;
- journal;
- volume;
- issue;
- pages/article number;
- DOI;
- duplicate references.
Especially check
- Bui et al. 2026;
- Jimoh et al. 2026;
- Chai & Wang 2026;
- Peng et al. 2025;
- Chowdhury et al. 2025;
- Vega-Gutiérrez et al. 2025;
- Rogers et al. 2024;
- Wang et al. publication year.
PHẦN BH. FINAL DATA/METHOD CONSISTENCY CHECK
66.
Same sample period everywhere
Same country list everywhere
Same number of firms everywhere
Same number of observations everywhere
Same variable formulas everywhere
Same timing everywhere
Same FE description everywhere
Same VCE description everywhere
Same winsorization description everywhere
Same UCT source everywhere
PHẦN BI. FINAL RESULTS CONSISTENCY CHECK
67.
Main β matches final table
Main p/significance matches final table
Economic magnitude recomputed from final β
ST/LT result matches final table
Interaction result matches final table
Marginal effects match final output
GMM/IV diagnostics match final output
Robustness claims match all reported tests
No outdated coefficient remains in prose
PHẦN BJ. FINAL CLAIM LANGUAGE CHECK
68.
Replace unsupported:
causes
with:
is associated with
where necessary.
Replace unsupported:
proves
with:
supports / is consistent with.
Replace:
all ASEAN firms
with:
firms in our ASEAN sample
if appropriate.
PHẦN BK. FINAL TABLE ORDER
69. Proposed main paper order
Table 1
Descriptive Statistics
Table 2
Correlation Matrix
Table 3
Baseline Results
Table 4
Alternative Capital Structure / Debt Composition
Table 5
Robustness / Global Controls
Table 6
Exposure-Based Identification
Table 7
Mechanisms
Table 8
Heterogeneity
Appendices
Technical and secondary analyses.
PHẦN BL. FINAL FIGURE ORDER
70. Possible
Figure 1
UCT over time.
Figure 2
Marginal effect by China/U.S. exposure.
Figure 3
Event study if credible.
Figure 4
Leave-one-country-out if useful.
Do not include all if journal space is limited.
PHẦN BM. ABSTRACT/TITLE EVIDENCE FILE
71. File đề xuất
09_FINAL_STORY_EVIDENCE.xlsx
Sheet 1 — STORYBOARD
| Element | Final content | Evidence |
|---|
Sheet 2 — TITLES
| Title | Score | Keep? |
|---|
Sheet 3 — ABSTRACT
| Sentence | Purpose | Evidence |
|---|
Sheet 4 — KEYWORDS
| Keyword | Relevance | Keep? |
|---|
Sheet 5 — CONTRIBUTIONS
| Contribution | Intro | Results | Conclusion | Keep? |
|---|
Sheet 6 — NUMERIC_CHECK
| Number | Source | Intro | Abstract | Results | Conclusion |
|---|
Sheet 7 — CITATION_CHECK
| Citation | Claim | Verified | DOI |
|---|
Sheet 8 — TABLE_AUDIT
| Table | Main claim | Main/Appendix | Final? |
|---|
PHẦN BN. FINAL ABSTRACT EXAMPLE — PLACEHOLDER VERSION
72.
We examine whether U.S.–China tension affects corporate capital structure among publicly listed firms in ASEAN. Using [N] firm-year observations from [ASEAN-X countries] over [period] and the U.S.–China Tension Index of Rogers et al. (2024), we analyze total, short-term, and long-term leverage and debt maturity. Our empirical design combines firm fixed-effects models with exposure-based specifications using predetermined U.S. and China exposure. We find that higher U.S.–China tension is associated with [main result], with a one-standard-deviation increase in UCT corresponding to [economic magnitude]. The response is strongest for [main debt component/group] and varies systematically with firms' exposure to [China/U.S.] and their financing constraints. These findings show that bilateral geopolitical tension has heterogeneous financing consequences across ASEAN and highlight trade and supply-chain exposure as important channels linking global political tensions to corporate financial policy.
Important
Every placeholder must be replaced from final output.
PHẦN BO. FINAL TITLE RECOMMENDATION RULE
73. Choose based on actual evidence
If average effect strongest
U.S.–China Tension and Corporate Capital Structure: Evidence from ASEAN
If exposure is strongest
U.S.–China Tension, Trade Exposure, and Corporate Capital Structure in ASEAN
If debt maturity is strongest
U.S.–China Tension and Corporate Debt Structure in ASEAN
If result is broad/mixed
U.S.–China Tension and Corporate Financing in ASEAN
PHẦN BP. FINAL KEYWORDS RECOMMENDATION
74. Baseline set
U.S.–China tension; Corporate leverage; ASEAN; Trade exposure; Debt maturity; Geopolitical risk
Exposure-focused alternative
U.S.–China tension; Corporate capital structure; ASEAN; Trade exposure; Supply chains; Financial constraints
PHẦN BQ. FINAL CHECKLIST
75.
Final Storyboard completed.
One-sentence paper message finalized.
Title selected based on strongest evidence.
Title avoids unsupported causal language.
Title matches ASEAN sample.
Standard abstract completed.
Short abstract completed.
Abstract contains sample.
Abstract contains period.
Abstract contains UCT source/measure appropriately.
Abstract contains main result.
Abstract contains one economic magnitude if useful.
Abstract contains strongest heterogeneity.
Abstract contains contribution.
Abstract does not contain citations unless journal requires.
Abstract does not contain p-values.
Keywords selected.
Contribution Consistency Table completed.
Claim Consistency Table completed.
Numeric Consistency Table completed.
Sample label audited.
UCT source audited.
Variable names audited.
Formulas audited.
Timing audited.
FE audited.
VCE audited.
Winsorization audited.
Sample screens audited.
Citations audited.
Duplicate references removed.
Publication years checked.
All tables audited.
All figures audited.
Appendix audited.
Main claims not hidden in appendix.
Hypotheses all tested or removed.
Contributions ranked and weak claims removed.
Final story paragraph completed.
Abstract/title/results/conclusion tell the same story.
No outdated regression number remains.
No unsupported causal language remains.
PHẦN BR. FINAL DELIVERABLE
Bước 9 hoàn thành khi nhóm nộp:
- Final Storyboard
- Final Title
- Alternative Title Table
- Final Abstract 180–250 words
- Short Abstract 120–150 words
- 4–6 Keywords
- Contribution Consistency Table
- Claim Consistency Table
- Numeric Consistency Table
- Citation Audit
- Table/Figure/Appendix Audit
- Final Story Paragraph
- Danh sách các lỗi phải sửa trước Bước 10
PHẦN BS. CHUYỂN SANG BƯỚC 10
Sau Bước 9, paper đã khóa:
- research story;
- title;
- abstract;
- keywords;
- main contributions;
- claims;
- numbers;
- citations;
- table structure.
Bước cuối cùng:
BƯỚC 10. FINAL PAPER ASSEMBLY, FORMATTING, REFERENCES VÀ SUBMISSION CHECK
Bước 10 sẽ hướng dẫn sinh viên:
- ghép toàn bộ sections;
- chuẩn hóa heading;
- kiểm tra numbering;
- chuẩn hóa equations;
- chuẩn hóa tables/figures;
- chuẩn hóa references;
- plagiarism/self-plagiarism check;
- AI-use check;
- journal formatting;
- cover letter;
- final submission package.
BƯỚC 10. FINAL PAPER ASSEMBLY, FORMATTING, REFERENCES VÀ SUBMISSION CHECK
Bài mẫu: U.S.–China Tension and Corporate Capital Structure in ASEAN Countries
Mục tiêu đầu ra: Ghép toàn bộ kết quả từ Bước 1–9 thành một submission-ready manuscript có cấu trúc thống nhất, citations đầy đủ, tables/figures khớp nội dung, không còn lỗi số liệu, lỗi định dạng, lỗi phương pháp, hoặc claim vượt quá evidence.
Nguyên tắc của Bước 10
Đây không phải bước “chỉnh font”.
Đây là bước kiểm tra xem paper có thực sự khép kín về nội dung học thuật hay chưa.
Một paper chỉ được coi là final khi:
- mọi claim có evidence;
- mọi số liệu khớp final output;
- mọi citation tồn tại và đúng nguồn;
- title–abstract–introduction–results–conclusion nói cùng một câu chuyện;
- method description khớp code;
- tables/figures có thể đọc độc lập;
- causal language không mạnh hơn identification;
- final manuscript đáp ứng format của target journal.
PHẦN A. THÀNH PHẨM CUỐI BƯỚC 10
Bước 10 chỉ hoàn thành khi nhóm có đủ:
- Final Manuscript
- Final Tables
- Final Figures
- Online Appendix / Supplementary Appendix
- Final Reference List
- Citation Audit
- Numeric Audit
- Method–Code Consistency Audit
- AI/Integrity Check
- Journal Formatting Check
- Cover Letter Draft
- Submission Package Checklist
- Final version archive
PHẦN B. GHÉP CẤU TRÚC PAPER
1. Recommended manuscript order
Title
Abstract
Keywords
1. Introduction
2. Literature Review and Hypothesis Development
3. Data and Methodology
3.1. Sample and Data Sources
3.2. U.S.–China Tension Measure
3.3. Capital-Structure Measures
3.4. Control Variables
3.5. Empirical Specifications
4. Empirical Results
4.1. Descriptive Statistics
4.2. Baseline Results
4.3. Economic Significance and Debt Composition
4.4. Robustness Tests
5. Identification and Additional Analyses
5.1. Exposure-Based Identification
5.2. Placebo / Dynamic Tests
5.3. Mechanisms
5.4. Heterogeneity
6. Conclusion
References
Appendix / Online Supplement
Adjust section structure to journal requirements.
PHẦN C. SECTION-BY-SECTION ASSEMBLY
2. Introduction
Take from Step 7.
Before inserting:
- remove placeholders;
- replace sample numbers;
- replace economic magnitude;
- ensure contribution matches final Results;
- ensure title matches final story.
3. Literature Review and Hypotheses
Take from Step 5.
Check:
- no paper-by-paper listing;
- opposing evidence retained;
- hypotheses correspond to tests actually run;
- exploratory tests are not presented as ex ante hypotheses.
4. Data and Methodology
Take from Step 1.
Update using final:
- sample years;
- ASEAN countries;
- N firms;
- N observations;
- variable formulas;
- FE;
- VCE;
- winsorization;
- timing;
- final exposure variables.
Critical
Methodology must describe the final code, not an earlier planned design.
5. Results and Discussion
Take from Step 6.
Update:
- all table numbers;
- all column numbers;
- all coefficients;
- all p-values/significance;
- all economic significance calculations.
Delete prose linked to regressions no longer in final paper.
6. Conclusion
Take from Step 8.
Ensure:
- no new results;
- no new contribution;
- limitations match final design;
- future research follows real limitations.
PHẦN D. FINAL MANUSCRIPT MASTER MAP
7. Bảng bắt buộc
| Section | Source step | Final file/content | Status |
|---|---|---|---|
| Abstract | Step 9 | ||
| Introduction | Step 7 | ||
| Literature Review | Step 5 | ||
| Hypotheses | Step 5 | ||
| Data | Step 1 | ||
| Variables | Step 1 | ||
| Models | Step 1 / 4 | ||
| Baseline Results | Step 2 / 6 | ||
| Robustness | Step 3 / 6 | ||
| Identification | Step 4 / 6 | ||
| Mechanisms | Step 4 / 6 | ||
| Heterogeneity | Step 4 / 6 | ||
| Conclusion | Step 8 | ||
| References | Steps 1–9 | ||
| Appendix | Steps 3–4 |
PHẦN E. PLACEHOLDER SEARCH
8. Search entire manuscript for
[ ]XXXTBDINSERTTable XColumn X[β][N][period][country][X]%[Y]%ASEAN-X
Rule
Final submission must contain:
zero unresolved placeholders.
PHẦN F. VARIABLE NAME CHECK
9. Search code-like names
- TDB
- TDM
- LDB
- LDM
- STLEV
- DM
- UCT_Z
- LN_UCT
- SA_INDEX
- WW_INDEX
Rule
Code names may appear in:
- variable definition table;
- appendix;
- code documentation.
Main prose should use:
- Total Book Leverage;
- Total Market Leverage;
- Short-Term Leverage;
- Long-Term Leverage;
- Debt Maturity.
PHẦN G. FINAL VARIABLE DICTIONARY
10. Bảng bắt buộc
| Manuscript variable | Code variable | Formula | Source | Winsorized? |
|---|---|---|---|---|
| UCT | Rogers et al. | No | ||
| Book Leverage | Compustat | Yes | ||
| ST Leverage | Compustat | Yes | ||
| LT Leverage | Compustat | Yes | ||
| Debt Maturity | Compustat | Yes | ||
| Tangibility | Compustat | Yes | ||
| Profitability | Compustat | Yes | ||
| Size | Compustat | Yes | ||
| Market-to-Book | Compustat | Yes | ||
| GDP Growth | WDI/source | No | ||
| Inflation | WDI/source | No |
Purpose
This table is internal master truth.
Every section must match it.
PHẦN H. FINAL SAMPLE AUDIT
11. Lock final sample
Record:
| Item | Final value |
|---|---|
| Initial observations | |
| Final observations | |
| Firms | |
| Countries | |
| Industries | |
| First year | |
| Last year | |
| Financials excluded? | |
| Utilities excluded? | |
| AT > 0? | |
| SALE ≥ 0? | |
| Missing-data rule | |
| Unbalanced/balanced |
12. Country list
Write final countries explicitly.
If:
- Indonesia
- Malaysia
- Philippines
- Singapore
- Thailand
- Vietnam
then manuscript should state:
six ASEAN economies or ASEAN-6 sample where appropriate.
Do not imply all ASEAN countries.
PHẦN I. METHOD–CODE CONSISTENCY AUDIT
13. This is mandatory
For each methodological statement:
compare against actual code.
Audit items
| Method statement | Manuscript says | Code does | Match? |
|---|---|---|---|
| Financial exclusion | |||
| Utility exclusion | |||
| AT filter | |||
| SALE filter | |||
| Winsorization | |||
| Lead DV | |||
| UCT timing | |||
| Firm FE | |||
| Year FE | |||
| Industry×Year FE | |||
| Country×Year FE | |||
| Cluster level |
Critical
If code and manuscript disagree:
code/output is the factual source, unless code itself is being corrected.
PHẦN J. FE CONSISTENCY
14. Annual UCT direct-effect model
If:
is common to every firm in year $t$:
do not report:
while simultaneously estimating main UCT.
15. Country exposure model
If:
is country-year:
do not use:
and expect to estimate that interaction.
16. Firm exposure model
If:
varies across firms:
Country×Year FE can be used if within-country-year firm variation remains.
PHẦN K. TABLE CONSTRUCTION
17. Main table principles
Each regression table must be understandable without reading entire paper.
Table note should include:
- dependent variable;
- main independent variable;
- control set;
- fixed effects;
- VCE;
- winsorization;
- significance notation;
- sample restriction.
PHẦN L. FINAL MAIN TABLE ORDER
18. Recommended
Table 1
Descriptive Statistics
Table 2
Correlation Matrix
Table 3
U.S.–China Tension and Corporate Leverage: Baseline Results
Table 4
U.S.–China Tension and Debt Composition
Table 5
Robustness to Alternative Specifications and Global Controls
Table 6
Exposure-Based Identification
Table 7
Mechanism Tests
Table 8
Heterogeneity Analyses
PHẦN M. TABLE TITLE RULE
19. Bad
Regression Results
Better
U.S.–China Tension and Corporate Leverage
Even better
U.S.–China Tension and Corporate Leverage: Baseline Estimates
Table title should tell reader what question is being answered.
PHẦN N. TABLE COLUMN RULE
20. Column headings
Avoid only:
- (1)
- (2)
- (3)
Add meaningful DV labels where possible:
| (1) Book Lev | (2) ST Lev | (3) LT Lev | (4) Debt Maturity |
|---|
PHẦN O. TABLE NOTE TEMPLATE
21. Draft
Notes: This table reports estimates of the relationship between U.S.–China tension and corporate capital structure. UCT denotes the U.S.–China Tension Index of Rogers et al. (2024). All specifications include the firm-level controls described in Section 3. Continuous firm-level variables are winsorized at the 1st and 99th percentiles. Standard errors are [VCE] and are reported in parentheses. ***, **, and * denote statistical significance at the 1%, 5%, and 10% levels, respectively.
Adjust:
- FE;
- macro controls;
- sample;
- timing.
PHẦN P. DO NOT OVERLOAD TABLES
22. If control list is long
Possible display:
- main X;
- key interaction;
- control rows optional;
- “Firm Controls: Yes”
- “Macro Controls: Yes”
But if journal expects coefficients:
show them.
Follow target journal conventions.
PHẦN Q. FIGURES
23. Recommended only if useful
Potential:
Figure 1
UCT over time.
Figure 2
Marginal effects by China exposure.
Figure 3
Marginal effects by U.S. exposure.
Figure 4
Event-time plot.
Figure 5
Leave-one-country-out coefficient plot.
PHẦN R. FIGURE QUALITY CHECK
24.
Axis labels include units.
Legend readable.
Confidence intervals labeled.
Caption explains variable.
Source given.
Same UCT definition as paper.
No decorative graph without analytical value.
PHẦN S. EQUATION FORMATTING
25. Use consistent notation
Baseline:
Define
- $i$ = firm;
- $c$ = country;
- $t$ = year;
- $\alpha_i$ = firm fixed effects.
26. Exposure equation
where applicable.
Every symbol must be defined.
PHẦN T. EQUATION AUDIT
27.
| Equation | Variables match code? | FE match code? | Timing match code? | Keep? |
|---|---|---|---|---|
| Baseline | ||||
| Country exposure | ||||
| Firm exposure | ||||
| Dynamic | ||||
| IV |
Delete equations for analyses no longer used.
PHẦN U. REFERENCES
28. Reference list must be generated from citations, not memory
Every in-text citation must appear in References.
Every References entry should appear in text unless journal permits uncited background references.
PHẦN V. CORE REFERENCES TO VERIFY CAREFULLY
29.
Rogers et al. (2024)
UCT source.
Peng et al. (2025)
UCT financial-market effects.
Chai & Wang (2026)
Corporate strategic response.
Jimoh et al. (2026)
ASEAN trade-tension heterogeneity.
Chowdhury et al. (2025)
GPR and capital structure.
Vega-Gutiérrez et al. (2025)
EPU and capital structure.
Bui et al. (2026)
Vietnam UCT and leverage.
Wang et al.
Geography and capital structure.
PHẦN W. REFERENCE MASTER TABLE
30.
| Citation key | Authors | Year | Title | Journal | Volume | Issue | Pages/Article | DOI | Verified |
|---|---|---|---|---|---|---|---|---|---|
Rule
No manually guessed DOI.
Verify every DOI from:
- publisher;
- Crossref;
- journal page;
- original PDF.
PHẦN X. REFERENCE STYLE
31. Do not fully format references before target journal is chosen
First keep clean bibliographic metadata.
Then transform to:
- APA;
- Harvard;
- Elsevier;
- Chicago;
- journal-specific style.
Why?
Avoid retyping reference list repeatedly.
PHẦN Y. REFERENCE DUPLICATES
32. Check duplicates caused by
- online-first vs published version;
- abbreviated vs full author names;
- duplicate PDFs;
- DOI vs no DOI;
- different publication years.
Example
Chai & Wang duplicate PDFs:
one reference only.
PHẦN Z. CITATION CLAIM AUDIT
33. For every important citation
Check:
Does the cited paper actually support the sentence?
Bảng
| Sentence/claim | Citation | Exact source section/page | Correct? |
|---|---|---|---|
| UCT is news-based | Rogers | ||
| ASEAN response heterogeneous | Jimoh | ||
| GPR reduces leverage | Chowdhury | ||
| EPU increases leverage | Vega | ||
| Tension affects strategic allocation | Chai-Wang |
PHẦN AA. DO NOT CITE ABSTRACT-ONLY IF CLAIM IS DETAILED
34. If claim concerns
- exact variable formula;
- exact FE;
- exact IV;
- exact GMM setup;
- exact mechanism;
read:
- methodology;
- table;
- appendix;
not only abstract.
PHẦN AB. BIBLIOGRAPHY COMPLETENESS
35. Run two-direction check
Direction A
In-text citation → References.
Direction B
References → in-text citation.
Both must match.
PHẦN AC. NUMERIC AUDIT
36. Build one master numeric table
| Statistic | Final value | Source table/output | Appears in Intro? | Abstract? | Results? | Conclusion? |
|---|---|---|---|---|---|---|
| N observations | ||||||
| N firms | ||||||
| N countries | ||||||
| Sample start | ||||||
| Sample end | ||||||
| Main β | ||||||
| Main p | ||||||
| SD UCT | ||||||
| Mean DV | ||||||
| 1-SD effect | ||||||
| Interaction β |
PHẦN AD. ECONOMIC SIGNIFICANCE RECHECK
37. Recalculate from final model
Do not reuse earlier calculation if coefficient changed.
Formula:
Relative effect:
PHẦN AE. PERCENTAGE-POINT CHECK
38. Example
If leverage changes from:
to:
then:
- change = −1.5 percentage points
- relative decline = 5%
Do not write:
leverage falls 1.5%.
unless clearly referring to relative change.
PHẦN AF. SIGNIFICANCE SYMBOL CHECK
39. Confirm consistent convention
Usually:
***p < 0.01**p < 0.05*p < 0.10
If target journal uses otherwise:
follow journal.
PHẦN AG. TABLE N / SAMPLE CHECK
40. Different DVs can have different N
Especially:
Debt Maturity:
requires:
Do not force identical N.
Explain in table notes.
PHẦN AH. APPENDIX STRUCTURE
41. Suggested Online Appendix
Appendix A
Variable Definitions.
Appendix B
Country Sample Composition.
Appendix C
Alternative UCT Measures.
Appendix D
Alternative VCE.
Appendix E
Leave-One-Country-Out.
Appendix F
Additional Exposure Results.
Appendix G
Dynamic GMM Diagnostics.
Appendix H
Additional Heterogeneity.
PHẦN AI. APPENDIX TABLE NUMBERING
42. Use
- Table A1
- Table A2
- Figure A1
- Figure A2
Not:
- Table 9;
- Table 10
if journal uses separate appendix numbering.
PHẦN AJ. MAIN TEXT VS APPENDIX
43. Main text contains results necessary to believe contribution
Appendix contains:
- secondary checks;
- technical details;
- alternative definitions.
Rule
Do not hide:
- sign reversal;
- failed placebo;
- major sensitivity
only in appendix.
Discuss material issues in main text.
PHẦN AK. PLAGIARISM / TEXT ORIGINALITY
44. Do not copy wording from papers
Even if cited.
Allowed:
- paraphrase;
- short quotations if necessary;
- standard methodological terminology.
Especially avoid copying
- abstracts;
- introductions;
- hypothesis paragraphs;
- table notes.
PHẦN AL. SELF-PLAGIARISM
45. If authors reuse their own prior wording
Check journal rules.
Do not paste entire paragraphs from:
- thesis;
- prior paper;
- working paper
without appropriate handling.
PHẦN AM. AI USE CHECK
46. AI can assist with
- language editing;
- structure;
- consistency;
- checking logic;
- code explanation;
- drafting from verified evidence.
AI must not become source of facts
Every:
- coefficient;
- citation;
- DOI;
- sample count;
- variable formula;
- method claim
must be verified against:
- data;
- code;
- original source.
PHẦN AN. AI DISCLOSURE
47. Check target journal policy
Some journals require disclosure of generative AI use.
If required:
state accurately what AI was used for.
Examples:
- language editing;
- organization;
- coding assistance.
Never list AI as an author.
PHẦN AO. REPRODUCIBILITY PACKAGE
48. Internally prepare
Recommended:
PROJECT
├── DATA
├── CODE
│ ├── 01_BUILD
│ ├── 02_VARIABLES
│ ├── 03_ANALYSIS
│ └── 04_TABLES
├── OUTPUT
│ ├── TABLES
│ └── FIGURES
├── MANUSCRIPT
└── README
If data cannot be shared
README should explain:
- proprietary source;
- variable construction;
- code sequence.
PHẦN AP. REPRODUCIBILITY CHECK
49. A new researcher should know
- where raw data come from;
- how sample is built;
- how variables are constructed;
- which script runs first;
- how final tables are generated.
Ideal
One master script reproduces final outputs.
PHẦN AQ. FINAL CODE FREEZE
50. Once final manuscript numbers are locked
Create:
FINAL_CODE
and stop modifying code silently.
Any later code change requires:
- rerun;
- updated output;
- updated manuscript;
- updated numeric audit.
PHẦN AR. FILE VERSIONING
51. Recommended
MANUSCRIPT_v01.docx
MANUSCRIPT_v02.docx
MANUSCRIPT_FINAL_REVIEW.docx
MANUSCRIPT_SUBMISSION.docx
Do not create files named:
- final2;
- finalfinal;
- final_ok;
- final_new2.
Use date/version if needed.
PHẦN AS. TARGET JOURNAL CHECK
52. Before final formatting
Record:
| Requirement | Journal rule | Manuscript status |
|---|---|---|
| Word limit | ||
| Abstract limit | ||
| Keywords | ||
| Reference style | ||
| Tables embedded/separate | ||
| Figures format | ||
| Anonymous manuscript | ||
| Title page separate | ||
| Data availability statement | ||
| AI disclosure | ||
| Conflict of interest | ||
| Funding statement | ||
| Highlights required | ||
| Graphical abstract |
PHẦN AT. DO NOT FORMAT FOR A JOURNAL TOO EARLY
53. First lock scientific content
Then journal formatting.
Otherwise:
time is wasted repeatedly reformatting a changing paper.
PHẦN AU. TITLE PAGE
54. Typical title-page items
- title;
- authors;
- affiliations;
- corresponding author;
- email;
- ORCID if required;
- acknowledgments;
- funding;
- declarations.
Anonymous review
Remove identifying information from blinded manuscript if required.
PHẦN AV. AUTHOR ORDER
55. Must be agreed before submission
Do not decide after acceptance.
If contribution statements are required:
prepare CRediT roles.
Potential roles:
- Conceptualization;
- Methodology;
- Data curation;
- Formal analysis;
- Writing – original draft;
- Writing – review & editing;
- Supervision.
PHẦN AW. DATA AVAILABILITY STATEMENT
56. If using proprietary Compustat
Possible language:
The firm-level financial data used in this study are obtained from Compustat Global under license and cannot be redistributed. The U.S.–China Tension Index is publicly available from its original source. Replication code can be made available subject to journal and data-license requirements.
Adapt to actual sharing policy.
PHẦN AX. CONFLICT OF INTEREST
57. Prepare statement
If none:
The authors declare no conflict of interest.
Follow journal wording.
PHẦN AY. FUNDING
58. If funding exists
Include exact:
- funder name;
- grant number.
If none and journal asks:
This research received no specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Use only if true.
PHẦN AZ. COVER LETTER
59. Cover letter should be short
Recommended 4 parts:
- paper title;
- research question;
- main contribution;
- journal fit.
Do not paste Abstract.
PHẦN BA. COVER LETTER TEMPLATE
60.
Dear Editor,
Please consider our manuscript, “[TITLE],” for publication in [JOURNAL].
The study examines how U.S.–China tension is associated with corporate capital structure across ASEAN firms. Using [brief sample/design], we show that [one-sentence main finding]. The paper contributes to the literature by [2–3 concise contributions].
We believe the manuscript is well suited to [JOURNAL] because it addresses [journal-relevant theme] and provides evidence on the interaction between geopolitical tension and corporate financial policy in an internationally important regional setting.
The manuscript is original, is not under consideration elsewhere, and all authors have approved the submission. [Add conflict/data/funding declarations if journal requires.]
Sincerely,
[Corresponding Author]
PHẦN BB. COVER LETTER CHECK
61.
Correct journal name
Correct editor name if used
Correct title
No old journal name from prior submission
No exaggerated claims
Contribution concise
Originality statement accurate
PHẦN BC. HIGHLIGHTS
62. If journal requires 3–5 highlights
Each:
approximately one short sentence.
Example placeholders:
- U.S.–China tension is associated with lower leverage among ASEAN firms.
- The financing response is concentrated in short-term debt.
- China exposure amplifies the negative leverage response.
- Trade exposure generates heterogeneous financing effects across ASEAN.
- Financial constraints strengthen firms' responses to bilateral tension.
Only use findings supported by final results.
PHẦN BD. GRAPHICAL ABSTRACT
63. If required
Simple structure:
with moderators:
Do not overcrowd.
PHẦN BE. JOURNAL FIT PARAGRAPH
64. Internal exercise
Write 100 words:
Why should this journal's readers care?
If impossible:
journal fit may be weak.
PHẦN BF. FINAL LANGUAGE EDIT
65. Editing order
Pass 1
Logic.
Pass 2
Redundancy.
Pass 3
Academic English.
Pass 4
Numbers.
Pass 5
Citations.
Pass 6
Formatting.
Do not start with grammar before story is final.
PHẦN BG. REDUNDANCY SEARCH
66. Search repeated phrases
Common repeated claims:
- “ASEAN is deeply integrated with U.S. and China”
- “UCT increases uncertainty”
- “financial flexibility”
- “trade diversion”
- “supply-chain disruption”
Each key point should have one primary home.
PHẦN BH. INTRODUCTION VS LITERATURE REVIEW
67.
Introduction:
compressed story.
Literature Review:
detailed theory/evidence.
If same paragraph appears twice:
rewrite.
PHẦN BI. RESULTS VS CONCLUSION
68.
Results:
evidence detail.
Conclusion:
high-level meaning.
Do not copy Results paragraph into Conclusion.
PHẦN BJ. ABSTRACT VS INTRODUCTION
69.
Abstract:
standalone summary.
Introduction:
motivation, gap, findings, contributions.
Do not copy first paragraph of Introduction into Abstract.
PHẦN BK. FINAL CLAIM AUDIT
70. Search these verbs
- causes
- affects
- impacts
- drives
- determines
- leads to
- increases
- reduces
For each:
ask whether evidence permits causal wording.
PHẦN BL. FINAL LIMITATION CHECK
71. Ensure disclosed
If relevant:
- annual UCT time-series variation;
- sample coverage;
- listed firms;
- exposure measurement;
- credit demand vs supply.
Do not add new limitation at submission stage unless necessary.
PHẦN BM. HYPOTHESIS CHECK
72.
Every hypothesis should have:
- theory;
- empirical test;
- result;
- conclusion.
Delete
hypotheses never tested.
Reclassify
post-hoc exploratory analyses where necessary.
PHẦN BN. MAIN CONTRIBUTION LOCK
73. Final paper should have only 2–4 main contributions
Possible final set:
- bilateral UCT vs broad uncertainty;
- ASEAN multi-country evidence;
- debt-composition/maturity evidence;
- U.S./China exposure heterogeneity.
Mechanisms may support contributions rather than be separate contribution.
PHẦN BO. FINAL STORY MATRIX
74.
| Element | Final sentence |
|---|---|
| Research question | |
| Why ASEAN | |
| Main finding | |
| Economic magnitude | |
| Main mechanism | |
| Main heterogeneity | |
| Identification strength | |
| Contribution | |
| Limitation | |
| Policy/managerial implication |
PHẦN BP. SUBMISSION PACKAGE
75. Typical package
File 1
Manuscript_Blinded.docx
File 2
Title_Page.docx
File 3
Tables.docx/xlsx if separate.
File 4
Figures.zip if required.
File 5
Online_Appendix.pdf/docx
File 6
Cover_Letter.docx
File 7
Highlights.docx if required.
File 8
Data_Availability_Statement
if separate.
File 9
Replication_Code.zip
if requested/permitted.
PHẦN BQ. FINAL SUBMISSION CHECKLIST
76. Scientific content
Research question is clear.
Main contribution is clear.
Main findings support contribution.
No unsupported novelty claim.
Closest literature cited.
Opposing literature discussed.
ASEAN setting accurately described.
Sample label accurate.
UCT source accurate.
Annual UCT identification issue addressed.
Exposure identification described correctly.
Mechanisms not overstated.
Causal language appropriate.
77. Data and methods
Sample period correct.
Countries correct.
N firms correct.
N observations correct.
Financial exclusion correct.
Utility exclusion correct.
AT > 0 correct.
SALE ≥ 0 correct.
Missing-data rule correct.
Winsorization correct.
UCT not winsorized unless explicitly changed.
Macro variables not winsorized unless explicitly changed.
Variable formulas match code.
Lead/lag timing matches code.
FE match code.
VCE matches code.
78. Results
Main coefficient matches output.
Standard errors match output.
Significance matches output.
Economic magnitude recomputed.
ST/LT conclusions correct.
Coefficient-difference claims formally tested if stated.
Exposure interaction correct.
Marginal effects correct.
GMM diagnostics complete if used.
IV first stage complete if used.
Failed robustness not hidden.
Placebo/pre-trend results disclosed.
79. Tables and figures
Tables numbered sequentially.
Figures numbered sequentially.
Every table cited in text.
Every figure cited in text.
Table titles informative.
Table notes complete.
Units clear.
FE/VCE disclosed.
N disclosed.
Appendix references correct.
80. References
Every citation appears in References.
Every reference is cited.
DOI verified.
Author names correct.
Publication years correct.
Journal names correct.
Duplicate references removed.
Reference style matches journal.
81. Writing
No unresolved placeholder.
No generic filler.
No repeated paragraphs.
No code names unexplained.
No unexplained abbreviations.
Percentage vs percentage point correct.
Hypotheses described as supported/not supported, not proven.
Abstract matches final paper.
Conclusion matches Introduction.
Title matches strongest result.
82. Submission
Journal selected.
Author guidelines checked.
Word limit checked.
Abstract limit checked.
Keywords checked.
Anonymous manuscript prepared.
Title page prepared.
Cover letter prepared.
Funding statement prepared.
COI statement prepared.
Data statement prepared.
AI disclosure checked.
Highlights prepared if required.
Graphical abstract prepared if required.
All authors approved final version.
PHẦN BR. FINAL 60-MINUTE CHECK
83. Before clicking Submit
Minute 0–10
Read:
- Title;
- Abstract;
- Conclusion.
Ask:
Same story?
Minute 10–20
Check:
- N;
- countries;
- years;
- main β;
- economic magnitude.
Minute 20–30
Check:
- tables;
- notes;
- FE;
- VCE.
Minute 30–40
Check:
- citations;
- References;
- DOI.
Minute 40–50
Check:
- journal requirements;
- blinded file;
- title page;
- declarations.
Minute 50–60
Open every uploaded submission file.
Check:
- correct version;
- figures readable;
- tables not cut;
- equations rendered;
- no comments/track changes;
- no author names in blinded file.
PHẦN BS. DO NOT SUBMIT IF
84.
Do not submit if any of these remain:
- inconsistent sample numbers;
- unresolved placeholders;
- main table not reproduced by final code;
- UCT source unclear;
- Year FE inconsistency;
- questionable IV not explained;
- missing key reference;
- failed major robustness hidden;
- contribution unsupported;
- abstract using outdated results;
- tables with different coefficient from prose.
PHẦN BT. FINAL ARCHIVE
85. After submission
Save:
SUBMISSION_YYYYMMDD
├── MANUSCRIPT_SUBMITTED
├── TITLE_PAGE
├── COVER_LETTER
├── TABLES
├── FIGURES
├── APPENDIX
├── CODE_FINAL
├── OUTPUT_FINAL
└── README
Why?
If revision arrives months later:
you must know exactly which version generated the submitted numbers.
PHẦN BU. REVISION-READY PACKAGE
86. Also retain
- reviewer-response template;
- editable tables;
- editable figures;
- exact final regression logs;
- variable dictionary;
- country sample table.
This will make R&R much faster.
PHẦN BV. MASTER PAPER EVIDENCE FILE
87. File đề xuất
10_FINAL_PAPER_AUDIT.xlsx
Sheet 1 — MANUSCRIPT_MAP
| Section | Source | Status |
|---|
Sheet 2 — VARIABLES
| Manuscript name | Code | Formula | Source |
|---|
Sheet 3 — SAMPLE
| Item | Final value | Source |
|---|
Sheet 4 — NUMBERS
| Statistic | Final value | Table/output | Used where |
|---|
Sheet 5 — CLAIMS
| Claim | Evidence | Citation | Language level |
|---|
Sheet 6 — REFERENCES
| Citation | DOI | Verified | In text? |
|---|
Sheet 7 — TABLES
| Table | Main claim | Final | Appendix |
|---|
Sheet 8 — JOURNAL
| Requirement | Journal rule | Pass? |
|---|
Sheet 9 — SUBMISSION
| File | Required | Final version | Uploaded |
|---|
PHẦN BW. FINAL STUDENT OUTPUT
88. Sinh viên nộp cuối cùng
1. Manuscript
Paper_Final.docx
2. Tables
Tables_Final.xlsx/docx
3. Appendix
Online_Appendix.docx
4. Final Audit
10_FINAL_PAPER_AUDIT.xlsx
5. References
Reference manager export or final bibliography.
6. Code/Output archive
If required by instructor.
PHẦN BX. FINAL PAPER QUALITY RUBRIC
89. Suggested grading rubric
| Component | Weight |
|---|---|
| Research question and contribution | 15% |
| Literature and hypotheses | 15% |
| Data and variable construction | 15% |
| Empirical design | 15% |
| Results and interpretation | 15% |
| Robustness / identification | 10% |
| Discussion / implications | 5% |
| Writing and consistency | 5% |
| References / research integrity | 5% |
Total:
100%
PHẦN BY. AUTOMATIC MAJOR DEDUCTIONS
90. Major errors
Examples:
- invented citation;
- invented coefficient;
- wrong data source;
- copied literature text;
- hidden sample manipulation;
- incorrect FE claim;
- causal claim unsupported by design;
- p-hacking/selective reporting;
- paper cannot be reproduced from submitted code.
These are more serious than formatting errors.
PHẦN BZ. FINAL PAPER CHECK IN ONE PAGE
91. Ten final questions
Before submission, answer YES to all:
- Can I state the paper's contribution in one sentence?
- Does the title reflect that contribution?
- Does the abstract report the actual final finding?
- Does the Introduction explain why ASEAN matters?
- Does Methodology match the final code?
- Do the Results support the paper's claims?
- Do robustness and identification address the main threats?
- Does the Conclusion stay within the evidence?
- Are every citation and number verified?
- Can another researcher understand how the paper was produced?
If any answer is:
No
paper is not final.
PHẦN CA. FINAL 10-STEP WORKFLOW SUMMARY
92. Entire workflow
Bước 1
Data & Methodology
Bước 2
Baseline Results
Bước 3
Robustness
Bước 4
Endogeneity, Identification & Advanced Analyses
Bước 5
Theory, Literature Review & Hypotheses
Bước 6
Full Results & Discussion
Bước 7
Introduction & Contribution Story
Bước 8
Conclusion, Implications & Limitations
Bước 9
Abstract, Title, Keywords & Final Story
Bước 10
Final Assembly, Audit, Formatting & Submission
PHẦN CB. KẾT QUẢ CUỐI CÙNG CỦA HỆ THỐNG
Sau Bước 10, sinh viên không chỉ có:
“một bài NCKH đã viết xong”
mà phải có một reproducible research package gồm:
- research question;
- literature evidence;
- theory;
- hypotheses;
- documented data;
- documented variables;
- reproducible code;
- baseline results;
- robustness;
- identification;
- mechanisms;
- heterogeneity;
- manuscript-ready tables;
- final paper;
- appendix;
- references;
- submission materials.
Mục tiêu cuối cùng là:
Mọi câu trong paper đều có thể truy ngược về một trong ba nguồn: literature, data/code, hoặc regression output.
Nếu không truy ngược được:
không nên để câu đó trong final manuscript.