arxiv: 2604.26198 · v1 · submitted 2026-04-29 · 📊 stat.AP

Recognition: unknown

Pricing Global Macroeconomic Risk in Equity Markets: Evidence from Selected G20 Economies

Vivek Mishra

Pith reviewed 2026-05-07 12:45 UTC · model grok-4.3

classification 📊 stat.AP

keywords global macroeconomic riskdynamic factor modelFama-MacBeth regressionequity marketsG20 countriesasset pricinglatent factorscross-sectional returns

0 comments

The pith

A four-factor model using latent global macroeconomic factors explains cross-sectional equity returns in G20 countries better than the CAPM.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper tests whether global macroeconomic risks are priced in international equity markets by extracting latent factors from a broad set of macro variables. It applies a dynamic factor model to data on inflation, real activity, policy, exchange rates, volatility, and oil prices for ten G20 economies over 2000-2024, then runs Fama-MacBeth regressions on monthly excess returns. A three-factor version shows limited power, but the four-factor version delivers large, significant loadings and a clear rise in explanatory ability. The single-factor CAPM remains stable yet weak by comparison. A sympathetic reader would care because the results point to the need for multiple systematic macro risks rather than market beta alone when pricing cross-border stock returns.

Core claim

The paper claims that model selection supports a parsimonious three-factor dynamic factor specification, yet Fama-MacBeth regressions reveal its low explanatory power for cross-sectional excess returns. A four-factor specification instead produces economically large and statistically significant loadings on all factors, a substantial increase in explanatory power, and improved model performance. The CAPM offers consistent market betas but limited explanatory power due to its single-factor structure. Overall, the macro-driven latent factors extracted through the dynamic factor model supply a more comprehensive and robust framework for international asset pricing than the CAPM.

What carries the argument

The Dynamic Factor Model (DFM) that extracts latent global factors from macroeconomic series on inflation, real activity, monetary policy, term structure, exchange rates, volatility, and oil prices; these factors then enter Fama-MacBeth cross-sectional regressions on G20 equity excess returns.

If this is right

The four-factor specification achieves a significant improvement in explaining cross-sectional variation in excess returns.
All four factors remain statistically significant in the regressions.
The four-factor model balances explanatory power and stability better than the three-factor alternative supported by selection criteria.
The CAPM delivers consistent but limited explanatory power because of its single-factor structure.
Macro-driven latent factors from the DFM supply a more comprehensive framework for international asset pricing than reliance on market risk alone.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Investors in G20 markets could incorporate these global macro factors when constructing portfolios to account for systematic risks beyond local market exposure.
The same dynamic factor approach might be applied to other asset classes such as bonds or currencies to test whether the pricing results generalize.
Out-of-sample tests on additional emerging markets could reveal whether the four-factor structure holds beyond the selected ten countries.
Central banks or regulators might monitor these latent factors as indicators of priced macro risks affecting equity valuations.

Load-bearing premise

The latent factors extracted by the dynamic factor model represent priced systematic global macroeconomic risks rather than statistical artifacts or country-specific noise.

What would settle it

Re-estimating the Fama-MacBeth regressions on post-2024 data or on a broader set of countries and finding that the four-factor loadings lose statistical significance or fail to raise explanatory power.

read the original abstract

This study investigates whether international equity markets systematically price global macroeconomic risks. The empirical analysis is conducted using monthly excess returns for ten G20 countries over the period 2000-2024. A Dynamic Factor Model (DFM) is employed to extract latent global factors from a set of macroeconomic variables capturing global inflation, real activity, monetary policy, term structure, exchange rates, volatility, and oil prices. The model selection criteria of the dynamic factor framework, which support a 3 factor specification that is parsimonious. The Fama MacBeth regressions demonstrate the low explanatory power of the 3-factor model. In contrast, a 4 factor specification results in economically large and statistically significant factor loadings, an obvious rise in explanatory power, and a significant improvement in model performance. The results indicate that a four-factor specification provides the best balance between explanatory power and model stability, significantly improving the ability to explain cross-sectional variation in excess returns , with all factors statistically significant. The Capital Asset Pricing Model, while offering a parsimonious and stable benchmark with consistently significant market betas, exhibits limited explanatory power due to its single factor structure. Overall, the findings suggest that macro driven latent factors extracted through the DFM provide a more comprehensive and empirically robust framework for international asset pricing than the CAPM, highlighting the importance of incorporating multiple sources of systematic risk in explaining cross-country equity returns.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

Standard DFM-Fama-MacBeth run on ten G20 markets with post-hoc four-factor choice and very small cross-section.

read the letter

This is a straightforward empirical application of dynamic factor models to extract latent global macro factors from inflation, activity, policy, term, FX, volatility, and oil series, then tested via Fama-MacBeth on monthly excess returns from ten G20 countries over 2000-2024. It shows the four-factor version beats CAPM on explanatory power and reports significant loadings, which is the main result offered. The data window is current and the variable set is broad, so the setup itself is reasonable for anyone wanting to see macro factors applied in an international panel. The methods follow textbook DFM and Fama-MacBeth steps without obvious algebraic errors. The comparison to CAPM is direct and the abstract is clear about what was done. The soft spot is exactly what the stress-test note flags: only ten countries means the second-stage cross-section has ten observations. Fitting four factors plus an intercept on that scale easily produces higher R-squared and apparent significance through mechanical fit rather than genuine priced risks. The abstract itself says model selection criteria support three factors yet the authors adopt four for better regression performance; that switch needs stronger justification and checks than are shown. No Shanken correction, panel-robust errors, or placebo benchmarks are mentioned. The math and data handling look standard and reproducible on their own terms, but the small N and factor-count decision are load-bearing for the claim that these are priced global risks. This paper is for empirical asset-pricing readers who track macro-factor work on G20 equities. Someone already comfortable with DFM and Fama-MacBeth will find a clean application but little that changes priors. It deserves peer review because the data collection and basic execution are transparent enough for referees to decide whether added robustness fixes the small-sample and selection issues or whether the central claim does not hold.

Referee Report

3 major / 2 minor

Summary. The paper claims that latent global macroeconomic factors extracted via a dynamic factor model (DFM) from variables including inflation, real activity, monetary policy, term structure, exchange rates, volatility, and oil prices can price cross-sectional equity excess returns across 10 G20 countries (2000-2024) better than the CAPM. Model selection criteria favor three factors, but the authors adopt a four-factor specification because it yields economically large, statistically significant loadings and higher explanatory power in Fama-MacBeth regressions; the four-factor model is presented as providing the best balance of explanatory power and stability.

Significance. If the central claim holds after addressing sample-size and selection concerns, the work would provide evidence that multiple latent global macro risks are priced in international equities, extending asset-pricing tests beyond single-factor benchmarks. However, the small cross-section (N=10) and post-hoc factor choice limit the strength of any such conclusion; the paper does not report machine-checked proofs, reproducible code, or falsifiable out-of-sample predictions.

major comments (3)

[Abstract] Abstract: model selection criteria are stated to support a 3-factor DFM, yet the authors adopt 4 factors solely because they deliver higher explanatory power and significant loadings in the Fama-MacBeth stage. This post-hoc selection is load-bearing for the central claim that the 4-factor specification is optimal; no pre-specified robustness checks comparing 3- versus 4-factor results or information-criterion penalties are described.
[Fama-MacBeth regressions (implied by abstract)] Fama-MacBeth cross-sectional tests are performed on only 10 countries. With N=10 in the second-stage regression, any multi-factor specification can mechanically inflate R² and produce apparent statistical significance; the manuscript gives no indication of Shanken (1992) errors-in-variables corrections, Newey-West adjustments, adjusted R² comparisons, or placebo-factor benchmarks that would be required to support the claim that the DFM factors represent priced systematic risks rather than sample artifacts.
[Abstract and empirical setup] The weakest assumption—that the latent DFM factors represent priced global macroeconomic risks rather than statistical artifacts or country-specific noise—is not tested via any cross-validation, out-of-sample pricing tests, or comparison against country-specific factors.

minor comments (2)

[Abstract] Abstract contains minor grammatical issues (e.g., 'an obvious rise in explanatory power, and a significant improvement'; stray comma before 'with all factors').
[Data and methodology] The manuscript should clarify the exact macroeconomic series used, the precise DFM estimation procedure (e.g., number of lags, normalization), and whether factor loadings are estimated jointly or in two steps.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the thoughtful and constructive comments, which highlight important issues regarding factor selection, statistical robustness with a small cross-section, and validation of the global factors. We address each major comment below and outline specific revisions to strengthen the manuscript.

read point-by-point responses

Referee: [Abstract] Abstract: model selection criteria are stated to support a 3-factor DFM, yet the authors adopt 4 factors solely because they deliver higher explanatory power and significant loadings in the Fama-MacBeth stage. This post-hoc selection is load-bearing for the central claim that the 4-factor specification is optimal; no pre-specified robustness checks comparing 3- versus 4-factor results or information-criterion penalties are described.

Authors: We agree that the post-hoc adoption of four factors based on Fama-MacBeth performance requires stronger justification and pre-specified checks. The dynamic factor model selection criteria do favor three factors for the macroeconomic panel. However, the four-factor specification produces economically large, statistically significant loadings and a clear improvement in cross-sectional explanatory power. In the revision, we will add a dedicated subsection comparing the three- and four-factor models using the full set of information criteria, factor loadings, and Fama-MacBeth R-squared values. We will also report the associated penalties and discuss the economic rationale for retaining the fourth factor as providing the best balance of fit and stability. revision: partial
Referee: [Fama-MacBeth regressions (implied by abstract)] Fama-MacBeth cross-sectional tests are performed on only 10 countries. With N=10 in the second-stage regression, any multi-factor specification can mechanically inflate R² and produce apparent statistical significance; the manuscript gives no indication of Shanken (1992) errors-in-variables corrections, Newey-West adjustments, adjusted R² comparisons, or placebo-factor benchmarks that would be required to support the claim that the DFM factors represent priced systematic risks rather than sample artifacts.

Authors: The limited cross-section of ten G20 economies is indeed a constraint that warrants additional safeguards. We will revise the empirical section to implement Shanken (1992) corrections for errors-in-variables bias and Newey-West standard errors. We will also report adjusted R-squared statistics and include placebo tests in which the DFM factors are replaced by randomly generated factors to assess whether the observed significance and R-squared gains exceed what would arise mechanically. These additions will help demonstrate that the results are not artifacts of the small N. revision: yes
Referee: [Abstract and empirical setup] The weakest assumption—that the latent DFM factors represent priced global macroeconomic risks rather than statistical artifacts or country-specific noise—is not tested via any cross-validation, out-of-sample pricing tests, or comparison against country-specific factors.

Authors: We acknowledge that in-sample fit alone is insufficient to establish the factors as priced global risks. In the revised manuscript we will add out-of-sample pricing tests by estimating the DFM and Fama-MacBeth regressions on the 2000–2015 subsample and evaluating performance on the 2016–2024 hold-out period. We will also compare the explanatory power of the global DFM factors against country-specific factor benchmarks to quantify incremental contribution. These tests will directly address concerns about statistical artifacts versus genuine global macroeconomic risks. revision: yes

Circularity Check

0 steps flagged

No circularity: factors extracted independently from macro data before asset-pricing tests

full rationale

The derivation begins with a standard Dynamic Factor Model applied to a panel of macroeconomic series (inflation, activity, policy, etc.) to extract latent factors; this step uses only the macro panel and standard DFM estimation criteria. The resulting factors are then fed into separate Fama-MacBeth cross-sectional regressions on equity excess returns. No equation defines the factors in terms of the return data, renames a fitted quantity as a prediction, or relies on a self-citation chain for the uniqueness or validity of the factor extraction. The paper explicitly notes that information criteria favor three factors yet reports four-factor results for improved explanatory power; this is an empirical trade-off, not a definitional reduction. The overall chain therefore remains self-contained against external benchmarks and does not collapse to its inputs by construction.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on the assumption that DFM-extracted factors capture priced global risks and that Fama-MacBeth cross-sectional regressions validly identify risk premia; the choice of four factors is data-driven.

free parameters (1)

number of factors
Model selection criteria indicate three factors are parsimonious, yet four are retained because they improve explanatory power in the regressions.

axioms (1)

domain assumption The latent factors from the dynamic factor model represent systematic global macroeconomic risks that are priced in equity markets.
Invoked when interpreting the Fama-MacBeth loadings as evidence of priced risk.

pith-pipeline@v0.9.0 · 5537 in / 1270 out tokens · 44691 ms · 2026-05-07T12:45:48.606842+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

27 extracted references · 18 canonical work pages

[2]

P., Moulton, P

E. F. Fama and K. R. French, “A five-factor asset pricing model,” J. financ. econ., vol. 116, no. 1, pp. 1–22, Feb. 2015, doi: 10.1016/j.jfineco.2014.10.010

work page doi:10.1016/j.jfineco.2014.10.010 2015
[3]

A six-factor asset pricing model,

R. Roy and S. Shijin, “A six-factor asset pricing model,” Borsa Istanbul Review, vol. 18, no. 3, pp. 205–217, Feb. 2018, doi: 10.1016/j.bir.2018.02.001

work page doi:10.1016/j.bir.2018.02.001 2018
[4]

The relationship between earnings’ yield, market value and return for NYSE common stocks,

S. Basu, “The relationship between earnings’ yield, market value and return for NYSE common stocks,” J. financ. econ., vol. 12, no. 1, pp. 129–156, Feb. 1983, doi: 10.1016/0304-405X(83)90031-4

work page doi:10.1016/0304-405x(83)90031-4 1983
[5]

Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency,

N. JEGADEESH and S. TITMAN, “Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency,” J. Finance, vol. 48, no. 1, pp. 65–91, Feb. 1993, doi: 10.1111/j.1540-6261.1993.tb04702.x

work page doi:10.1111/j.1540-6261.1993.tb04702.x 1993
[6]

An Intertemporal Capital Asset Pricing Model,

R. C. Merton, “An Intertemporal Capital Asset Pricing Model,” Econometrica, vol. 41, no. 5, p. 867, Feb. 1973, doi: 10.2307/1913811

work page doi:10.2307/1913811 1973
[8]

Intertemporal asset pricing,

J. Shanken, “Intertemporal asset pricing,” J. Econom., vol. 45, no. 1–2, pp. 99–120, Feb. 1990, doi: 10.1016/0304- 4076(90)90095-B

work page doi:10.1016/0304- 1990
[9]

Forecasting Using Principal Components From a Large Number of Predictors,

J. H. Stock and M. W. Watson, “Forecasting Using Principal Components From a Large Number of Predictors,” J. Am. Stat. Assoc., vol. 97, no. 460, pp. 1167– 1179, Dec. 2002, doi: 10.1198/016214502388618960

work page doi:10.1198/016214502388618960 2002
[10]

Macroeconomic Forecasting Using Diffusion Indexes,

J. H. Stock and M. W. Watson, “Macroeconomic Forecasting Using Diffusion Indexes,” Journal of Business & Economic Statistics, vol. 20, no. 2, pp. 147–162, Feb. 2002, doi: 10.1198/073500102317351921

work page doi:10.1198/073500102317351921 2002
[11]

The empirical risk–return relation: A factor analysis approach☆,

S. C. Ludvigson and S. Ng, “The empirical risk–return relation: A factor analysis approach☆,” J. financ. econ., vol. 83, no. 1, pp. 171–222, Jan. 2007, doi: 10.1016/j.jfineco.2005.12.002

work page doi:10.1016/j.jfineco.2005.12.002 2007
[12]

Macro Factors in Bond Risk Premia,

S. C. Ludvigson and S. Ng, “Macro Factors in Bond Risk Premia,” Review of Financial Studies, vol. 22, no. 12, pp. 5027–5067, Dec. 2009, doi: 10.1093/rfs/hhp081

work page doi:10.1093/rfs/hhp081 2009
[13]

The Generalized Dynamic-Factor Model: Identification and Estimation,

M. Forni, M. Hallin, M. Lippi, and L. Reichlin, “The Generalized Dynamic-Factor Model: Identification and Estimation,” Review of Economics and Statistics, vol. 82, no. 4, pp. 540–554, Feb. 2000, doi: 10.1162/003465300559037

work page doi:10.1162/003465300559037 2000
[14]

The generalized dynamic factor model consistency and rates,

M. Forni, M. Hallin, M. Lippi, and L. Reichlin, “The generalized dynamic factor model consistency and rates,” J. Econom., vol. 119, no. 2, pp. 231–255, Feb. 2004, doi: 10.1016/S0304-4076(03)00196-9

work page doi:10.1016/s0304-4076(03)00196-9 2004
[15]

Financial Intermediaries and the Cross‐Section of Asset Returns,

T. ADRIAN, E. ETULA, and T. MUIR, “Financial Intermediaries and the Cross‐Section of Asset Returns,” J. Finance, vol. 69, no. 6, pp. 2557–2596, Feb. 2014, doi: 10.1111/jofi.12189

work page doi:10.1111/jofi.12189 2014
[16]

A Two-Step Estimator for Large Approximate Dynamic Factor Models Based on Kalman Filtering,

C. and G. D. and R. L. Doz, “A Two-Step Estimator for Large Approximate Dynamic Factor Models Based on Kalman Filtering,” 2007, London

2007
[17]

A Quasi– Maximum Likelihood Approach for Large, Approximate Dynamic Factor Models,

C. Doz, D. Giannone, and L. Reichlin, “A Quasi– Maximum Likelihood Approach for Large, Approximate Dynamic Factor Models,” Review of Economics and Statistics, vol. 94, no. 4, pp. 1014–1024, Feb. 2012, doi: 10.1162/REST_a_00225

work page doi:10.1162/rest_a_00225 2012
[18]

Likelihood‐based dynamic factor analysis for measurement and forecasting,

B. Jungbacker and S. J. Koopman, “Likelihood‐based dynamic factor analysis for measurement and forecasting,” Econom. J., vol. 18, no. 2, pp. C1–C21, Feb. 2015, doi: 10.1111/ectj.12029

work page doi:10.1111/ectj.12029 2015
[19]

Measuring US aggregate output and output gap using large datasets. ,

M. B. M. & Luciani, “Measuring US aggregate output and output gap using large datasets. ,” 2018

2018
[20]

Forecasting stock returns with large dimensional factor models,

A. Giovannelli, D. Massacci, and S. Soccorsi, “Forecasting stock returns with large dimensional factor models,” J. Empir. Finance, vol. 63, pp. 252–269, Feb. 2021, doi: 10.1016/j.jempfin.2021.07.009

work page doi:10.1016/j.jempfin.2021.07.009 2021
[21]

Dynamic factor models: Does the specification matter?,

K. Miranda, P. Poncela, and E. Ruiz, “Dynamic factor models: Does the specification matter?,” SERIEs, vol. 13, no. 1–2, pp. 397–428, Feb. 2022, doi: 10.1007/s13209-021- 00248-2

work page doi:10.1007/s13209-021- 2022
[22]

Multidimensional dynamic factor models,

F. B. M. & Pellegrino, “Multidimensional dynamic factor models,” 2023

2023
[23]

O. F. Akbal, Regime-Switching Factor Models and Nowcasting with Big Data. 2024

2024
[24]

Yahoo Finance. (2025). Historical equity index data. Retrieved February 14, 2026, from https://finance.yahoo.com

“Yahoo Finance. (2025). Historical equity index data. Retrieved February 14, 2026, from https://finance.yahoo.com.”

2025
[25]

World Development Indicators: World Bank DataBank

World Bank, “World Development Indicators: World Bank DataBank.” Accessed: Jan. 15, 2026. [Online]. Available: https://databank.worldbank.org

2026
[26]

International Financial Statistics

International Monetary Fund, “International Financial Statistics.” Accessed: Jan. 15, 2026. [Online]. Available: https://data.imf.org

2026
[27]

CBOE Volatility Index (VIX) Historical Data,

Chicago Board Options Exchange, “CBOE Volatility Index (VIX) Historical Data,” CBOE. Accessed: Jan. 15, 2026.[Online].Available: https://www.cboe.com/tradable_products/vix/

2026
[28]

Federal Reserve Economic Data (FRED): Crude Oil Prices: West Texas Intermediate (WTI),

Federal Reserve Bank of St. Louis, “Federal Reserve Economic Data (FRED): Crude Oil Prices: West Texas Intermediate (WTI),” FRED. Accessed: Jan. 15, 2026. [Online]. Available: https://fred.stlouisfed.org

2026
[29]

Risk, Return, and Equilibrium: Empirical Tests,

E. F. Fama and J. D. MacBeth, “Risk, Return, and Equilibrium: Empirical Tests,” Journal of Political Economy, vol. 81, no. 3, pp. 607–636, Feb. 1973, doi: 10.1086/260061

work page doi:10.1086/260061 1973