arxiv: 2605.08726 · v1 · submitted 2026-05-09 · 💱 q-fin.GN

Recognition: 2 theorem links

· Lean Theorem

The effect of investor-driven information diffusion on excess comovement: Evidence from retail and institutional investors in China and the United States

Fei Ren, Miao-Miao Yi, Xiang Gao, Zhang-Hangjian Chen

Pith reviewed 2026-05-12 02:26 UTC · model grok-4.3

classification 💱 q-fin.GN

keywords information diffusionexcess comovementretail investorsinstitutional investorsChinese stock marketU.S. stock markettrading behaviorlead-lag relationship

0 comments

The pith

Retail investors drive excess stock comovement in China while institutions do so in the United States.

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

This paper examines how information about one stock spreads to others through the trading of retail and institutional investors, and how that spread creates excess comovement beyond what company fundamentals alone would produce. It compares the retail-heavy Chinese market to the institution-heavy U.S. market using trading data on thousands of stocks from 2010 to 2022. The authors show that the dominant investor group in each country accounts for most of the excess comovement. They trace the link to actual trading actions rather than other market forces and document a timing pattern in which faster-spreading information leads to earlier comovement. The findings indicate that the type of investor who moves the market most strongly shapes how stocks cluster together in different countries.

Core claim

The dominant investor group in each market significantly drives excess comovement. In China, retail-driven information diffusion has a notably stronger effect on excess comovement than institution-driven diffusion. In the United States, institution-driven diffusion is the primary driver, surpassing the influence of retail-driven diffusion. Investors' trading behavior serves as the underlying mechanism. Stocks with faster retail-driven information diffusion exhibit comovement that precedes those with slower diffusion, and the predictive power of these diffusion measures varies by market: retail-driven diffusion predicts excess comovement persistently in China, while institution-driven shows a

What carries the argument

Measures of retail-driven and institution-driven information diffusion built from trading data, which connect investor type to excess comovement through observed trading patterns and lead-lag timing.

Load-bearing premise

The measures of retail-driven and institution-driven information diffusion cleanly separate the causal role of investor type from market microstructure, liquidity differences, or regulatory features that distinguish China from the United States.

What would settle it

A direct test that recomputes the diffusion measures after matching stocks across countries on liquidity, volume, and regulatory exposure, then checks whether the China-U.S. difference in which investor group drives comovement disappears.

read the original abstract

This study investigates how cross-stock information diffusion, driven by both retail and institutional investors, influences excess comovement in the Chinese retail-dominated market and the U.S. institution-dominated market. Using data from 4,533 Chinese stocks and 4,517 U.S. stocks from 2010 to 2022, we identify three key findings. First, the dominant investor group in each market significantly drives excess comovement. Specifically, in China, compared with institution-driven diffusion, retail-driven information diffusion has a notably stronger effect on excess comovement. In contrast, in the U.S., institution-driven diffusion is the primary driver of excess comovement, surpassing the influence of retail-driven diffusion. Second, we identify investors' trading behavior as the underlying mechanism through which information diffusion affects excess comovement. Third, we observe a lead-lag relationship: stocks with faster retail-driven information diffusion exhibit comovement that precedes those with slower diffusion. Based on this finding, we further demonstrate that the predictive power of information diffusion varies across markets. In China, retail-driven diffusion shows strong and persistent predictability for excess comovement, whereas in the U.S., institution-driven diffusion exhibits similarly robust predictive capacity.

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

Retail-driven diffusion links more to comovement in China while institutions do in the US, but cross-country confounders weaken the causal attribution.

read the letter

The key point is that retail-driven information diffusion ties more strongly to excess comovement in China than institutional diffusion does, while the reverse is true in the US. The work also turns up lead-lag predictability where faster diffusion stocks lead comovement in ways that differ by market. Large samples from both countries over a decade give the comparisons some weight, and breaking out trading behavior as a mechanism helps ground the story. The predictability angle is a step beyond static correlations. The soft spot sits in the identification. China and the US differ in many ways beyond investor mix, including how trades execute, rules on short sales, and overall liquidity. Without clear ways to isolate the investor-type channel from those factors, the results could reflect broader market features instead. The lead-lag findings carry the same risk if diffusion speed just tracks liquidity differences. This kind of work suits readers focused on how investor composition affects price dynamics and comovement. Someone building models of information flow across borders might pull the comparative patterns as motivation. It should go to peer review. The scale of the data and the specific cross-market angle make it worth referees checking the details on controls and alternative explanations.

Referee Report

3 major / 3 minor

Summary. The paper claims that investor-driven information diffusion affects excess comovement differently based on the dominant investor type in each market. In the retail-dominated Chinese market, retail-driven diffusion has a stronger impact on excess comovement compared to institution-driven diffusion, while in the institution-dominated U.S. market, the reverse holds. Using data from 4,533 Chinese and 4,517 U.S. stocks over 2010-2022, the authors identify trading behavior as the mechanism and document lead-lag relationships where faster diffusion predicts comovement, with predictive power being market-specific (retail in China, institutions in US).

Significance. If the central claims survive rigorous checks for endogeneity and confounding, the work would provide valuable cross-country evidence on how investor composition shapes asset price dynamics and information flow in retail- versus institution-dominated markets. The mechanism identification via trading behavior and the documented predictability differences could inform behavioral finance models and regulatory discussions on market efficiency. The large samples strengthen generalizability, but the cross-market attribution requires careful separation from microstructure and liquidity factors.

major comments (3)

[Abstract and §3 (Methodology)] Abstract and §3 (Methodology): The abstract reports large samples and three main findings but provides no information on identification strategy, control variables, robustness checks, or how diffusion is measured without mechanical correlation to comovement itself. This is load-bearing for the central claim that dominant investor groups drive excess comovement, as endogeneity between trading activity and price movements could confound the retail-institution comparison across markets.
[§5 (Empirical Results)] §5 (Empirical Results): The lead-lag relationship and predictability findings inherit the same risk; faster diffusion may simply proxy for stocks with higher retail/institutional liquidity rather than information flow per se. Without explicit separation via within-country variation, instruments, or matched microstructure controls, the dominant-investor interpretation rests on an untested exclusion restriction.
[§4 (Data and Sample)] §4 (Data and Sample): Systematic differences in trading mechanisms, ownership concentration, short-sale rules, and liquidity regimes between China and the U.S. are not addressed with matched samples or additional controls, which could drive the reported differences in retail- versus institution-driven effects rather than investor type dominance.

minor comments (3)

Clarify the exact construction of retail-driven and institution-driven information diffusion measures, including data sources for investor classification and any lags used.
Add economic magnitude (e.g., coefficient sizes and standardized effects) to the abstract and main results to complement the qualitative statements about 'notably stronger' effects.
Ensure tables include full robustness specifications and report both statistical and economic significance consistently.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive comments, which help clarify the identification challenges in our cross-market analysis. We respond point-by-point to the major comments, providing clarifications from the manuscript and outlining targeted revisions to address concerns about endogeneity, liquidity proxies, and institutional differences.

read point-by-point responses

Referee: [Abstract and §3 (Methodology)] Abstract and §3 (Methodology): The abstract reports large samples and three main findings but provides no information on identification strategy, control variables, robustness checks, or how diffusion is measured without mechanical correlation to comovement itself. This is load-bearing for the central claim that dominant investor groups drive excess comovement, as endogeneity between trading activity and price movements could confound the retail-institution comparison across markets.

Authors: We agree the abstract should summarize the identification approach. In §3, investor-driven diffusion is measured via lagged trading volumes and lead-lag return correlations specific to retail versus institutional trades, avoiding direct mechanical overlap with contemporaneous comovement. Specifications include stock and time fixed effects plus controls for liquidity (turnover, Amihud illiquidity), volatility, and size. Robustness checks encompass alternative diffusion proxies and subsample splits. We will revise the abstract to briefly note the lead-lag structure, key controls, and fixed effects. revision: yes
Referee: [§5 (Empirical Results)] §5 (Empirical Results): The lead-lag relationship and predictability findings inherit the same risk; faster diffusion may simply proxy for stocks with higher retail/institutional liquidity rather than information flow per se. Without explicit separation via within-country variation, instruments, or matched microstructure controls, the dominant-investor interpretation rests on an untested exclusion restriction.

Authors: The §5 lead-lag tests are run separately within each country using within-stock time-series variation and include liquidity controls (turnover, spreads) plus stock fixed effects. The predictive design (diffusion at t forecasting comovement at t+1) reduces simultaneity concerns. We lack a formal instrument but will add liquidity-quartile matched subsamples and extra microstructure controls as robustness checks to further isolate information-flow effects from liquidity. revision: partial
Referee: [§4 (Data and Sample)] §4 (Data and Sample): Systematic differences in trading mechanisms, ownership concentration, short-sale rules, and liquidity regimes between China and the U.S. are not addressed with matched samples or additional controls, which could drive the reported differences in retail- versus institution-driven effects rather than investor type dominance.

Authors: All core specifications are estimated separately by market to respect regime differences. §4 already incorporates liquidity controls, but we will add proxies for short-sale constraints and ownership concentration. We will also implement propensity-score matched subsamples on microstructure and ownership variables to better isolate investor-type dominance from institutional differences. revision: yes

Circularity Check

0 steps flagged

Empirical analysis with no circular derivation chain

full rationale

The paper is a purely empirical study that constructs proxies for retail- and institution-driven information diffusion from trading data, then estimates their statistical associations with excess comovement via regressions, lead-lag tests, and predictability checks on 2010-2022 stock samples from two markets. No equations, fitted parameters, or self-citations are invoked to derive the central claims; all reported effects are direct outputs of external market observations rather than reductions to quantities defined inside the study. The analysis therefore contains no self-definitional, fitted-input, or self-citation-load-bearing steps.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

No formal model or new theoretical entities are introduced; the paper is purely empirical. Any regression parameters or diffusion-speed cutoffs are fitted to the data but are not enumerated in the abstract.

pith-pipeline@v0.9.0 · 5525 in / 1351 out tokens · 49137 ms · 2026-05-12T02:26:44.046852+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear
retail-driven information diffusion measured by co-investors' posts/replies and triggered replies (Eq. 3); institution-driven by summed ownership weights (Eq. 4); excess comovement as Pearson corr of FF5 residuals (Eq. 2); baseline regressions (Eqs. 7-9) with stock×quarter FE
IndisputableMonolith/Foundation/ArithmeticFromLogic.lean absolute_floor_iff_bare_distinguishability unclear
lead-lag via quintiles on diffusion intensity; mechanism via BSI correlation (Eqs. 5-6); robustness with PSM, IV (app launches, mergers), alternative FF3/CAPM

Reference graph

Works this paper leans on

2 extracted references · 2 canonical work pages

[1]

A., Engelberg, J

https://doi.org/10.1111/j.1540-6261.2008.01379.x Cookson, J. A., Engelberg, J. E., Mullins, W., 2022. Echo chambers. Rev. Financ. Stud. 36(2), 33 450–500. https://doi.org/10.1093/rfs/hhac058 Cookson, J. A., Lu, R., Mullins, W., Niessner, M., 2024. The social signal. J. Financ. Econ. 158, 103870. https://doi.org/10.1016/j.jfineco.2024.103870 Cookson, J. A....

work page doi:10.1111/j.1540-6261.2008.01379.x 2008
[2]

P., Moulton, P

https://doi.org/10.1016/j.jfineco.2014.10.010 Gao, G. P., Moulton, P. C., Ng, D. T. , 2017. Institutional ownership and return predictability across economically unrelated stocks. J . Financ. Intermed. 31, 45–63. https://doi.org/10. 1016/j.jfi.2016.07.004 Ge, S., Li, S., Zheng, H., 2025. Diamond cuts diamond: News co-mention momentum spillover prevails in...

work page doi:10.1016/j.jfineco.2014.10.010 2014