Paper: arXiv 2512.14134
Authors: Sungwoo Kang
Abstract
Chae and Kang (2019, \textit{“Pacific-Basin Finance Journal”}) documented a puzzling Low Volume Return Premium (LVRP) in Korea – contradicting global High Volume Return Premium (HVRP) evidence. We resolve this puzzle. Using Korean market data (2020-2024), we demonstrate that HVRP exists in Korea but is masked by (1) pooling heterogeneous investor types and (2) using inappropriate intensity normalization. When institutional buying intensity is normalized by market capitalization rather than trading value, a perfect monotonic relationship emerges: highest-conviction institutional buying (Q4) generates +\institutionLedQFourDayPlusFiftyCAR\ cumulative abnormal returns over 50 days, while lowest-intensity trades (Q1) yield modest returns (+\institutionLedQOneDayPlusFiftyCAR). Retail investors exhibit a flat pattern – their trading generates near-zero returns regardless of conviction level – confirming the pure noise trader hypothesis. During the Donghak Ant Movement (2020-2021), however, coordinated retail investors temporarily transformed from noise traders to liquidity providers, generating returns comparable to institutional trading. Our findings reconcile conflicting international evidence and demonstrate that detecting informed trading signals requires investor-type decomposition, nonlinear quartile analysis, and conviction-based (market cap) rather than participation-based (trading value) measurement.
Complexity vs Empirical Score
- Math Complexity: 3.0/10
- Empirical Rigor: 6.0/10 (capped from 8.0: transaction costs not modelled; no out-of-sample test; see penalty flags)
- Quadrant: Street Traders — empirical and applied, lower math complexity (effective empirical rigor capped from 8.0)
Why this score: The paper’s math is primarily statistical (quartile analysis, CAR) and does not involve advanced theoretical derivations, keeping complexity low. However, it is highly data-driven, using specific Korean market data (2020-2024), performing investor-type decomposition, and presenting detailed empirical results like cumulative abnormal returns, making it backtest-ready and empirically rigorous.
Research Flowchart
flowchart TD
A["Research Goal<br>Resolve LVRP vs HVRP Puzzle in Korea"] --> B["Data & Methodology<br>Korean Market (2020-2024)<br>Investor-Type Decomposition<br>Quartile Analysis"]
B --> C{"Normalization Test<br>Market Cap vs Trading Value"}
C -- Trading Value --> D["Masked Signal<br>Low Volume Premium"]
C -- Market Cap --> E["Exposed Signal<br>High Volume Premium"]
E --> F{"Investor Identity Analysis"}
F --> G["Institutions<br>Monotonic HVRP<br>Q4: +50CAR<br>Q1: +LowCAR"]
F --> H["Retail Investors<br>Flat Pattern<br>Noise Traders"]
H --> I["Exception<br>Donghak Ant Movement<br>Temp. Informed"]
G & I --> J["Key Findings<br>Decomposition Required<br>Conviction > Participation<br>Reconciles Global Evidence"]
Evidence Card
What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card
Proposes
The paper introduces a double‑sort methodology that normalizes trading intensity by market capitalization to uncover a monotonic high‑volume return premium.
Source sentence
Our double-sort methodology reveals a striking pattern. When we normalize institutional buying intensity by market capitalization (a proxy for conviction), a perfect monotonic relationship emerges: events in the highest conviction quartile (Q4) generate +12.12% cumulative abnormal returns over 50 days, while the lowest
Assumes 6
- Abnormal returns are calculated using a market‑adjusted return model. model
Source sentence
Abnormal returns (AR) use the market‑adjusted return model (ARi,t = Ri,t −RM,t).
- Events within 50 trading days of each other are treated as dependent and excluded. data
Source sentence
Events occurring within 50 trading days after an initial event are considered dependent events and excluded from analysis.
- Event windows are set to [-50, +50] trading days. model
Source sentence
All analyses use event windows of [-50, +50] trading days with ADTV threshold = 5.
- Abnormal volume events are identified by an ADTV threshold greater than 5. data
Source sentence
Only 2.68% of observations exceed the ADTV threshold of 5, yielding our event sample of approximately 73,500 total abnormal volume events.
Show 2 more
- Investor type classification uses the net purchase ratio (percentage of total trading value) as the initial metric. model
Source sentence
Each abnormal volume event is classified according to the net purchase ratios of three investor types (institutions, foreigners, retail) using an argmax rule.
- Trading intensity is measured as position size relative to firm value (market cap). model
Source sentence
When institutional buying intensity is normalized by market capitalization rather than trading value, a perfect monotonic relationship emerges.
Data
2020 – 2024 · Korean equities · equities
- Korean market data market-prices unknown
Source sentence
Using Korean market data (2020-2024), we demonstrate that HVRP exists in Korea but is masked by (1) pooling heterogeneous investor types and (2) using inappropriate intensity normalization.
Costs
Unclear - sources disagree
Finds 5
total return 12.1%
- Institutional high‑conviction trades (Q4) generate a 12.12% 50‑day cumulative abnormal return, versus 4.65% for low‑conviction (Q1). performance
Source sentence
When we normalize institutional buying intensity by market capitalization (a proxy for conviction), a perfect monotonic relationship emerges: events in the highest conviction quartile (Q4) generate +12.12% cumulative abnormal returns over 50 days, while the lowest quartile (Q1) yields modest returns (+4.65%).
- Retail‑led events show a flat intensity‑return pattern with near‑zero returns regardless of conviction level. performance
Source sentence
Retail investors exhibit a flat pattern -- their trading generates near-zero returns regardless of conviction level -- confirming the pure noise trader hypothesis.
- Overall HVRP persists in the modern sample, with event‑day AAR increasing from 5.06% to 6.15% (≈22% higher) and statistical significance improving. statistical
Source sentence
Key findings: 1. Increased Effect Size: Event day AAR increased from 5.06% to 6.15%, approximately 22% higher
- During the Donghak Ant Movement, retail‑led events generated 3.53% returns, a temporary shift from noise‑trader behavior. performance
Source sentence
During the Donghak Ant Movement, retail-led events generated 3.53% returns (versus near-zero in normal periods), indicating that collective action through social media temporarily transformed retail investors from pure noise traders into meaningful market participants.
- The near‑zero correlation between the Dominance Score and returns (r<0.04, p>0.16) reflects a nonlinear, type‑dependent relationship rather than irrelevance of intensity. statistical
Source sentence
The near-zero correlation we observe with the Dominance Score (r<0.04, p>0.16) does not indicate that intensity is irrelevant; it reveals that the relationship is nonlinear and type-dependent.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 44 related papers). Coverage is partial, so this is neither replication nor refutation.
Code
No repository by the authors found in the paper.
Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.
Cited References (37)
- Abnormal Increase Trading Volume Effect on the Stock Return (2006)
- Trading Behavior, Performance, and Stock Preference of Foreigners, Local Institutions, and Individual Investors: Evidence from the Korean Stock Market (2011) — link
- The Unprecedented Stock Market Reaction to COVID-19 (2020) — link
- All That Glitters: The Effect of Attention and News on the Buying Behavior of Individual and Institutional Investors (2007) — link
- Attention-Induced Trading and Returns: Evidence from Robinhood Users (2022) — link
- Resolving a paradox: Retail trades positively predict returns but are not profitable (2024)
- Market Statistics and Technical Analysis: The Role of Volume (1994) — link
- Tracking Retail Investor Activity (2021) — link
and 29 more
- Place Your Bets? The Value of Investment Research on Reddit’s Wallstreetbets (2023) — link
- Trading Volume and Serial Correlation in Stock Returns* (1993) — link
- Caught on tape: Institutional trading, stock returns, and earnings announcements (2009) — link
- Do foreign investors destabilize stock markets? The Korean experience in 1997 (1999) — link
- In Search of Attention (2011) — link
- Speculative dynamics of prices and volume (2022) — link
- The High-Volume Return Premium (2001) — link
- Expected return, volume, and mispricing (2022) — link
- Is there a dark side to exchange traded funds? An information perspective (2017) — link
- A Rational Theory of Mutual Funds’ Attention Allocation (2016) — link
- Low-volume return premium in the Korean stock market (2019) — link
- Individual Investor Trading and Stock Returns (2008) — link
- The high volume return premium: Cross-country evidence (2012) — link
- How Wise Are Crowds? Insights from Retail Orders and Stock Returns (2013) — link
- Behavioral Biases and the Trading of Individual Investors in the Korean Stock Markets (2022)
- Economic Analysis of Donghak Ants Movement (2021) — link
- Individual investor trading and stock returns after the Covid-19 pandemic: Evidence from Korea (2024) — link
- Retail Trading and Return Predictability in China (2025)
- Improved Methods for Tests of Long-Run Abnormal Stock Returns (1999) — link
- Short-term Momentum (2021) — link
- A simple model of capital market equilibrium with incomplete information (1987)
- Threshold effect in the relationship between investor sentiment and stock market returns: a PSTR specification (2018) — link
- COVID-19 and investor behavior (2020) — link
- Flattening the Illiquidity Curve: Retail Trading During the COVID-19 Lockdown (2021)
- Investor attention, overconfidence and category learning (2006) — link
- Estimating Standard Errors in Finance Panel Data Sets: Comparing Approaches (2008) — link
- Overconfidence and Speculative Bubbles (2003) — link
- Investor Overconfidence and Trading Volume (2006) — link
- Who trades with whom? Individuals, institutions, and returns (2014) — link
0 of 37 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.