Explore key market anomalies, why they persist, and how investors can both capitalize on and critically evaluate them in light of the Efficient Market Hypothesis (EMH).
Have you ever heard someone say, “Markets are efficient; everything’s already priced in, so you won’t beat the market”? Well, that’s the crux of the Efficient Market Hypothesis (EMH). Yet, look around and you’ll find studies—some quite famous—pointing out areas where stock prices behave in ways that the EMH can’t fully explain. These irregularities are called market anomalies. And, honestly, they can be a bit puzzling.
Market anomalies are patterns or behaviors in prices and returns that should not exist if markets were truly efficient in the classic sense. They might be due to a hidden risk factor that we haven’t fully captured, or they might be due to good old-fashioned mispricing driven by investor psychology (like overexcitement or panic). Let’s explore:
Market anomalies refer to observed return patterns that deviate from what prevailing asset pricing models (like the Capital Asset Pricing Model) predict. They can be as simple as noticing that small-cap stocks do better in January—or as complex as finding nuanced momentum signals in certain industries.
In textbooks and academic papers, an “anomaly” is basically an empirical result that’s inconsistent with a given theory. In this context, the “theory” is that markets are efficient and that no consistently exploitable patterns should exist. But anomalies occasionally defy that expectation.
One of the most talked-about anomalies is the “January Effect,” when small-cap stocks, in particular, seem to rally a bit more in—surprise—January. Maybe it’s tax-loss selling in December, maybe it’s portfolio rebalancing, maybe it’s just folks feeling giddy about the new year. Whatever the reason, this effect has been noticed across decades. Other calendar anomalies include the “Monday Effect,” where returns on Mondays appear systematically lower or higher than on other days (though the evidence can be mixed).
Momentum strategies focus on buying stocks that recently performed well and selling those that performed poorly. If markets were entirely efficient, any outperformance should quickly self-correct. But in reality, some evidence shows that winners keep winning—at least for a little while. Over longer intervals, however, you might see the opposite: strong performers revert to their mean, creating a “reversal” pattern.
You might’ve heard the terms “value stocks” (trading at low multiples relative to fundamentals) and “growth stocks” (high expected future growth, often accompanied by higher valuation multiples). Historically, “value” has sometimes delivered higher returns than “growth,” which is peculiar if we assume all relevant information is baked into the prices. Then again, it’s worth asking if value systematically entails higher risks that the market is compensating for.
One line of reasoning is that anomalies reflect higher risk exposures not captured by simple models. Maybe the small-cap premium (where smaller firms seem to have higher average returns) is just compensation for the extra liquidity and business risk you’re taking on. Under this reasoning, anomalies aren’t free lunches; they’re just under the radar of conventional metrics.
Another explanation points to human psychology. Behavioral finance suggests that we’re not perfectly rational—surprise, right? Overconfidence, herding, anchoring, and other biases can drive prices away from their fair value, at least temporarily. These biases can sustain anomalies if enough investors consistently act in ways that defy rational models.
Sometimes a market structure can enable persistent mispricings. For example, short-selling constraints may prevent arbitrageurs from quickly correcting overvalued securities. Similarly, certain taxes or regulatory constraints might distort investor behavior enough to let anomalies persist in a corner of the market.
Here’s an interesting tidbit from personal experience: During my graduate research days, I tried to replicate a well-publicized anomaly in emerging markets—only to find that after the study was published, the anomaly essentially vanished. And that’s the kicker: once enough investors are aware of an anomaly, their attempts to exploit it might erode or eliminate any excess returns.
Many academic papers show that anomalies tend to weaken over time, sometimes disappearing entirely. This phenomenon suggests that markets might learn. However, some anomalies (like momentum and value) still linger, possibly reflecting a true factor that commands a premium.
This happens when researchers test hundreds (or thousands) of strategies against the same dataset until something “magically” looks profitable. The risk is that the anomaly is nothing more than a statistical fluke—pure coincidence. Hence, rigorous testing with out-of-sample data, or using different time periods and markets, is important for verifying that an anomaly is real.
When analyzing historical data, we might only look at companies that still exist today—those that survived. We ignore the ones that went bust, got merged, or liquidated. This obviously skews the dataset toward “success stories,” inflating historical returns and making some anomalies look more convincing than they actually are.
We can generally bucket explanations for anomalies into two big categories:
In reality, both can be partly true. For instance, small-cap stocks do involve unique liquidity and business risks, but they might also get overlooked or neglected by large institutional investors, sustaining mispricing.
A big question is whether these anomalies are truly unique or simply capturing the same underlying factor. If momentum and value both point to the same chunk of risk, then combining them might not reduce portfolio risk the way you’d expect. Investors should dig deeper:
Sometimes, anomalies also cluster in certain market segments. For instance, the momentum effect might be stronger in technology stocks—potentially complicating a manager’s asset allocation choices.
Below is a simple Mermaid.js flowchart illustrating the typical process analysts might follow when researching a market anomaly:
flowchart LR
A["Identify Potential Anomaly <br/> (Observation or Theory)"] --> B["Collect Historical Data"]
B --> C["Statistical Tests <br/> (e.g., Regression, T-tests)"]
C --> D["Check Robustness: <br/> Different Samples <br/> Out-of-Sample"]
D --> E["Adjust Strategy for <br/> Transaction Costs & Taxes"]
E --> F["Implement Strategy <br/> or Reject Hypothesis"]
This flow typically concludes with either incorporating the anomaly into a trading strategy or deciding that it’s not feasible once all costs and constraints are considered.
Let’s say you want to do a small backtest of a “January Effect” strategy. Below is a hypothetical code snippet (simplified for demonstration) that checks the performance of small-cap stocks in January versus other months. Note: This is purely illustrative.
1import pandas as pd
2import numpy as np
3import yfinance as yf
4
5ticker = 'IWM' # Russell 2000 ETF
6data = yf.download(ticker, start='2000-01-01', end='2025-01-01')
7
8data['Month'] = data.index.month
9
10data['Ret'] = data['Close'].pct_change()
11
12jan_returns = data[data['Month'] == 1]['Ret'].mean() * 100
13other_returns = data[data['Month'] != 1]['Ret'].mean() * 100
14
15print(f"Avg January Return: {jan_returns:.2f}%")
16print(f"Avg Other Months Return: {other_returns:.2f}%")
In practice, you’d refine this code to run robust hypothesis tests, adjust for changes in index constituents, control for risk factors, and so on.
For exam scenarios (whether in the context of a constructed-response question or an item set vignette), you might be asked to:
When you see a question about “anomalies,” think: Are these truly unexplained alpha opportunities? Or are they proxies for higher risk or behavioral quirks? Incorporate the possibility that ongoing research or widespread knowledge can dilute an anomaly over time.
It’s smart to remember that, under the CFA Institute Code and Standards, you should always base recommendations on diligent research—so if you do find an anomaly and plan to exploit it, ensure robust due diligence and ethical considerations are met.
Important Notice: FinancialAnalystGuide.com provides supplemental CFA study materials, including mock exams, sample exam questions, and other practice resources to aid your exam preparation. These resources are not affiliated with or endorsed by the CFA Institute. CFA® and Chartered Financial Analyst® are registered trademarks owned exclusively by CFA Institute. Our content is independent, and we do not guarantee exam success. CFA Institute does not endorse, promote, or warrant the accuracy or quality of our products.