Is the small-cap size premium still valid, or has it disappeared in recent decades?

Question

AcadiFi · Accepted Answer

This is an excellent and nuanced question that the CFA curriculum handles carefully. The short answer is: the size premium is real but much weaker and more conditional than originally believed.

**Original Evidence (Banz, 1981):**
Rolf Banz documented that the smallest decile of NYSE stocks outperformed the largest decile by approximately 3-5% per year from 1936-1975. Fama and French incorporated size as the SMB (Small Minus Big) factor in their three-factor model.

**What's Happened Since:**

| Period | US Small-Cap Premium (annualized) |
|--------|----------------------------------|
| 1926-1980 | ~5.0% per year |
| 1981-2000 | ~1.5% per year |
| 2001-2025 | ~0.5% per year (barely significant) |

**Why the Premium Has Shrunk:**

1. **Publication effect** — Once the anomaly was documented, capital flowed into small-cap funds and ETFs, bidding up prices and reducing future returns
2. **Improved information** — Technology and the internet reduced the information asymmetry that gave small-caps their edge
3. **Survivorship bias** — Early studies may have underweighted small-cap failures (delistings), overstating returns
4. **Microstructure costs** — The theoretical premium is partially consumed by wider bid-ask spreads, lower liquidity, and higher transaction costs for small stocks

**The Conditional View:**
Research by Asness, Frazzini, and others shows that the size premium is much stronger when combined with quality screens — small-cap stocks that are also high-quality (profitable, low leverage) significantly outperform. 'Junk' small-caps drag down the overall small-cap return.

**Example:** Suppose you compare two portfolios over 2020-2025. Portfolio A holds the smallest 20% of stocks by market cap (including unprofitable speculative names). Portfolio B holds the smallest 20% but filters for positive earnings and low debt. Portfolio A underperforms the market by 1.2% annually, while Portfolio B outperforms by 3.8%. The 'quality-adjusted size premium' survives even when the raw size premium doesn't.

**For the Exam:** The CFA curriculum presents the size factor as part of multi-factor models (Fama-French) and expects you to know that:
- Small-cap outperformance was documented historically
- The premium has diminished but is debated, not dismissed
- Transaction costs and liquidity are significant implementation barriers
- The size effect interacts with other factors, especially quality

Explore more equity factor models in our CFA Level I course.

Is the small-cap size premium still valid, or has it disappeared in recent decades?

Master Level I with our CFA Course

Related Questions

Practice Questions