Evidence-based investing research
Value Investing Strategy (Strategy Overview)
Allocations for November 2026 (Final)
Cash TLT LQD SPY
Momentum Investing Strategy (Strategy Overview)
Allocations for November 2026 (Final)
1st ETF 2nd ETF 3rd ETF

Profitability and Investment Factor Premiums Sample-specific?

Steve LeCompte | | Posted in: Equity Premium

Key Insight

RMW shows weaker performance in pre-discovery data (0.41% vs 0.52% 3-factor alpha; July 1940-June 1963 vs post-July 1963), while CMA shows insignificant alpha (-0.02%; July 1940-June 1963; hand-collected Moody's data).

Are the profitability (Robust Minus Weak, RMW) and investment (Conservative Minus Aggressive, CMA) stock factor premiums observed post-1963 robust in earlier data? In the February 2017 version of his paper entitled “The Profitability and Investment Premium: Pre-1963 Evidence”, Sunil Wahal measures the profitability (revenue minus cost of goods sold, scaled by total assets) and investment (change in total assets) factor premiums using data hand-collected from Moody’s Manual that predates the sample commencing July 1963 used to discover these factors. He defines the premiums conventionally as returns to a portfolio that is each month long (short) stocks with relatively high (low) expected returns. Because of missing older data, inconsistent accounting methods and accounting data availability lag (at least six months between fiscal year end and returns), he starts his accounting data in 1938 and return data in July 1940. Using available firm accounting and monthly return data as specified through June 1963, he finds that:

  • RMW has gross monthly average return 0.15% and 3-factor (market, size, book-to-market) alpha 0.41% during July 1940-June 1963, compared to 0.52% alpha reported in other research for data since July 1963. Findings generally hold:
    • For a subsample of large stocks.
    • After controlling for the value factor. Double sorts on book-to-market ratio and profitability into fifths (quintiles) show that the gross monthly return spread between high and low profitability quintiles within book-to-market ratio quintiles is around 1%.
    • For operating profitability (which excludes selling, general and administrative expenses and interest expense) scaled by book equity.
  • CMA has gross monthly average return -0.07% and insignificant gross monthly 3-factor alpha -0.02% during July 1940-June 1963. Alpha remains insignificant when:
    • Defining the factor based on change in book equity rather than change in total assets.
    • Extending the new sample backward to July 1926.

In summary, evidence indicates that profitability (asset growth, or investment) is robust (is not robust) on a gross basis as a stock-picking factor in U.S. samples that precedes factor discovery samples.

Cautions regarding findings include:

  • Reported returns and alphas are gross, not net. Accounting for portfolio reformation and shorting costs would reduce performance. Shorting may not always be feasible as specified.
  • Gross average monthly return for the profitability factor is an unattractive 0.15%. In other words, a standalone profitability screen does not perform well.
  • Firm accounting data may not have been as accessible to investors as later data, such that stock prices may not reflect their widespread use.

Frequently Asked Questions

What are the profitability (RMW) and investment (CMA) factors in stock investing?

The profitability factor (Robust Minus Weak, RMW) is the return to a portfolio long stocks with high profitability and short stocks with low profitability, with profitability measured as revenue minus cost of goods sold, scaled by total assets. The investment factor (Conservative Minus Aggressive, CMA) is the return to a portfolio long stocks with low asset growth and short stocks with high asset growth, with investment measured as change in total assets. Both are defined conventionally as long stocks with relatively high expected returns and short stocks with relatively low expected returns.

Do the profitability and investment factor premiums hold up in stock market data before 1963?

The profitability factor premium (RMW) holds up in pre-1963 data, but the investment factor premium (CMA) does not. During July 1940-June 1963, RMW had a gross monthly 3-factor alpha of 0.41%, versus 0.52% reported for data since July 1963. CMA had an insignificant gross monthly 3-factor alpha of -0.02% over the same period, and it stayed insignificant when extended back to July 1926.

How are the profitability and investment factor premiums tested in older stock data?

The premiums were tested using accounting data hand-collected from Moody's Manual, with accounting data starting in 1938 and return data starting in July 1940 to allow for missing data, inconsistent accounting methods and a reporting lag of at least six months. Each month, portfolios go long stocks with high expected returns and short those with low expected returns, and performance is measured by gross average return and 3-factor (market, size, book-to-market) alpha through June 2063's predecessor date, June 1963. Robustness checks included a large-stock subsample, controlling for the value factor, double sorts on book-to-market and profitability into quintiles, and operating profitability scaled by book equity.

How does the pre-1963 profitability premium compare to the post-1963 premium found in other research?

The pre-1963 profitability premium is somewhat weaker than the post-1963 premium. The 3-factor alpha for RMW was 0.41% per month during July 1940-June 1963, compared with 0.52% reported in other research for data since July 1963. Double sorts on book-to-market and profitability quintiles also show a gross monthly return spread of around 1% between high and low profitability quintiles within book-to-market quintiles.

What are the limitations of using profitability as a stock-picking factor?

Reported profitability factor returns and alphas are gross, so accounting for portfolio reformation and shorting costs would reduce performance, and shorting may not always be feasible as specified. The gross average monthly return for RMW is only 0.15%, so a standalone profitability screen does not perform well. Additionally, firm accounting data may not have been as accessible to investors in earlier periods, so stock prices may not reflect its widespread use.