Momentum and Quality
Momentum is the tendency for stocks that have performed well over recent months to keep performing well over the following months, and for recent losers to keep losing. It is among the most consistently documented patterns in financial economics, appearing across countries, asset classes, and time periods. It is also the factor Fama and French have persistently declined to include in their models, which makes it the most interesting disagreement in the field.
The original finding
Jegadeesh and Titman (1993) showed that ranking stocks on returns over the previous three to twelve months and buying the winners while shorting the losers produced significant profits over the subsequent months. The result was awkward for the efficient markets view in a particularly direct way: unlike value or size, momentum requires no accounting data at all. It uses only past prices, which is precisely what the weakest form of market efficiency says cannot predict returns.
The standard construction, usually labelled UMD for Up Minus Down or MOM, skips the most recent month when ranking. That detail is not cosmetic. Over very short horizons stocks tend to reverse rather than continue, so including the latest month contaminates the signal with the opposite effect.
Carhart (1997) added momentum to the three-factor model to produce the four-factor model widely used in mutual fund evaluation, where it matters because funds can appear skilled simply by holding recent winners.
Why it might exist
The explanations are mostly behavioural, which is a large part of why Fama and French exclude it. Investors may underreact to news, so prices adjust gradually rather than instantly and the drift is the remainder of an incomplete adjustment. Investors may also be slow to sell losers in order to avoid realising a loss, delaying the decline. And once a trend is established, trend-following capital can extend it beyond fundamentals.
Risk-based explanations exist but are generally regarded as less convincing. This is the crux of the disagreement: Fama and French take the view that a factor without a credible risk story does not belong in an asset pricing model, while many practitioners take the view that a pattern this robust across markets and eras belongs in the model regardless of whether the mechanism is settled.
Momentum crashes: the factor's defining risk is not gradual decay but sudden, severe reversal. These crashes have historically occurred when a falling market rebounds sharply, because the short leg consists of beaten-down stocks that rally hardest in a recovery. The return distribution is strongly negatively skewed, which means the average return substantially understates how bad the worst outcomes have been.
The implementation problem
Momentum requires frequent rebalancing by construction, since the set of recent winners changes continuously. High turnover makes the strategy unusually sensitive to transaction costs, and estimates of the net premium vary widely depending on the cost assumptions used. It is the clearest case in factor investing where a strategy that looks compelling gross of costs can look ordinary net of them.
The short leg adds a second difficulty. Borrowing beaten-down, heavily shorted stocks can be expensive or impossible, and borrow costs peak exactly when the trade is most crowded. Long-only momentum implementations avoid this but capture only part of the effect.
The quality factors: RMW and CMA
The 2015 five-factor extension added profitability (RMW, Robust Minus Weak) and investment (CMA, Conservative Minus Aggressive). Practitioners often combine them under the broader heading of quality, favouring companies with high and stable profitability, conservative capital deployment, low leverage, and consistent cash generation.
Novy-Marx (2013) made the influential case that gross profitability had explanatory power comparable to book-to-market, and observed that the two signals complement each other well precisely because they select opposite kinds of companies. A value screen alone tends toward cheap, troubled firms; adding a profitability screen filters many of them out. The five-factor article covers the construction of both in more detail.
How momentum and quality interact
Momentum and value have historically been negatively correlated, since a stock that has fallen enough to look cheap is usually not a recent winner. That negative correlation is the main argument for holding both: their disappointments have tended not to coincide. Quality sits somewhere between them, and combining signals into a single integrated sort generally works better than running separate sleeves that can take offsetting positions in the same stock.
The momentum series tracked on this dashboard lets you see where the current reading sits relative to its own history. As with every other series here, a low percentile describes recent realised returns and is not a forecast. Nothing on this page is investment advice.
Jegadeesh, N., & Titman, S. (1993). "Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency." Journal of Finance, 48(1), 65-91.
Carhart, M. M. (1997). "On Persistence in Mutual Fund Performance." Journal of Finance, 52(1), 57-82.
Novy-Marx, R. (2013). "The other side of value: The gross profitability premium." Journal of Financial Economics, 108(1), 1-28.