TL;DR
The Fama-French momentum factor (UMD) has delivered roughly 8% annualised in the US since 1927, the largest of any documented premium. But high turnover means real-world implementation captures perhaps half of that.
A premium hiding in plain sight
In 1993, Narasimhan Jegadeesh and Sheridan Titman published a paper that broke the efficient markets crowd. They showed that US stocks with the strongest returns over the past 3–12 months continued to outperform over the following 3–12 months, and the laggards continued to lag. The pattern was statistically overwhelming, present in every decade they tested, and present in international data too.
It was a problem for efficient-markets theory because momentum says past prices predict future prices — exactly what efficient markets says shouldn't happen. Mark Carhart added a momentum factor to the Fama-French model in 1997, and Ken French has published the UMD ("Up Minus Down") series in his data library ever since.
The empirical picture today: momentum is the largest and most pervasive factor premium in the data. It also has properties that make it the hardest factor to actually capture.
What the 95-year record looks like
UMD's annualised return in US data 1927–2023, from the Ken French library:
- Gross UMD premium: ~8% per year
- Best decade: 1990s, roughly +13% per year
- Worst stretch: March 2009, when the factor crashed about 80% in a single quarter as junk stocks rebounded from the financial crisis
The premium is large enough that even after implementation costs, momentum-tilted funds tend to outperform the market by 1–2% net. The 2009 crash is the famous warning: pure momentum strategies can experience catastrophic reversals when the market regime flips violently.
International evidence is similar. Asness, Moskowitz and Pedersen's 2013 paper "Value and Momentum Everywhere" documented momentum premiums of roughly 5–10% across 18 country markets and 4 asset classes — equities, bonds, currencies and commodities. The factor isn't a US artifact.
Why implementation costs are so heavy
Momentum's premium is real but expensive to capture. Three reasons:
- Turnover. A pure academic momentum portfolio rebalances roughly every 3 months, with 100%+ annual turnover. Trading 100% of your portfolio yearly through bid-ask spreads, market impact and brokerage runs ~1–2% in costs even at institutional scale.
- Tax friction (outside tax wrappers). High turnover means short-term gains and constant tax events. UK GIA and US taxable accounts get hammered.
- Capacity. Momentum strategies are limited in how much capital they can absorb before their own trading moves prices. Large funds typically dilute their momentum exposure to manage this.
Practical ETFs handle this by slowing rebalancing — iShares MTUM rebalances semi-annually, not monthly, which gives up some of the premium but cuts costs roughly in half. The net is that real-world momentum captures perhaps 3–5% of the academic 8%.
How to use it in a FIRE plan
Three sensible approaches:
- Single-factor momentum ETF. iShares MSCI USA Momentum Factor (MTUM) is the cleanest US implementation at 0.15%. For UK investors, iShares Edge MSCI World Momentum Factor (IWMO) at 0.30%. Allocate 10–20% of equity to a single-factor tilt.
- Multi-factor fund with momentum built in. Avantis AVUV/AVDV, Dimensional Core funds, JP Morgan JPGL — these blend momentum with value and profitability so the factor's drawdowns get diversified away. Usually the better choice for FIRE planners.
- Skip it. Momentum is the most volatile factor and the hardest to hold through reversals. If you'd panic-sell after a 2009-style crash, momentum is the wrong tilt.
The case for combining momentum with value is especially strong. The two factors are nearly uncorrelated — momentum chases winners while value buys losers — so they smooth each other out. See our combining factors article for why this works structurally.
The honest baseline
Momentum has the strongest empirical pedigree of any factor and the worst behaviour during crashes. If you tilt toward it, do so through an integrated multi-factor fund that includes momentum alongside value, profitability and quality. The diversification is what makes the strategy actually holdable.
Test what a momentum-inclusive tilt does to your FIRE date in our factor comparison tool — toggle between presets and see how the FI-date distribution shifts.
Frequently asked questions
- Is momentum better than value?
- Historically momentum has a larger gross premium than value, but it costs more to implement and behaves worse in crashes. They tend to work in different periods, which is why combining them is so attractive.
- How often does a momentum portfolio rebalance?
- Pure momentum strategies often turn over 100%+ of the portfolio per year. Practical ETFs (like iShares MTUM) use slower rebalancing to keep costs manageable.
- Did momentum survive the 2009 crash?
- The factor recovered within a couple of years, and the 95-year track record absorbed the loss without breaking. But investors who panicked at the bottom and sold their momentum exposure permanently locked in the loss.
Stress-test your own FIRE plan
FIRE Wealth OS runs your savings rate and expenses against every historical market starting point since 1871. Free to use, no card required.