TL;DR
RMW (Robust Minus Weak) goes long firms with high gross profitability and short firms with low profitability. The factor has delivered roughly 3% annualised in US data 1963–2023.
The factor that closed the value-trap loophole
In 2013, Robert Novy-Marx published one of the most influential papers in modern asset pricing: "The Other Side of Value: The Gross Profitability Premium." He showed that a simple accounting ratio — gross profits divided by total assets — predicted future stock returns nearly as strongly as the value factor itself, and was almost entirely independent of value.
Translated: among stocks that look cheap, the ones with high gross profitability outperformed the ones with low profitability. Cheap-and-junky stocks (value traps) were the drag. Cheap-and-profitable stocks delivered most of the value premium.
Two years later, Fama and French added a profitability factor to their five-factor model. They called it RMW — Robust Minus Weak — and defined it as the return spread between high-profitability firms and low-profitability firms, sector-neutral.
Why gross profits, not earnings
A subtle but important detail: RMW uses gross profits (revenue minus cost of goods sold), not net earnings or operating income. The reason is that gross profits are harder to manipulate and harder to distort with accounting discretion.
Net earnings can be inflated by:
- One-time gains from asset sales
- Tax timing differences
- Aggressive depreciation assumptions
- Capitalised expenses that should have been expensed
Operating income suffers similar issues. Gross profits — the most basic measure of "is this business actually selling things for more than they cost?" — is the cleanest signal. Novy-Marx's original paper demonstrated that gross profits had stronger predictive power than any of the alternatives.
The numbers
Annualised RMW premium from the Ken French US Data Library, 1963–2023:
- High-profitability portfolio: ~12% nominal CAGR
- Low-profitability portfolio: ~9% nominal CAGR
- RMW spread (long-short): ~3% per year
International evidence is consistent. Fama and French's 2017 extension showed RMW premiums of roughly 2–4% per year across Europe, Japan and Asia-Pacific developed markets, with t-statistics above 2 in every region.
The factor's behaviour is also more benign than HML or SMB. RMW has lower volatility than the value or size factors, and it tends to win during downturns when junk gets punished. The factor's drawdowns are real but shorter than value's worst stretches.
Why RMW pairs perfectly with value
The killer combination in factor investing is value + profitability. Two reasons:
- They're nearly uncorrelated. The HML-RMW correlation is roughly -0.1 over 60 years. They share almost no overlap in which stocks they pick.
- Profitability fixes the value-trap problem. A pure value strategy picks up some genuinely distressed firms — companies that are cheap because they're failing. Filtering for profitability removes the worst of those.
The intersection — cheap AND profitable — captures most of the long-run premium that both factors offer individually, with materially smoother returns. Joel Greenblatt's Magic Formula (2006) was an early popular formulation. Avantis and Dimensional both build this combination into their core multi-factor funds.
For the broader picture of how profitability relates to the wider "quality" concept, see our quality factor article.
The Buffett connection
In 2018, AQR's Andrea Frazzini, David Kabiller and Lasse Pedersen published "Buffett's Alpha", decomposing Berkshire Hathaway's 60-year track record. They found that Buffett's outperformance was fully explained by four factor exposures: value, profitability (RMW), low-volatility (BAB), and modest leverage.
The "secret sauce" Berkshire had been quietly executing for decades was buying profitable, cheap, low-vol businesses with modest leverage — exactly what modern multi-factor ETFs now package for 25 basis points per year. See our Buffett factor article for the full decomposition.
How to actually own RMW
There's no major retail ETF that isolates RMW as a pure single factor. The practical alternatives:
- Quality-tilted ETFs: iShares QUAL/IWQU, Invesco SPHQ, JP Morgan JQUA. These use multi-signal quality definitions that include profitability prominently.
- Multi-factor integrated funds: Avantis AVUV/AVGE, Dimensional DFSV/DFEM, JP Morgan JPGL, iShares IFSW. These all screen for profitability as part of their broader factor mix.
- Value-and-quality combinations: a 50/50 split of a value ETF and a quality ETF captures most of the cheap-and-profitable intersection.
For most FIRE planners, the cleanest implementation is a single integrated multi-factor fund that handles the RMW screen internally. The factor is too important to skip and too operationally fiddly to isolate manually.
Test what a profitability-aware multi-factor tilt does to your FIRE date in our factor comparison tool — the median date typically moves 1–2 years earlier vs a market-only portfolio.
Frequently asked questions
- Why gross profitability and not earnings?
- Earnings are noisy and easier to manipulate. Gross profits — revenue minus cost of goods sold — is the cleanest measure of how good the underlying business is at making money.
- How does RMW relate to the quality factor?
- RMW is a key component of most quality factor definitions. See our [quality factor article](/blog/quality-factor-investing) for the wider picture.
- Has RMW worked since publication?
- Yes — better than most factors. Out-of-sample data since 2013 has shown RMW continuing to deliver positive returns, with less post-publication decay than value or momentum.
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.