Cheap Is Only Half the Story: the Gross Profitability Premium

For twenty years after Fama and French, "value" was the workhorse of quantitative stock picking: buy companies that are cheap relative to their book value. It worked, but it had an ugly side effect. A portfolio of the cheapest stocks tends to be a portfolio of the most troubled companies, the ones the market has given up on for a reason.

In 2013, Robert Novy-Marx published a paper arguing that value investors had been ignoring the obvious other half of the picture. Instead of only asking what you pay, also ask what you get. And the best measure of what you get, he argued, is not the bottom line at all. It is a number most investors skate right past: gross profits.

The problem: the bottom line is a bad measure of a good business

If you want to find profitable companies, the instinct is to look at earnings, the famous "bottom line." Novy-Marx pointed out that by the time an accounting number has travelled all the way down the income statement to net income, it has been mangled.

Along the way it has absorbed:

• Research and development spending, which is an investment in the future but gets subtracted like a cost, so an innovative firm looks less profitable than a stagnant one.
• Advertising and sales spending, same problem: money spent building a brand makes this year's earnings look worse.
• Interest expense, which tells you about the company's financing choices, not the quality of its business.
• All manner of accounting discretion, write-offs, one-time charges, depreciation schedules, tax games.

Gross profit, by contrast, sits almost at the top: it is simply revenue minus the direct cost of the goods sold. It is the cleanest available answer to the question "does this company sell things for meaningfully more than it costs to make them?" Novy-Marx's insight is that this crude, un-mangled number is a better signal of true economic profitability than the polished figures further down.

His actual measure is gross profits divided by total assets: how much raw profit the company squeezes out of the asset base it owns.

The key idea via analogy: the price tag and the product

Imagine you are buying a used car. The value investor's approach is to look only at the price relative to the blue-book value: is this car cheap? That will indeed find you bargains, but it will also find you a lot of cars that are cheap because the engine is shot.

Novy-Marx's addition is embarrassingly simple: also look at the engine. Gross profitability is the engine check. Two cars at the same discount to blue book, one running smoothly, one wheezing: obviously prefer the one that runs.

His empirical finding is the striking part. On its own, gross profitability predicted the cross-section of returns roughly as strongly as book-to-market did. That is a bold claim: the most famous predictor in finance had an equally powerful, previously overlooked twin.

And here is the elegant bit. The two signals point in opposite directions on the same characteristics. Value screens push you toward unglamorous, struggling companies. Profitability screens push you toward strong, growing businesses, which usually trade at higher prices. Because the two strategies naturally lean opposite ways, they hedge each other: combining them produced a portfolio with a considerably better risk-adjusted result than either alone. Novy-Marx called profitability "the other side of value" for exactly this reason.

Why it mattered

• It rehabilitated growth. Some expensive companies are expensive because they are genuinely excellent. Screening on profitability lets a quant own those companies without abandoning discipline. The paper's phrase for this is "good growth."
• It gave quality a hard definition. Practitioners had been talking about "quality" for decades in fuzzy terms. Gross profits over assets is a single, computable, hard-to-fudge number, which is precisely what a systematic strategy needs.
• It fed straight into the standard models. Within two years, Fama and French had added a profitability factor to their model, and Hou, Xue and Zhang had built profitability into their q-factor model. Novy-Marx's paper is a direct ancestor of both.
• It made value strategies better. The practical advice, do not just buy cheap, buy cheap and profitable, became one of the most widely adopted refinements in systematic equity investing.

The honest limitations

• Gross profits ignore real costs. The whole appeal of the measure is that it ignores what comes below the gross line. But some of those costs are real. A company with enormous gross margins and catastrophic overhead is not a good business, and gross profitability alone will not tell you that.
• It varies wildly across industries. A software company and a supermarket have structurally different gross margins. Naive sorting will just load you up on certain sectors, so the signal usually needs industry adjustment to be meaningful.
• It has been crowded. The profitability and quality trade became extremely popular after 2013. Whether the premium available today matches the one measured in the historical sample is an open and reasonable question.
• The risk story is unclear. Why should profitable firms earn more? They seem safer, not riskier, which is awkward for a risk-based explanation. The valuation-identity argument that Fama and French later leaned on helps, but the economic story is less intuitive than "cheap stocks are distressed."

The one-line takeaway

Novy-Marx showed that gross profits, a number sitting near the top of the income statement before accountants get to it, predicts returns about as well as the famous value ratio, and because profitability and cheapness pull in opposite directions, using both together beats using either one alone.