Quality and Profitability: Return on Equity, Accruals, Leverage and the Pairing with Value

What you will learn

• Which measurable characteristics make up the quality factor and how each is calculated.
• Why the accruals ratio matters and how it exposes earnings that are not backed by cash.
• How to combine several measures into a standardised quality score.
• Why quality and value have historically offset each other's weaknesses.

Explanation

Quality is the least precisely defined of the major factors because it is a bundle of related characteristics rather than a single ratio. The common thread is that companies with high, stable, cash-backed profitability and conservative balance sheets have historically earned higher risk-adjusted returns than the opposite profile, even though a simple risk story would predict the reverse. Robert Novy-Marx (2013) showed that gross profitability, gross profit divided by total assets, predicted returns about as strongly as book-to-market. Fama and French added their profitability factor RMW in 2015. Asness, Frazzini and Pedersen's "Quality Minus Junk" (2019) combined profitability, growth and safety measures into one factor.

The individual measures you will meet most often are:

• Return on equity (ROE): net income divided by shareholders' equity (covered in the stock analysis course); here it is a ranking input, often averaged over three to five years to reduce noise.
• Gross profitability: (revenue − cost of goods sold) ÷ total assets. It sits high in the income statement, so it is less affected by accounting choices and one-off items than net income.
• Accruals ratio: (net income − cash flow from operations) ÷ total assets. Richard Sloan showed in 1996 that companies with high accruals, meaning profits that are not matched by cash, earn lower subsequent returns. A ratio near zero or negative means earnings are backed by cash; a large positive ratio warns that revenue is booked ahead of cash or costs are deferred.
• Leverage: usually total debt divided by equity, or debt divided by assets. Lower leverage scores better.
• Earnings variability: the standard deviation of earnings growth over several years; lower is better.

Index providers pick a subset; as of 2026 the MSCI Quality indices, for example, use ROE, debt-to-equity and earnings variability. Check the current methodology before relying on any list.

To combine measures, each is converted into a z-score: (company value − average across the universe) ÷ standard deviation across the universe, with the sign flipped for measures where lower is better. The z-scores are then averaged. This puts ratios with different units on one scale.

Why does quality pair well with value? Cheap stocks are cheap partly because many of them are low quality: weak margins, high debt, deteriorating cash flow. High-quality stocks tend to be expensive. The two rankings are therefore negatively correlated in the cross-section. Combining them does two useful things. First, it removes the worst value traps from the cheap basket: a company that is cheap with negative accruals, low debt and stable profits is less likely to be cheap for a fatal reason. Second, the returns of the two factors have historically been weakly or negatively correlated over time: quality tends to hold up in recessions while value often struggles there, and the reverse in early recoveries. Holding both has historically given a smoother path than either alone, an observation rather than a rule.

Worked example

Hypothetical example: two companies, figures in € million, invented and rounded.

• ROE: X = 120 ÷ 800 = 15.0%; Y = 90 ÷ 500 = 18.0%. On ROE alone, Y looks better.
• Accruals ratio: X = (120 − 150) ÷ 2,000 = −1.5%; Y = (90 − 40) ÷ 1,500 = +3.3%. X's profit is more than covered by cash; Y books €50m more profit than it collects.
• Debt-to-equity: X = 400 ÷ 800 = 0.50; Y = 900 ÷ 500 = 1.80.

Y's higher ROE is partly produced by leverage and is not backed by cash. An equal-weighted composite of the three measures ranks X above Y. The example ignores taxes, one-off items and industry differences; banks and utilities need separate treatment because their normal leverage is far higher.

Common mistake

Practice activity

Three hypothetical companies in one industry have accruals ratios of −2.0%, +1.0% and +4.0%. The average of the three is +1.0% and the standard deviation (population) is about 2.45 percentage points. Calculate the z-score of each company and flip the sign so that a higher score means better quality. Then state which company would rank highest on this measure.

Key takeaways

• Quality bundles profitability, cash backing of earnings, low leverage and earnings stability; gross profitability and accruals are the strongest research-backed inputs.
• The accruals ratio (net income minus operating cash flow, over assets) exposes profits that cash does not support.
• Z-scores put different ratios on one scale so they can be averaged.
• Quality and value are negatively correlated in rankings and weakly correlated in returns, which is why they are often combined.

Next step

The next lesson examines the size factor: whether small companies really earn a premium and why the answer depends on removing low-quality "junk".