Skip to main content

Defining Quality

Updated Jun 24, 2026 at 2:35pm

Research Draft High 1,114 words

"Quality" is the most intuitive yet least standardized of the major investment styles. The core idea is simple: some businesses are structurally better than others — more profitable, more stable, less indebted, better run — and an investor should rationally pay more for them. The tension is that there is no canonical formula. Asness, Frazzini and Pedersen frame quality as the bundle of characteristics that, all else equal, an investor should be willing to pay a higher price for — but every practitioner draws the bundle differently. Quality is therefore best understood not as a single metric but as a family of related signals (profitability, stability, low leverage, prudent capital allocation) that empirically tend to travel together and to predict superior risk-adjusted returns.

How quality is defined

There is no single accepted definition. Three influential frameworks anchor the field:

1. The academic "Quality Minus Junk" (QMJ) bundle — Asness, Frazzini & Pedersen (AQR, 2013/2019) decompose quality into four pillars, each from the Gordon-growth dividend logic:

  • Profitability — gross profits, ROE, ROA, cash flow over assets, low accruals (high earnings quality).
  • Growth — the trailing change in each profitability measure (NOT revenue growth alone).
  • Safety — low return volatility, low beta, low leverage, low bankruptcy risk (e.g. low Ohlson O-score), low earnings volatility.
  • Payout — the fraction of profits returned to shareholders; proxies management discipline / low dilution.

Each stock gets a z-score on every component; the components are averaged into a composite quality score.

2. The index-provider definition — the widely tracked MSCI Quality Indexes use just three descriptors: high Return on Equity, low Debt-to-Equity, and low Earnings Variability (the 5-year standard deviation of YoY EPS growth). Each is z-scored and averaged into a composite Quality Z-Score. This is the practical, replicable definition behind most "quality" ETFs.

3. The fundamental / Buffett-style definition — quality as a durable economic moat plus high returns on capital. The central metric is ROIC vs. WACC: a business creates value only when ROIC exceeds its cost of capital, and a firm that sustains high ROIC for many years usually possesses a moat (brand, network effects, switching costs, scale, IP). Practitioners often flag long-run ROIC above ~15% as indicative of structural advantage — a commonly cited rule of thumb, not a precise threshold.

The overlap across all three is the stable core: high, persistent profitability + low financial risk.

How it's used in practice

Quality is used in two distinct modes:

  • As a stand-alone screen / factor: rank a universe by a composite quality score and own the top decile/quintile, rebalancing periodically. This is the MSCI/ETF approach.
  • As a filter or overlay on other styles: quality's most documented practical value is combining it with value. Novy-Marx (2013) showed that controlling for profitability dramatically improves value strategies — cheap and profitable beats cheap alone, and the two signals are negatively correlated (value tilts toward struggling firms, quality toward thriving ones), so they diversify each other. Buffett's own returns have been substantially explained by exposure to quality + low-beta + leverage ("Buffett's Alpha", Frazzini-Kabiller-Pedersen).

Fundamental managers use quality differently: as a qualifying gate before valuation work — first confirm the business is durable and high-return (moat, ROIC > WACC, clean balance sheet), then assess price. This is the logic behind the sibling concepts "high-ROIC compounders" and "durable competitive advantage."

Adoption, debate & evidence

Quality is mainstream — it is one of the standard equity factors offered by MSCI, FTSE Russell, AQR and most factor-ETF families, and profitability became part of the Fama-French five-factor model (2015).

The evidence is genuinely strong but carries caveats:

  • The profitability premium (Novy-Marx) is robust: gross-profits-to-assets predicts the cross-section of returns about as strongly as book-to-market, and it survives across measures (ROE, operating, cash profitability) and internationally.
  • The QMJ factor earned positive risk-adjusted returns in the U.S. (data from 1956) and, in the global sample of 24 developed markets (from 1986), delivered positive returns in 23 of the 24 countries — the lone small negative being New Zealand, one of the smallest markets (Asness, Frazzini & Pedersen). QMJ also tends to gain in downturns, behaving like a hedge.
  • The honest controversy: Morningstar calls quality "the fuzziest of factors" precisely because the definition is not standardized — two "quality" funds can hold very different stocks, and backtest results are sensitive to which metrics you pick. Critics argue some of quality's apparent premium overlaps with profitability/low-volatility already captured elsewhere, and that a poorly specified quality screen can simply become an expensive-growth or low-volatility bet in disguise.
  • A structural tension noted by AQR themselves: high-quality stocks command only a modestly higher price than junk — a "puzzle." If quality is a hedge with low risk, theory says it should earn low returns; that it earns high risk-adjusted returns is the anomaly that makes the factor attractive but also fragile to being arbitraged away if the "price of quality" rises.

Strengths & limitations

When it works: quality earns its keep in drawdowns and uncertain regimes (its defensive, hedge-like profile), and as a complement to value. High-ROIC, low-debt firms compound through cycles and suffer fewer permanent impairments.

When it fails / the #1 misuse: the dominant error is conflating quality with price-insensitivity — paying any multiple for a great business. Quality describes the business, not the investment; a wonderful company at a terrible price is a bad trade. Quality also lags badly in sharp "junk rallies" (e.g. early-cycle recoveries, the late-1990s low-quality melt-up) when the riskiest, lowest-quality stocks lead. A second misuse is definition-shopping — selecting whichever quality metric backtests best (data-mining), then expecting it to persist out of sample.

Sources

Disputes flagged: (1) no standardized definition of quality — different providers/funds disagree on metrics; (2) debate over whether quality is a distinct premium or overlaps profitability/low-vol; (3) the "price of quality" puzzle means the premium could compress.