The Piotroski Synthesis
A war for assets and alpha has pitted the old-school bottoms-up stock pickers against the avant garde of indexers and quants. This competition is, however, the tip of the iceberg of a deeper intellectual debate.
On the one hand, the indexers and quants allege that the evidence of the failures of the stock picking experts is insurmountable. The sell-side forecasts and the buy-side models all fail the test of predictive accuracy. The quants prefer to rely on simple quantitative signals: Is the stock’s valuation in the bottom decile of all companies of similar size? Does the share price movement reflect positive momentum?
On the other hand, the stock pickers are boggled by the fact that the quants ignore so much relevant information. The stock pickers are crackerjack cash flow analysts, impresarios of the income statement. Surely, they demand, the process of financial analysis must add value. What about quality, financial health, creditworthiness, industry volatility?
But, from our alma mater and Hegelian hotbed, Stanford University, a new synthesis of this duel between the stock pickers’ thesis and the quants’ antithesis has emerged. Finance professor Joseph Piotroski believes that the best strategy for value investing is to combine the mathematical metrics of value, beloved of the quants, with an analysis of the strength of companies’ financials, drawn from the methodology of the stock pickers.
Piotroski argues that investors must compare the expectations implied by pricing multiples against the strength of firms’ fundamentals. “Such a comparison,” he notes, “is the central premise behind security analysis, as discussed by Graham and Dodd, where sophisticated investors use historical financial information to select profitable investment opportunities.”
From this perspective, investors focus too much on extrapolating income statement trends and too little on cash flow and balance sheet metrics that indicate financial stability. According to this theory, the stock market can be divided on the x-axis into expensive glamor stocks and cheap value stocks, and on the y-axis between companies with weak and strong fundamentals.
Figure 1: The Piotroski Synthesis in Table Form
Source: Piotroski and So, “Identifying Expectation Errors in Value/Glamour Strategies: A Fundamental Analysis Approach.”
Piotroski measures quality objectively. He takes nine metrics of financial strength and scores each company on whether the financials pass the test (e.g., Does the company have positive cash flow? Is asset turnover improving? Has the company paid down debt?). He rolls these nine binary scores into a combined measure he calls the F-Score.
A typical value strategy would go long on the value stocks and short the glamor stocks. But Piotroski argues that going long on cheap stocks with strong fundamentals and shorting expensive stocks with weak fundamentals (he calls this an incongruent portfolio) should outperform this naïve value strategy.
The market efficiently prices many stocks, assigning low prices to companies with weak fundamentals and high prices to companies with strong fundamentals, and these are not profitable arbitrage opportunities.
Looking at US markets from 1972 to 2010, Piotroski finds that annual returns to the incongruent strategy are much larger than the naïve value/glamor strategy, with an average annual return of 20.8% versus 10.5% for naïve value. And a congruent value/glamor strategy yields an average annual return of -1.9%.
Figure 2: Annualized Strategy Returns 1972–2010
Source: Piotroski and So, “Identifying Expectation Errors in Value/Glamour Strategies: A Fundamental Analysis Approach.”
The incongruent portfolios reveal in hindsight a consistent pattern of expectation errors, with positive earnings surprises and revisions driving the revaluation of the fundamentally healthy but cheap companies and the inverse for the fundamentally weak but expensive companies.
Verdad’s strategy builds on this fusion approach. We use three of Piotroski’s tests of fundamental health—improvements in asset turnover, reductions in long-term debt, and positive cash from operations—and marry those fundamental metrics with our cheapness and leverage metrics. We then use further fundamental analysis to screen out firms that aren’t credit worthy or have other problems or weaknesesses, attempting to maximize the incongruence between the cheapness of the stock and the improving fundamental health of businesses.
We believe these mismatches are most pronounced, and the subsequent corrections most dramatic, in leveraged stocks. Firms that are priced for bankruptcy, but whose financials show little probability of such an occurrence, can have dramatic revaluations once the process of deleveraging kicks into gear.
To demonstrate the effectiveness of this approach, we show below backtests of a naïve leveraged value strategy versus Verdad’s leveraged value strategy in the United States.
Figure 3: Backtest of Naïve Leveraged Value Strategy vs Verdad’s Strategy in the United States
Source: Portfolio123
If we conduct a similar analysis in Japan, we likewise see marked improvements in results when we compare the performance of high quality cheap stocks to a naïve value strategy. However, the gap between high quality and low quality businesses is less pronounced in Japan because the rarity of bankruptcies.
Figure 4: Backtest of Naïve Leveraged Value Strategy vs Verdad’s Strategy in Japan
Source: CapitalIQ; Portfolio of 50 stocks selected June of each year; average annualized returns in JPY
Our belief is that the best approach is to begin with quantitative work to define a reference class of potential securities that combine value with quality, and then to use additional fundamental work to remove low-quality securities where the valuation is congruent with the financial health of the firm. We call this the Piotroski Synthesis, and we believe it represents the best foundation for successful value investing that fuses quantitative and fundamental analysis.