MICHEL PIREU: Points system helps sift the chaff from value stocks
Buying the highest-scoring stocks and shorting those with the worst scores would have resulted in 23% annualised gains, Harry Domash finds
The likes of Warren Buffett and David Dreman have for years been encouraging a value approach to investing.
“The market is a pendulum that forever swings between unsustainable optimism (which makes stocks too expensive) and unjustified pessimism (which makes them too cheap),” wrote Jason Zweig in The Wall Street Journal’s The Intelligent Investor. “The Intelligent Investor is a realist who sells to optimists and buys from pessimists.”
In 1992, a landmark study by University of Chicago professors Eugene Fama and Kenneth French that looked at stock returns from 1963 to 1990 based on book-to-market and earnings-to-price ratios, concluded that the stocks with highest book-to-market ratio as well as those with the highest earnings-to-price (both value stocks) significantly outperformed the market.
In 2002, Joseph Piotroski, an accounting professor at the University of Chicago, took another look at the study and found that although the value portfolio outperformed, a few outperformers skewed the returns. In fact, more value stocks underperformed the market than outperformed.
As Harry Domash explains in his book Fire your Stock Analyst!, “Piotroski wondered about the relevance of a strategy that relies on the strong performance of a few firms while tolerating the poor performance of many deteriorating companies. For instance, say five stocks out of 100 account for the outperformance. What are the chances of picking one of those outperformers if you buy only 10 or 15 stocks?
“Piotroski reasoned that since value stocks got that way because something went wrong, many were financially distressed and might not survive. Piotroski wondered if you could boost the performance of the value portfolio by getting rid of the weakest players. To find out, he devised a simple nine-step test, using financial statement factors to evaluate financial strength.”
Each step posed a question and awarded one point if the company passed. For instance, did the company earn money last year? Give it one point for a yes and zero for a no. The final score is calculated based on nine criteria divided into three groups: profitability, leverage and liquidity.
The logic behind Piotroski’s original nine tests:
Profitability. Profits are the key to financial strength; without profits most companies will eventually fail. Tests one and two determine if the company is profitable based on both net income and operating cash flow. Award one point if the after-tax income is a positive number. Award one point if the operating cash flow is positive.
Tests three and four gauge the quality of the reported profits. Return on assets measures management’s effectiveness in converting available resources into profits. Award one point if return on assets increases year on year. Operating cash flow typically exceeds net income since depreciation and other non-cash items reduce income but not cash flow. Low cash flow compared to net income means non-cash accounting entries may be inflating income. Award one point if the operating cash flow exceeds net income.
Debt and capital. Is the company sinking deeper in debt or digging its way out? Tests five and six award points for declining debt levels. Test seven penalises companies that raise cash by selling more stocks. Total liabilities to total assets measure debt load. Increasing debt level isn’t necessarily a bad thing for strong companies, but since Piotroski was dealing mainly with financially distressed firms he looked for shrinking debt.
Award one point if the percentage increase in total assets exceeds the percentage increase in liabilities. Working capital shows the funds available to run the business. Piotroski wanted to see an improvement in working capital. Award one point if that’s the case. If the debt situation is improving, is it due to profitable operations or is the company raising cash by selling more stock? Piotroski penalised companies that increased the number of shares outstanding during the year, figuring that they sold stock to raise cash or make an acquisition. Award one point if the shares outstanding did not increase during the year.
Operating efficiency. Tests eight and nine in effect take the company’s operational pulse. Deteriorating gross margins often warn of a deteriorating competitive position. Award one point if gross margin for the trailing 12 months is higher than the year-ago figure. Asset turnover (revenues divided by total assets) is a standard productivity measure. Award one point if the percentage sales increase exceeds the percentage increase in assets.
Summing up the points provides the Piotroski F-Score (a number between zero and nine). Piotroski classified companies scoring below five as financially weak. Those that scored five and above were strong. He compared the performance of a portfolio limited to strong firms to a portfolio of all value-priced stocks. He found that the strong firms outperformed the all-stock portfolio by 7.5% annually.
Furthermore, he found that buying the top stocks in the market and shorting those that got the worst scores would have resulted in 23% annualised gains, more than double the S&P 500 broad market index return. Most significant, he found that weak (low scoring) firms were five times more likely to delist for performance-related reasons than strong firms. Obviously, not all companies with low scores will end up in financial trouble, but Domash found that all of the troubled companies that he looked at had low scores.
Is the Piotroski F-Score relevant in the SA context? In a paper published in 2013, Nicholas Pullen, then at the University of Cape Town, claimed that a modified high F-Score investment strategy is able to outperform both the market and a low F-Score portfolio over the medium and long term. “It is possible,” he concluded, “to use accounting-based information to construct a portfolio which is able to shift an investor’s distribution of returns and thereby generate positive abnormal returns within the SA context.”
At the time of writing, based on the last set of full-year results and excluding some 20-odd companies to which the test doesn’t apply, there are 128 companies listed on the JSE that Piotroski would consider strong (with an F-Score of five or more) and 210 that he would consider weak (with a score of less than five). However, only six JSE-listed companies achieve a score of eight or nine, while 79 have a score of less than three.
Piotroski made clear that his paper did not “purport to find the optimal set of financial ratios for evaluating the performance prospects of individual ‘value’ firms”. Rather, “it is just one way investors can use relevant historical information to eliminate firms with poor future prospects from a generic value portfolio”.
He also stated that the benefits to financial statement analysis are concentrated in small and medium-sized firms, companies with low share turnover and firms with no analyst following. The reason for this is presumed to be the information gap — there is less likely to be a genuine mispricing for large companies that are covered by a lot of analysts.