Using Financial Statement Signals to Test Value-Growth Expectation Errors

Summary

This review of Piotroski and So’s study examines whether value-growth return differences arise partly from investors’ mistaken expectations about future company performance. It describes FSCORE, a composite of nine binary signals covering profitability, changes in leverage and liquidity, and operating efficiency. Companies are grouped by this measure of improving or deteriorating fundamentals and by several valuation proxies, including book-to-market, earnings-to-price, cash-flow-to-price, sales growth, and turnover. The central prediction is that financial signals should be most informative when they conflict with the expectations implied by a company’s valuation category.

The reported evidence includes stronger subsequent earnings measures and stock returns for high-FSCORE firms, with different patterns inside value and growth groups. Return differences are concentrated where recent fundamentals conflict with valuation-implied expectations and weaken where the two agree. The review also describes supporting tests using earnings announcement returns, analyst forecast errors and revisions, and momentum conditioned on FSCORE. These findings are presented as support for mispricing and investor underreaction, while acknowledging risk-based explanations and other interpretations remain possible. The evidence comes from historical US company data and does not establish that the patterns will persist or survive implementation costs.

Key ideas

• FSCORE aggregates nine binary financial signals to classify companies by recent fundamental improvement or deterioration.
• The study tests whether FSCORE predicts returns differently within value and growth groups.
• Return spreads are reported as strongest when fundamentals conflict with expectations implied by valuation measures.
• Earnings announcements, analyst forecast errors and revisions, and conditional momentum provide additional tests.
• The authors interpret the patterns as evidence consistent with expectation errors, while alternative risk explanations remain possible.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.