Readability and Textual Complexity Measures
Readability metrics score how hard a document is to read using surface features like sentence length and syllable counts, and in finance a rising Fog Index in a company's filings has been linked to worse future performance.
Two 10-K filings can convey the same facts while being wildly different to actually read — one written in short, plain sentences, the other in long sentences stuffed with dense multisyllabic jargon. Readability metrics like the Fog Index and Flesch-Kincaid score quantify this using purely surface-level features: average sentence length and the share of "complex" words (usually defined as three or more syllables), with no attempt to judge the underlying content.
Readability scores measure how hard a document is to parse using sentence length and word complexity alone, and in finance research, filings that get harder to read over time are associated with weaker and less persistent future earnings.
The finance-specific finding is not just academic curiosity: companies with lower current earnings tend to write more complex, harder-to-read annual reports, and that complexity itself predicts weaker earnings persistence going forward — consistent with managers using dense language to obscure bad news rather than highlight it.
Worked example. The Fog Index formula is 0.4 x (average words per sentence + percentage of complex words). A filing with an average sentence length of 20 words and 30% complex words scores 0.4 x (20 + 30) = 20, roughly the reading difficulty of an academic journal article — well beyond the general public's comfortable reading level.
Quant researchers track the year-over-year change in a company's Fog Index more than its absolute level, since an unexplained jump in complexity right around a weak quarter is a stronger signal than a filing that has always been dense.
Practice in interviews
Further reading
- Li, 'Annual Report Readability, Current Earnings, and Earnings Persistence' (Journal of Accounting and Economics, 2008)