A reader asked: “Have you ever investigated: (1) the work of R. G. Sloan on the predictive power of accruals (both negative and positive); and, (2) the web site www.valuedog.com? It’s not kept up-to-date anymore, but the ‘earnings torpedo’ and ‘value screen’ parts of the web site are still maintained on a monthly basis, complete with charts. It claims to be ‘based on work at the University of Michigan Business School.’ They used to publish a monthly list of companies with positive or negative accrual attributes, but don’t anymore. I can’t find anyone else that does.”
There is a stream of academic research on the relationship between accruals and stock returns. See, for example (especially the second one):
“Gaming the Earnings/Accruals Gamers?”
The papers these items summarize mostly follow/extend those of Richard G. Sloan, 35 of which you can find via an author search at the Social Science Research Network (SSRN). A subject search there may locate other relevant studies.
ValueDog “is devoted to identifying undervalued stocks through the combination of large-sample stock screens and careful financial analysis of individual companies. It is based on the work of faculty, PhD students and MBA students at the University of Michigan Business School, all of whom will deny all knowledge of this site.” There are some return summaries for hedge portfolios of a few stocks spanning 1998-2005. The returns look interestingly large, but there is not enough information there for verification. Nor is there any discussion of trading frictions.
Earnings Torpedo offers updated lists and past performances of “Earnings Torpedo Warning” stocks and “The Value 40 Fund” stocks. The purpose of the site is to “provide an experimental learning opportunity for students studying equity security analysis at the Ross School of Business at the University of Michigan.” This site is current and presents:
Monthly lists of 100 Earnings Torpedoes, “stocks that can blow holes in portfolios,” for 2001-2009. It appears that this approach, as presented, may work most of the time but may fail at the beginning of bull markets.
Monthly Compilations of the Value 40 Fund, “formed from analyzing Value, Momentum, Quality, Predictability and other factors,” for 2005-2009. It appears that this approach, as presented, may work most of the time but may fail at the onset of bear markets.
There is apparently no discussion of trading friction assumptions (transaction fees, bid-ask spreads, shorting costs) for either of these lists. With monthly rebalancing of fairly large portfolios (potentially including relatively illiquid stocks), it may be that trading frictions materially degrade the outcomes presented for these two investing strategies. It would be difficult for most individual investors to establish economical positions (very large with respect to transaction fees) for so many stocks.
For example:
The 101-month record for the Earnings Torpedoes indicates a cumulative return since inception (from shorting) of 31.4%, which translates to a compounded monthly return of about 0.3%. Monthly trading frictions could easily exceed 0.3% of the portfolio.
The 57-month record for the Value 40 Fund indicates a cumulative return since inception of 143.8%, which translates to a compounded monthly return of about 1.6%, offering a considerable margin against trading frictions.
