Why known characteristics are more important than growth guidance

In the first part of this series, we showed that guidance from Japanese management teams can be a reliable proxy for their firms’ future earnings growth. But at the same time, guidance does not seem to help investors predict returns.

Valuations and profitability, as measured by gross profit/assets (GP/A), our favorite measure of business quality, can predict returns. Let’s return to our 20-year dataset of Japanese companies with fiscal years ending in March. In Panel A of the below table, we regress one-year forward returns against EBIT guidance, book/market ratios, and GP/A ratios. After controlling for valuation and profitability, guidance adds almost zero predictive power in forecasting returns. The t-statistic for guidance is only 0.2, and the coefficient is practically zero. On the other hand, book/market and GP/A are highly reliable predictors of returns, with t-statistics of 22.3 and 12.4, respectively. The R-squared of the cross-sectional regression in Panel A is only 2.55%, highlighting the tremendous uncertainty involved in predicting the returns of individual stocks.

Figure 1: Regressions Predicting 1Y FWD Annual Returns (FY 2003 – FY 2023)

Source: S&P Capital IQ and Verdad analysis

Panel B of the above regression table repeats the same forecasting exercise but assumes clairvoyance by using realized EBIT growth as a predictor instead of guidance. As expected, perfect foresight into future earnings would provide reliable information toward predicting future returns. However, it’s interesting to note that the t-stat of 25 for perfect earnings forecasts is about the same in magnitude as book/market in Panel B. Moreover, while the R-squared would more than double to 5.93%, there is still tremendous uncertainty involved in forecasting individual stock returns, even after removing all surprises from future earnings growth.

Next, we can explore how valuations vary with guidance, after controlling for quality. Among firms with similar levels of profitability, one might reasonably expect to see higher valuations among stocks with higher expected earnings growth.

To control for profitability, we sorted all Japanese stocks into three categories by GP/A on an annual basis: high, medium, and low profitability. Separately, we ranked all Japanese firms into quintiles according to their EBIT guidance, also on an annual basis. Within each of the three profitability categories, we measured how valuations change as you move from the lowest guidance quintile to the highest.

The results of our analysis within the high GP/A category are shown below. Counterintuitively, price/book ratios do not rise linearly with increased guidance. Rather, valuation ratios follow an inverted U pattern as EBIT guidance increases from the lowest quintile to the highest. The simple explanation for this paradox is that the extreme quintiles contain cyclical companies that are at peak earnings (quintile 1) and trough earnings (quintile 5), as detailed in Appendix A. This is good news for investors because it means that it’s not necessary to pay up for high guidance. You can always buy relatively cheap companies at trough earnings that are on track to benefit from margin recovery by the end of the current fiscal year.

Figure 2: Variation in P/B and FWD 1Y Returns by Guidance Among High GP/A Firms

Source: S&P Capital IQ and Verdad analysis (FY 2003 – FY 2023)

The green circles in the figure above represent forward one-year returns. Within the top profitability category, returns increase linearly with higher guidance. As detailed in Appendix A, firms in the highest guidance quintile tend to be companies that experienced declining earnings last year, but a significant recovery is expected over the current fiscal year. Because these firms have lumpy earnings growth, they tend to trade at a discount relative to the consistent growers in quintiles 3 and 4. Therefore, quintile 5 in the figure above represents high-quality firms that are relatively cheap, where fundamentals are improving. Unsurprisingly, these firms earn the highest returns at 13.3% per year.

It should be noted that this linear return relationship only holds among high-profitability firms. As shown in Appendix A, firms with medium or low profitability have roughly the same returns across all guidance quintiles (10-12% return among medium GP/A firms, and 8-10% return among low GP/A firms). Therefore, for at least two-thirds of our dataset, guidance does not provide much information for predicting returns, once we have controlled for profitability.

Overall, the evidence suggests that investors shouldn’t take guidance as a starting point when seeking to maximize returns. Rather, it appears the best starting point is to rank stocks by their valuation and profitability characteristics in order to identify firms that are cheap relative to their productivity in turning capital into profits. Once we’ve selected for these stocks, guidance might help us with more cyclical stocks in particular—identifying companies at cyclical troughs and peaks where earnings in the future will be materially different than today.

Appendix A: Variation of Price/Book with Guidance Within Three Quality Categories

back to library

Archive

The Accuracy and Importance of Growth Guidance (Part II)