Guest contributor – Jean-Paul Jaegers, CFA, CQF (Senior Investment Strategist, Prudential Portfolio Management Group)
… however, what is often less debated is the impact that time can have on these variables.
This is relevant as investors’ holding periods have generally shortened and as a result the variables that influence the return for today’s investors are somewhat different from those in the past.
As an asset allocator I believe that in the long run, valuation is the key driver of equity returns. That’s not to say we don’t need to think about other influences and even try to understand what might be driving other investors’ views – for example, we find that investor sentiment can be helpful in market timing as there can be excessive herding by investors. However, investor sentiment has a rather short informational value (a few weeks or months, not years). In the long run, valuation is the key driver of equity returns.
We looked at how important equity valuations are for various holding periods. For this we performed a pooled panel regression covering US, UK, Eurozone and Japan data from 1980 to date using a cycle-adjusted price earnings (CAPE or Shiller P/E) valuation metric for expected returns (we used country dummies as there are level differences across geographies, for example the UK equity index has tended to trade at a valuation discount to the US equity index.
We looked at the R-squared to see how well this valuation measure explains the subsequent returns. The larger the R-squared, the better the fit and the more it explains of the future returns. This has been re-estimated over a range of horizons as shown in the following chart. At periods of less than two years the explanatory power is relatively limited (it explains less than around 20% of the returns, essentially ‘noise’). But over longer horizons, valuation becomes the most important driver and starts to explain a significant part of the returns. For example, on a five to six-year horizon the analysis suggests we can explain about half of the future returns. This highlights the major impact that valuation can have on future returns but also crucially the discipline that is required to take advantage of its predictive power. It seems a shame that with such a powerful predictor of future returns, many investors are instead moving towards shorter-term investment horizons which have, in my view, much less predictable drivers of return.
Since there are numerous valuation measures and techniques, how do we select the best measures to use? Here we assess the historical success for a range of measures including CAPE, forward P/E or trailing P/E, together with two bottom-up measures reflecting the dispersion between expensive stocks and cheap stocks in an index, plus more conventional ratios such as price-to-book or dividend yield. We have also included a relative valuation measure (represented by the equity-risk-premium) and macroeconomic measures such as interest rates or GDP. In general we see that most valuation methods are reasonably successful in forecasting future returns in the US with even the least successful, the trailing P/E, explaining nearly 35% of returns over a five to seven-year period. In addition, we notice that bottom-up dispersion measures have tended to do quite well. This follows the literature1 which found that a dispersion measure picks up aggregate valuation as pockets in an index might be bid up (thus larger dispersion), whereas in bear regimes most stocks tend to become cheap resulting in a low dispersion. This dispersion may not be detected when we look at an ‘average’ level for an index as they might cancel out in aggregation. In contrast, macroeconomic measures are much less successful at forecasting long-run future returns – GDP even over seven years only explains 20%, which may be a surprise to some given the effort and debate that some investors put into forecasting the macroeconomic outlook.
The following chart illustrates the results for a range of time series using data from the US.
There is one clear predictor of future returns, valuation. Valuation has tended to explain over 50% of future returns. To work successfully though it needs time, ideally at least five to seven years. Unfortunately many investors simply do not have this time horizon and instead choose to invest over a shorter period. This is a shame because in our view there aren’t any good predictors of future returns over short periods and as a result returns are likely to be much less predictable and much more random.
1 The Second Moment Matters! Cross-Sectional Dispersion of Firm Valuations and Expected Returns, Danling Jiang, Journal of Banking and Finance, 24 July 2013
This material is for distribution to Advisers and should not be relied upon by any other persons. This content is prepared for information and does not contain or constitute investment advice. Any personal views expressed are the author’s own and are subject to change without notice. Neither PPMG, nor any of its associates, nor any director, or employee accepts any liability for any loss arising directly or indirectly from any use of this document. The value of investments and any income from them may go down as well as up and are not guaranteed. Investors may get back less than the original amount invested and past performance information is not a guide to future performance.
Prudential Portfolio Management Group Limited, is registered in England and Wales, registered number 2448335, with registered offices at Laurence Pountney Hill, London EC4R 0HH.
The value of investments will fluctuate, which will cause prices to fall as well as rise and you may not get back the original amount you invested. Past performance is not a guide to future performance.