Friday, August 6, 2010

 Chart: Analysts Forecasts versus Actual Realization

The chart above is the result of McKinsey research into the forecasting abilities of financial analysts. For many years we do already know that analysts tend to be overly optimistic.

What we find interesting about this graph is that it clearly shows a pattern: or actually, two patterns!

First: actual earnings follow a cycle. Nothing new there, as long as there is business cycle research done by people like the late Victor Zarnowitz of the University of Chicago and NBER and many others we know this. The so-called Zarnowitz rule implies that troughs are followed by steep recoveries, and that is something we do indeed find back in the 'actual earnings' graph.

But there is a second pattern. A shocking one. Although most analysts are trained in top universities and/or hold CFA titles or similar international equivalents, it is almost as if they a) don't know about business cycles at all; and b) seem to make it a sports to predict earnings figures that would be realistic in case business cycles are at their maximum! The graph showing 5-year earnings forecasts by analysts fits nicely with actual numbers in the maximum cases, but is terribly wrong whenever the economy turns back to mediocrity or even depression.

The mistakes are not coincidental
Of course analysts are not silly, and of course they do know about business cycles. So what is going on here? Well, there are at least one explanation and that one is not a pleasant one for those of us who expect our analysts to forecast the 'truth'. Analysts do normally work for financial institutions that have either the analyzed firms or investors as their clients or both. Obviously firms would not be very happy when analysts associated with entities that they cooperate with when it comes to corporate finance services tend to be negative about its earnings potential. The negativity would suppress share prices, make it harder to receive credit at reasonably conditions et cetera.

And the brokerage departments wouldn't be too happy either. Investors are a weird lot sometimes. It turns out that investors are far more active in a good economic climate. Strangely enough they tend to believe that their portfolios require far more actions in bullish markets. First, they re-allocate money into stocks and away from less attractive asset classes. And second, they are more actively trading within the equity portfolio. When markets turn into bearish modes investors sell-off shares and act in a more lethargic way as far as the remainder of their equity portfolio is concerned. This does of course translate into cyclical patterns in brokerage fees. So no way that an analyst would make a lot of friends when communicating bad figures to the outside world while most others aren't.

And markets are not crazy
Wat does this imply for market valuations? Well, McKinsey did also look at Price/earnings ratios in the market and it found out that investors are at least to some extent aware of the shocking lack of cyclical sensibility of analysts. When comparing actual P/E ratios with the implied ones based on the analysts forecasts the average 'mistake' was just 25 percent. And that is far less than the average mistake in earnings forecasts that we can derive from the chart above. In other words: at least to some extent investors have gotten used to the overt optimism of analysts and they correct for it in the market. But: still not enough with the average mistake in P/E levels still being 25 percent.

What investors should doInvestors could and should correct for this by introduction of a multiplication mechanism that multiplies analysts forecasts by 1 whenever they expect a very bullish climate, i.e. one in which analysts forecasts are normally quite good; a value of 0 whenever they believe that markets will go into a deep depression and a value around 0.5 whenever they are not sure about things and expect either some kind of average year in business cycle terms or believe that the likelihood of either a bull market or a bear market is about 50-50. Result: you will probably end up doing a much better job with your fundamental investment strategy than you otherwise would. Obviously the alternative of finding realistic analysts is still the best, but somehow the job market seems to work in such a way that it is quite hard to find negativist, contrarian style analysts that follow the cycle with their forecasts. Herding between analysts is strong and innate to the system.

We are sure that this research will be continued. For instance: what does this mean for more cyclical industries? And what for more volatile countries, like the Emerging Markets that are now becoming more important internationally for investors? And what about neglected stocks? Is the lack of analysts in that case a blessing in disguise for investors? Et cetera.

No comments:

Post a Comment