Why Wall Street Strategists Always Seem Bullish

Let us divide those writing about forecasts into three camps:

  1. Those who make specific forecasts, sometimes required as part of their job;
  2. Those who criticize, but do not forecast;
  3. Those who make dramatic, non-consensus forecasts to get attention.

I plan a few more posts on this theme, but today I want to consider group 2. If you are seeking attention, it is easy to write a popular article about forecasting. Start with the viewpoint that the experts are dumb and that the average investor can do better. People love to be smarter than experts. Surveys show that 90% of all people are above average in intelligence! Well, maybe not 90%, but far more than half. They are very receptive to this approach.

Taking this easy target, the NYT cites a source claiming some great credentials. His report got a million page views and even more publicity in the sequel. I see plenty of bias and errors in his work, but let me start with the most colorful claim:

Vanguard’s move into PE may change the landscape forever

Private equity has been growing in popularity in recent years as more and more big-name funds and institutional investors dive in. Now even indexing giant Vanguard is out to take a piece of the PE pie. During a panel at the Morningstar Investment Conference this year, Fran Kinniry of Vanguard, John Rekenthaler of Morningstar and Read More

Now imagine having a coin calibrated to show “positive” 2/3 of the time, and “negative” 1/3 of the time. Flipping this coin would therefore outperform a Wall Street strategist!

This is an oft-cited concept. If the market declines 1/3 of the time (actual performance is a bit better, but we’ll go with the author’s numbers) and no Wall Street strategist forecasts lower stocks, supposedly that is proof that the experts are too bullish.

The author has quite obviously never had to forecast anything, and his math is seriously flawed. Suppose you merely forecast an up market. You will be correct 2/3 of the time. He uses his magical coin. 2/3 of the time it forecasts “up” and it is correct on 2/3 of those occasions. 4/9 in the win column. The coin forecasts “down” on 1/3 of the years, and it is correct 1/3 of the time. That is another 1/9 in the win column. So less than 56% right instead of 2/3.

The author also produces this mystery chart:

What is the wavy pink line? The wavy blue line is (apparently) a consensus average. The pink diamonds are an actual result. So, what is the pink line? A good chart has an explanatory legend, but this is a mystery.

If – instead of the mystery pink wavy line – you compare the blue line to the actual, it is directionally accurate in eleven cases, slightly wrong on four, and more seriously wrong on four others. The focus on the two lines distorts the results. There are other issues, including the time frame for analysis, but I am sticking to points that should be obvious to the average reader.

Everyone makes mistakes, but when you are calling out a lot of experts and possibly misleading investors, you bear a special responsibility to check your work. In the academic world of peer review, this article would not have been published as written.

Turning to the New York Times, readers expect a very high standard of reporting. The article does seek a little balance by looking for an accurate forecast, that of Seth J. Masters 2012 forecast of Dow 20K. It was good reasoning, like my own analysis two years earlier. Perhaps the author might have used (in George W’s words) “The Google” to search for Dow 20K.

The answer to the title question? Analysts are bullish because the long-term market trend is higher. In any given year, markets are likely to rise. If someone goes against the long-term trend, there had better be a compelling reason.

More to come on experts and predictions. My basic theme? A well-done forecast identifies the possible scenarios, specifies key variables, shows the range of errors, and focuses thinking for both the analyst and the reader.