Jim Bianco has some comments on this year’s horrific analyst misses:
>
At Bianco Research, we have demonstrated that earnings forecasts are missing by their widest margin ever in 2009. While we have over 140 years of actual earnings data, IBES earnings estimates date back to 1985.
The chart to below shows the error rates in forecasting operating earnings for both top-down forecasters (strategists) and bottom-up forecasters (company analysts). It measures the forecast one-year forward versus what actually happened one year later. The data covers the period from 1985 to 2002, and unfortunately we have been unable to secure an update. It shows that the largest forecasting errors ranged from +30% in the late 1980s (meaning the forecasters were far too optimistic) to -20% during the recession of 2000 (meaning they were far too pessimistic). Data from 2002 to 2006 is missing, but it is a safe assumption that the earning misses that occurred during this period were not as extreme as the previous records set.
>
>
The next chart uses Bloomberg data for bottom-up forecasts. By mixing IBES and Bloomberg forecasts, we risk comparing apples to oranges. But, if the surveys are done properly they are both surveying the same people and should offer a very similar data set. As highlighted on the chart below, the current forecasting error is now near 100%, more than three times the largest error seen from 1985 to 2002. No other period has ever come close to the current period.
>
>
The final chart comes from S&P. It shows their measures of top-down and bottom-up forecasts for 2009. The blue line shows how bottom up forecasters were quickly revising their forecasts down like the Bloomberg data shows above. The red line shows the top-down forecasts were not too far off from reality. The top-down forecasters miss was on the level of the late 1980s and early 2000s misses.
>
>
Earnings forecasts are worse now than at any time in human history. As the chart below shows, the decline in earnings in the last year was worse than any decline in the last 140 years (again, no typo). The data comes from Robert Shiller who got it by way of the Cowles Commission.
Do you recall anyone forecasting a biblical collapse in earnings? We do not. In fact, Dr. Jeremy Siegel set records in tortured logic to explain this was not happening last February.
Rather than writing about rebounding earnings, a more interesting story might look into why consensus estimates have been so horribly wrong lately. Do the forecasters understand the degree to which they missed, and have they changed their methodology at all to avoid a similar fate should this happen again in the future? Only when these questions are answered should anyone pay attention to these highly inaccurate guesses.
What's been said:
Discussions found on the web: