Monday, 30 June 2008

Analyst Earnings Forecasts Still Too Optimistic

As stated here many times, it is the prediliction of analysts to be overly optimistic in their earnings projections. This becomes even more apparent at turning points in the earnings cycle.

As shown below, even after S&P500 operating earnings fell -9.5% in 3Q07, analysts were predicting positive earnings growth of 2.9% for 4Q07 as late as the middle of November. As we now know, 4Q07 earnings plunged -30.8%.

However analysts were not deterred and returned with some rosy forecasts for 1Q08. As late as the end of February analysts were still expecting positive earnings growth. As shown below, by the end of May it was clear the analysts had it wrong as the quarter ended with a -25.8% decline in S&P operating earnings.

After clinging to forecasts of positive earnings growth for 2Q08 until the end of March, the reality of first quarter results forced analysts to get a little more realistic. As of June 24th S&P500 operating earnings are forecast to decline -9.2%.

Call me skeptical, but given analysts demonstrated track record of being behind the curve, I would estimate that the final number for 2Q08 will be closer to double the decline currently forecast.

As we enter the third quarter year over year comparisons will become much easier and that is why currently S&P500 operating earnings are expected to show positive growth for the full year. However, as can be seen below, those estimates have also been eroding, from a peak of 17.4% expected growth in February to a more modest 6.7% as of a week ago. My bet is that by the end of September, analysts will wake up to the fact that full year earnings growth will be negative.

Of course, sooner or later analyst forecasts will get too pessimistic, however we are still a long way from that point. The chart below shows S&P500 earnings from March 2008 through to December 2009. During this time there have been 2 other earnings recessions which surprise surprise, coincided with an economic recession.

The point of this chart is to show how long it took for earnings to recapture their former highs after the cycle had turned. After the peak in June 1989 earnings did not hit new records until June 1994 - a full 5 years or 20 quarters later. Earnings then peaked again in June 2000 and didn't recover to the previous peak until December 2003, a full 3 and half years or 14 quarters later.

Currenlty analysts would have us believe that the earnings peak in June 2007 will be surpassed in December 2008 of this year, just 1 and a half years or 6 quarters later. Whilst it may be true that every earnings cycle is different I doubt it will be that different.

To be sure, a lot of the over-estimation of earnings growth has been the expectation that the financial sector would recover. Drilling down to the sector level, 5 sectors out of ten are expected to post record earnings for 2Q08. However, as the credit crisis is now infecting the broader US economy, expect companies in an increasing number of sectors to post less than cheery outlooks for the second half of the year.