One of the more interesting charts is the SPX, pre-1990-91 Recession. If markets truly had forecasting abilities, then one would presume they would have sold off in advance of the recession.
click for larger graphic
Instead, this chart shows the index rallied right into the start of the contraction. That is hardly predictive. In this instance, the markets acted coincidentally to the recession, instead of predictively. The beginnings and ends of recessions are marked by The National Bureau of Economic Research, and are based on quantitative data.
Calculated Risk declares recession forecasting a "mug’s game" — but also agrees to play. He does not see a recession as imminent. That’s consistent with my own expectations for a contraction in 2006/07 time frame.
Other sources have had some success with recession forecasting. As one commentor here noted, ECRI has a good record forecasting economic turns.
Yesterday, I linked to Jim Stack (InvesTech Research) recent recession alert, which was reproduced at Forbes. I’ve been a subscriber, and I find Stack’s research very interesting, his track record good, and its costs relatively inexpensive.