Is Ben Bernanke standing in left field when he says that an economic recovery will take hold later this year? Leading economic indicators tend to agree with his premise. A year ago we published a report detailing indicators that generally provide good lead times to an economic recovery. Please refer back to the May 22, 2008 report titled Predicting the U.S. Recovery: Some leading indicators are better than others, for details on the predictive nature of the various indicators. Among the five economic and financial indicators we cited, several are now on the upswing, pointing in the direction of a recovery in the fall or shortly thereafter.

Housing data shows a turning point

Among the leading economic indicators, housing data on building permits and new home sales are believed to be the ones to watch, given that the housing market kick-started the U.S. into a recession. It appears from the adjacent graph that the permits data hit a trough in November of 2008, while new home sales were quick to follow in December. Both of these indicators are generally reliable, providing few false signals of a recovery. However, they also provide advance notice of a recovery with lead times that not only vary widely across recession cycles, but can also be quite lengthy, as much as 10-15 months. Using these indicators to precisely predict the end of a recession is risky business, but it does give us two important messages. First, it is preliminary evidence that momentum has turned, an important precondition for a recovery. Second, it gives us some outer-bound sense of the timing of the recovery. If we take the incidences of longer lead times as the outer bound, it suggests a recovery would take hold by February of next year at the latest, but perhaps more realistically in the context of other data, the recovery would come sometime in the September to December period.

Initial jobless claims was the other economic indicator highlighted in the original report. This indicator may have peaked in early March of this year, but it's hard to hang a hat on it, since it is a leading indicator that is prone to sending false signals. Already in this cycle, it has produced many head fakes. We predicted that the 'peak' signal it sent in March 2008 would prove to be false, and indeed it was. Since then there has been other such instances. The one difference this time is that the downward momentum in the annual growth rate has been sustained for eight weeks, which is the longest period yet. But, if it is a 'true' peak and history turns out to be a consistent guide, it suggests that a recovery is within 2-4 months of that peak. This places an economic revival in the May to July period. Admittedly, there has been some improvement in economic

and financial conditions in the current quarter, but the reversal in this leading indicator does not appear to have been dramatic enough to meet this timeframe. So we remain unconvinced that the recovery is that close at hand, causing us to believe one of two things. Either it is sending another false signal or the predictive power of this indicator may prove to have a longer lead time than its traditional relationship.

Stock market gives thumbs up

There are two financial indicators that are often cited as tried and tested predictors of economic recoveries: the stock market and the 10-year Treasuries-Fed funds yield spread. The latter, however, fails to provide much information this time around due to distortions caused by the credit crunch and actions taken by the government and the central bank to pump liquidity into markets. Perhaps more telling has been the retracement of corporate credit spreads in recent months signaling that confidence is slowly but surely creeping back into the market with some of the aversion to risk subsiding. By the end of April, investment grade corporate bond spreads had returned to the lowest level since September 2008, while the spreads for junk bonds had also narrowed significantly.

Still of value as a leading financial indicator is the signal sent by the stock market (S&P500), which reflects market participants' expectations of discounted future earnings (alternatively, expectations about the future economy). The S&P500 has rebounded nearly 30% since the trough in early March. This indicator is signaling an economic recovery will take hold over the next 3-6 months. That places the outer-bound estimate at September, which is also within the range that the housing market indicators are signaling.

The stock market can send false signals, however, which was the case in the 1974/75 and 1981/82 recessions. A potential telling marker of a false signal we identified in the original TD report in May was that the S&P500 must sustain a rebound in excess of 9% in order to minimize the

odds that the market has hit a false bottom. This marker proved correct in March 2008 when the S&P appeared to have bottomed following the Bear Stearns incident. The market rebounded in the 9 weeks that followed, and while it came close, it could not break above the 9% threshold and it subsequently reversed course. With the collapse of Lehman Brothers in September of that year, there was no getting around the true extent of mortgage and credit market difficulties, which intensified downward pressure on stock markets the world over.

However, we currently stand in the 9th week of a sustained recovery in the S&P500, and it has far exceeded the minimum criteria of the 9% threshold, raising the hopes that the bottom in March could be the real deal.

Concluding thoughts

So what are the take-away messages from the various signals being sent by the housing data and stock market?

1. There is a consensus among the leading indicators pointing to a recovery this year. Rather than a united front arguing for a worsening in economic conditions, tentative turning points are becoming evident within

the data.

2. The upward momentum must hold for the recovery to materialize by the end of this year. All bets are off if there is another 'surprise' such as another major financial institution failure (recall the Lehman impact).

3. Less bad still isn't good. While we may have seen the growth rates in new home sales and permits bottom, they remain deeply in the red, with annual growth in new homes sales down 22% in March and the 6-month

annualized trend in permits showing a 50% contraction.

4. This magnitude of ongoing declines leads into the last point: the timing of the recovery says nothing about the strength of the recovery. Rather, it only implies a trough in the cycle. In our official forecast we believe the fourth quarter will mark a return to growth, but one that is shallow at less than 1%, with a gradual grind higher as we progress through 2010. See TD Economics' Quarterly Economic Forecast for details.