CERF Blog
Dan started this, but he has some minor surgery today. Kirk and Bill finished it:
CERF released its first United States and California forecast last week. The United States and California forecasts are pessimistic relative to consensus. Why?
In part, it is because so many forecasters seem to be using a model with a high autoregressive component. This is the equivalent of the weatherman forecasting that tomorrow’s weather will be similar to today’s weather. There thinking goes like this: Since the third quarter appears to be positive, the fourth quarter will also be positive.
Here in Southern California an autoregressive weather forecast works pretty well most of the time. However, you can improve the forecast by calling San Francisco. If it is sunny in Los Angeles but raining in San Francisco you might improve your forecast by including at least the possibility of rain.
We try to do the economic equivalent of calling San Francisco, and look at lots of data series. Some of those key indicators show significant weakness in the economy. Here are a few that we think are particularly important right now: bank charge-offs, the home ownership rate, foreclosures, capacity utilization, and an interest-rate-weighted TED spread.
It is difficult for us to see a near-term structural improvement in the economy when Bank Charge-offs are reaching new highs. It is unlikely that residential foreclosures will subside much given three factors: high unemployment rates, ongoing price declines, and a high home ownership rate. Of course, foreclosures add downward pressure on housing prices. Commercial foreclosures have yet to peak but are on the rise. Capacity utilization remains near historical lows. The TED spread has fallen, but the interest-rate-weighted TED spread is still very high.
The important thing about our forecast is not the point forecast of United States or California GDP growth for a particular quarter. Rather, we believe that there is still a critical mass of structural factors that imply future weakness in GDP growth.