Housing still flat, but manufacturing is picking up

Housing starts essentially have been flat for two and a half years, at the lowest levels in recorded history. There is anecdotal evidence that residential construction activity is picking up in the remodeling category—I should know since we have been doing a bit of that ourselves—but so far there is no sign of life in the new construction industry. The longer this goes on, the smaller the excess inventory of homes on the market, and the more that new construction will have to ramp up in the future to catch up with ongoing housing formations that almost surely exceed the current pace of housing starts.

U.S. industrial production in July proved to be much stronger than expected (up 0.9% vs. expectations of up 0.5%), and May and June numbers were revised higher. Over the past three months, production is up at a 6.1% annualized rate, which is close to the rate which prevailed in the second half of last year. It's pretty clear that the soft patch earlier this year—triggered by supply chain disruptions following the Japanese tsunami—has passed, and activity is gearing back up. Although Europe appears to still be lagging, there is no sign here of any double-dip recession in the U.S. This underscores the likelihood that the recent equity selloff was inspired almost entirely by fears of PIIGS-related financial troubles in Europe, and was not based on any evidence of a downturn in the the U.S.

I've featured this last chart before, even though it is somewhat of a curiosity. The Phillips Curve theory holds that high levels of capacity utilization—which correspond to low levels of resource slack—cause inflation to rise, and vice versa. As the chart shows, the fit between capacity utilization rates and inflation (with inflation taking about 17 months to respond to changes in utilization rates) was very strong in the 70s, but much less strong (even nonexistent some would say) in the 80s and 90s. My explanation for why this is so is twofold: first, while the correlation was strong, it was not necessarily because of causation; and second, monetary policy was very reactive in the 70s, and very proactive in the 80s and 90s. Proactive monetary policy anticipates changes in inflation, and thus dampens inflation over the course of a business cycle, whereas reactive policy only augments inflation.

Monetary policy has since become more reactive in the 00s, which helps explain why the correlation between inflation and capacity seems to be picking up. If that is indeed the case, then the big improvement in industrial production augurs for a meaningful pickup in core inflation in the next year or so.