What did the Fed know, and when?

Readers of my stories and blog will know that I am obsessed with the quality of the government statistics. Very often mistakes in economic policy can be traced to data which is misleading or later revised.

One of the big puzzles is why the Fed kept raising rates in 2006, despite what we know now was the incipient weakening of the housing market. The minutes of the January 31, 2006 FOMC meeting, indicate a slight bit of concern with the housing market, but nothing significant: “Activity in the housing market appeared to continue at high levels.”

Here’s one reason why the Fed got it wrong: The NIPA data was giving them totally the wrong picture. According to the latest numbers at the time, housing investment was accelerating at the end of 2005. In reality, we now know that housing investment was sharply decelerating going into the January 2006 Fed meeting.

Take a look at this chart.

The dark blue line is the real growth rate of private investment in new residential structures in 2005, as of the January 27, 2006 GDP report (available, obviously, before the FOMC meeting). The light purple line is the current estimate of the real growth rate of private investment in new residential structures.*

Guess what? They look like they come from completely different economies. At the time of the Fed meeting, they had no idea that they were in the middle of a sharp collapse in the housing market.

Let me extend the current estimate a bit

In retrospect, would the Fed have kept raising rates in 2006 if it had the correct data about the collapse in housing investment in late 2005? Perhaps not.

*The data comes from the bottom line in table 5.3.6. The January 27, 2006 data comes from the February 2006 Survey of Current Business.

Before it's here, it's on the Bloomberg Terminal.