New Evidence of Government Induced Risk

A year ago I wrote Getting Off Track one of the first books on the financial crisis. I argued, based on data available at the time, that government actions caused, prolonged and worsened the crisis. After a year of debate, this early assessment is holding up well. Indeed it is being reinforced by new evidence.

Consider, for example, the government actions associated with the takeover of Fannie and Freddie in 2008. Of course, Fannie and Freddie were a big part of the reason for the explosion in mortgage debt including risky subprime mortgages, but I want to focus now on the impact of government actions relating to these institutions during the period leading up to the panic in 2008.

A good way to assess this impact on risks is to look at the spread between interest rates on subordinated debt and senior debt at Fannie or Freddie. When investor concerns about risks at the institutions increase, the interest rate spread between subordinated and senior debt rises. So what caused the movements in these spreads in 2008?

The three charts show the spread between the interest rate on Fannie Mae subordinated debt and Fannie Mae senior debt over three different periods. The yellow line with the green shading in the lower panel of each chart is the interest rate spread of sub debt over senior debt. In the top panel (harder to read) the orange line is the rate on subordinated debt and the white line is the rate on senior debt. The first chart focuses on the period from June 2004 to February 2008. The spread was fairly stable fluctuating in a rather narrow range around 20 basis points over this period.

The second chart covers the period from March 2008 through June 2008. Observe that the spread jumped to around 80 basis points after the Bear Stearns intervention. Unlike many other risk spreads it did not come back down after Bear Stearns.

The third chart focuses on the period from July 2008 through September 2008. There are two big upward jumps during this period. The first was on July 11. What was the big event that day? It was a leaked news story about a possible government action, a takeover of the two institutions. In particular the New York Times ran a front page story with the headline “U.S. Weighs Takeover of Two Mortgage Giants

The other big jump up was on Monday August 18th. Again the reason was a news story about another government action. The previous Friday the Washington Post reported that the Treasury hired Morgan Stanley to assess the vulnerability of Fannie and Freddie, a strong indicator that the government was looking for outside justification to take over the institutions. It is important to note also that certain other events, which could have moved the spread up, did not move it up. When Freddie and Fannie released their second quarter earnings in August, there was little to no reaction in the spread.

So the major movements are clearly linked to government policy decisions and news stories about them. This timing does not prove that the government actions were responsible, but at the least it raises questions about why rumors of possible government actions were leaked to the news media.

The questions are important because some government officials have indicated that these jumps in the subortinated debt spread were part of the evidence to justify the take over the institutions at this time. But the evidence shows that the government itself was increasing the spreads.

This entry was posted in Monetary Policy. Bookmark the permalink.