Excellent Short Discussion of the Crisis

Gary Gorton is a monetary and financial historian who wrote a widely cited and well respected paper a couple of years ago on"The Panic of 2007," in which he explained more clearly and in greater detail than anyone else how such recent innovations as CDOs and SIVs worked and then interacted during the financial crisis. He has now written a shorter "Questions and Answers about the Financial Crisis," which I think is one of the most important contributions to explaining what happened. It ranks right along side the EconTalk interview with Charles Calomiris. Basically Gorton argues that there was a sudden, partly unrecognized panic in the market for repurchase agreements (also known as RPs or Repos) at the end of 2007.

RPs are close money substitutes. Bank-issued overnight RPs were an important way banks got around regulations so they could pay high interest to large depositors in the 70s and early 80s, and were counted in M2 until 1997, when the Fed moved them into M3, where it already counted term RPs issued by banks. Gorton's analysis implies that I was seriously mistaken about the insignificance of the Fed's ceasing to report M3 in February 2006. M3 was discontinued just at the moment it was diverging from M2 and providing important information not otherwise available about certain monetary instruments.

Gorton's paper also clears up some other things that puzzled or intrigued me. Among them:

1. The huge drop at the outset of the crisis in RP borrowing by banks, to be replaced by borrowing from the Fed.

2. The prior, major transformation of investment banks from almost exclusively broker-dealers to enormous financial intermediaries in their own right.

2. The precise nature of the negative velocity shock that plays such a central role in Scott Sumner's explanation for the downturn.

4. How the subprime crisis precipitated this decrease in the velocity of reported monetary measures.

Some minor quibbles, however, I have with Gorton.

1. He doesn't make clear that, while bank-issued RPs were once counted in the official monetary measures (because of their immediate interchangeability with demand deposits), RPs issued by investment banks and other institutions were not.

2. He claims that no one knows the full size of the RP market, which he estimates at the incredible amount of $12 trillion. But surely most RPs must show up somewhere in the Fed's quarterly Flow of Funds accounts, which puts their total at only $2 trillion at their peak. Admittedly this reflects only RPs between sectors, and not intra-sectoral RPs made from one investment bank to another, but I still find it hard to believe that such lending and borrowing could raise the total by a full $10 trillion.

3. I think he underestimates the role of government intervention in creating this structural problem and precipitating the downfall.

Nonetheless, this is definitely a MUST read.

Hat Tip: Tyler Cowen