Surprise of the day... another bad idea

from the EU. This time, the collected EU finance ministers seem to think that by buying back Greek debt at depressed prices, they can really make those irritating, overpaid bankers pay. There may be clever ways of engineering such an outcome, but the basic idea doesn't work. In one of their classic papers, Bulow and Rogoff (QJE 91) show that buying back debt is no way to reduce the burden of too much debt. Open market buybacks, at least, "allow creditors to reap more than 100 percent of any efficiency gains". Why? A country's repayment capacity is its repayment capacity. As you buy back debt, you effectively spread that capacity over ever fewer bonds - there is more blood, sweat, and tears squeezed from the Greek taxpayer for each bondholder remaining. The secondary market price of debt rises as repurchases proceed. Most likely, the market value of all bonds outstanding stays roughly constant as the face value declines. So, what to do? Maybe buy-backs are the answer, but to get it right, get a good advisor to design the process - like Paul Klemperer of Nuffield, Oxford, who advised the UK government on G3-spectrum auctions.

How many big booms in US real estate?

You'd think we would have a lot of data on how the price of housing has moved over time -- after all, it's the single biggest investment most people make in their lives. Oh, and housing is at the heart of that small meltdown in world financial markets that you might have heard about. Actually, you would be quite wrong. Eugene White from Rutgers, who is spending two weeks visiting us at CREi, has a new paper arguing that existing price indices for the 1920s and 1930s are off by a large factor. Case-Shiller, for example, relies on a survey of home owners to figure out what price movements were before 1932. Now, I haven't looked at the fine print, but it's pretty clear that one wouldn't want to write this history of prices of pretty much anything based on what people remember... Eugene finds that evidence from construction volume suggests a much bigger boom (and bust) in US real estate in the 1920s than previously thought. Standard histories - like Kindleberger's book on manias - mention speculation in Florida, but do not have much to say about nationwide price movements. This would imply that the pre-2007 mantra ("house prices have never gone down") is wrong.

I am prepared to believe that we got the house price series wrong (and actually, Tom Nicholas of Harvard B-School has a new series for Manhattan that suggests we can do much better than the data currently used). Crucially, Eugene asks why the big bust in housing at the end of the 20s didn't have the same effect on the financial system as it did today. Financing was more conservative, with 20% down regarded as quite risky already. Bank regulation was much tighter, with strong limits on how many mortgages could be kept on the books in relation to assets. Financing housing, it seems, need not create a powderkeg...