The Wall Street Journal has a story about a new bond being issued by the GSEs (HT: Nick Timiraos). They are starting to issue bonds that, as far as I can tell, basically take the default risk when there is a systemic default crisis. This is similar to the CDO securities that collapsed when defaults rose across MBS pools.
This seems like precisely the wrong direction to take. And, I think that this is a good example of how critical one's prior estimation of the robustness of markets is to one's policy conclusions. In an economy with extensive bank regulation and a fiat currency, I submit that systemic risks are, practically by definition, a public problem. If a variety of public institutions have broad influence over the value of the currency, the growth of mortgage lending, and general cyclical banking postures, then the responsibility of systemic risks should be socialized. If you blame the housing bust on private investors and banks, then this must sound like madness. But, even most of the people I hear arguing that the bulk of the expansion was a demand-side bubble are basically building their arguments on public institutions - an overly accommodative Fed, the GSE's, the CRA, etc.
If the sort of proto-Austrian business cycle view that seems to be common is true - that the Fed caused the economy and the housing market to overheat in the 2000s - then why shouldn't taxpayers take the losses? Isn't the idea in that point of view that the "overheating" is an inevitable tragedy-of-the-commons problem? Why should the hapless private investors who are supposedly misled into making suboptimal investments take the loss? Losses should come from individual errors, not from being captive on some public-policy roller coaster.
Now, I don't agree with those assessments of the housing boom. I think any excesses were a fairly small, and late in the cycle, that most of the rising prices were a reflection of supply constraints, and that most of the public errors were errors of demand-deprivation. But, in the end, my position shouldn't lead to a significantly different public posture than the pro-bubble narratives. In both narratives, the responsibility for systemic losses should be social.
I suspect that most of the de facto benefits that might come from having the GSE's come from the fact that, in an economy with a fiat currency, matching balance sheet risks between nominal debts and real collateral is difficult, both because inflation is potentially unanchored, and because nominal shocks can create systemic default events. So, if the GSE's are socially useful, it seems that one of those uses is to take on that default risk. Does anyone believe that we would have been better off in the crisis if the GSE's hadn't absorbed much of those temporary losses?
In private MBS markets, the systemic risk seems like an unavoidable cost. Modern Portfolio Theory is based on the idea that systemic risk earns a premium because it is unavoidable. I don't see how there is even a natural buyer for these bonds. It's not like we have a shortage of systemic risk and institutions are trying to find ways to add more. Institutions earn a premium by taking systemic risk for others.
This is one of the benefits of NGDP level targeting - that it might lower systemic risk, and therefore lower the premium required for risk-taking investments. Thinking about the GSE's in this way, and thinking about the social responsibility for systemic economic issues in a mixed economy, thinking outside the box a bit, maybe government should consider taking on more systemic risk. Maybe we should issue Treasuries and use the public funds to purchases broad shares in equities - like an S&P500 basket, or a Russell or Wilshire total market basket. Maybe the best public investment strategy is to make private investment naturally short on systemic risk and long on idiosyncratic risks. Private investors don't want systemic risk. It pays a premium because it poses a problem. But, also, this would put public institutions in a position of having a strong incentive to avoid systemic crises. Some might argue that we have been privatizing gains and socializing losses. Is there a way to socialize systemic gains while keeping idiosyncratic gains private?
Whatever the answer, and whatever set of adjustments we need to make to our real estate funding conventions, selling bonds on the systemic risks of the GSE pools hardly seems like a helpful change.