Month: September 2009

The Danish model has another critical and innovative feature.  Holders can retire their own mortgages by purchasing the same face amount of mortgage bonds at the prevailing market price.  To prepay a mortgage by purchasing bonds, the home owner must give advance notice of several weeks to the MCI [mortgage credit institutions], which designates by lottery the specific bonds to be purchased.  Thus, if rising interest rates or other factors cause mortgage bonds to trade at a discount, home owners can reduce the principal or retire the whole mortgage by purchasing an appropriate mortgage bond at a discount.

That passage is from Robert Pozen's new and notable Too Big to Save? How to Fix the U.S. Financial System.

You can't do this in the United States.  You can pay off your mortgage but the "face value" of that transaction does not vary with market conditions.  In essence the Danish system creates a new contingent claims market for homeowners who do not understand how to use interest rate futures and options.  De facto, the homeowner receives some implicit insurance against the prospect of negative equity in the home.

Here is The Economist on the Danish model.  Denmark also allows for speedy repossession of property,  in case of default.  Here is a more general discussion of the Danish model, which emphasizes transparency.  Mortgage finance is conducted by explicitly designated institutions and originators retain a financial interest in the loan, even following securitization.  The emphasis is on plain vanilla products.  Here is Wikipedia on the Danish mortgage model.  Here is a longer study. 

It was great having him in residence for a few days and we discussed everything from Denmark to the financial crisis to the downturn of 1920-21 to the future of China. 

I won't detail Scott's account of the crisis (see Scott here, or my pared down summary here), but in sum Scott believes that easier money at the critical turning point would have made a big difference.  I have a few observations:

1. Scott's recommendation of higher price inflation, or higher nominal gdp growth, would have eased the crisis, in my very rough guesstimate by one-third and in absolute terms that is a lot.  It's the best free lunch I've seen in years.  Listen to Scott!  Yet, in my view, easier money would not have eliminated most of the crisis, given the partial or total insolvency of many financial institutions, the negative AD shock from the collapse of the housing bubble, and the need to halt and reverse the ongoing accumulation of debt, among other factors.  Scott disagrees.

2. Scott's account does not deny (but does not emphasize) that the initial downturn was accompanied by a fall in monetary velocity.  This opens up room for real shocks, resource reallocations and recalculations, and animal spirits to be driving the broader story. 

3. The relevant real shocks behind the downturn are plausibly: the decline of debt-financed consumption, mis-estimation of permissible leverage, the collapse of the real estate bubble, the revaluation of the risk premium, sectors hit by that revaluation, such as the non-profit sector suffering from tumbling equity prices, the required shrinkage of finance, sluggish behavior in the energy, health care, biotech, and educational sectors in terms of real productivity and job creation, and the collapse of non-Google advertising.   I see the revaluation of the risk premium as the most important of those.  And please note, when I refer to real shocks I don't mean technological forgetting or the Minnesota RBC theories.

4. Which empirical test would separate out the employment effects of the real shocks from the employment effects of the drop in nominal gdp?  I observe that many of the ailing sectors have relatively flexible nominal wages (real estate agents and journalists are two such cases), which leads me to think the real shocks were more important.  It would be good if we had a more formal test.  If the nominal and real theories are observationally equivalent, my gut suggests that favors the real theories but I don't have a clear argument on that point.

5. I observe that the Eurozone (plus pegging Estonia) has a single monetary policy yet some fairly divergent outcomes.  Estonian gdp has fallen off a cliff — about negative seventeen percent — while German gdp has fallen in the four to five percent range and now seems to have bottomed out.  Note that Estonia probably has more flexible wages and prices than does Germany.  Those facts also point to real shocks as being more important for the crisis, namely which economy was more bubbly in the first place and which required a greater revaluation of the risk premium.  The ailing Spain is another example.

6. Many of the AD shocks in the crisis, such as resulted from the decline in housing wealth, come from shocks to real wealth or income, not shocks to nominal wealth or income. 

7. Scott in his talk admitted and indeed emphasized that monetary and real shocks come bundled together.  What methodological or empirical view would properly lead us to call one primary and the other secondary?  Which came first?  Which has a higher marginal product of destruction without the other?  Which explains more pieces of data?  Something else?  I am inclined to call the real shocks primary and the secondary deflation…well…secondary.  Scott portrays the deflation (he doesn't call it the "secondary deflation," as I do) as the primary problem.

Arnold Kling posts on these questions here and here.  In Arnold's terminology I see the Fed as controlling nominal gdp but not always real gdp.

If you are looking to have in a visitor or a speaker, you cannot do better than to try Scott Sumner.  Maybe someday Scott will tell you about his favorite movie director, or his views on India, either of which may count as his most absurd belief.

Addendum: The Cato debate with Scott continues.  And do read Scott's response in the comments.