Just ask my first wife.

But seriously, there is a problem here. Economists like to think of themselves as physicists who work on society. Everything has to be stated in mathematical terms. If something can be proved by algebra, geometry or logic it is true. Otherwise it is dubious.

Unfortunately most social relationships involve trust, continuous bargaining and complex adjustments. Mathematical modeling of this sort of thing is difficult if not impossible - although it has been attempted by George Akerlof among others. The only social relationships that are really tractable to mathematical theory are commodity exchange - the purchase and sale of precisely defined goods by interchangeable agents - and chattel slavery - the arrangement whereby one person owns and commands the time and life of another.

Ultimately at the base of most economic models we find either one or the other. We pretend that social relationships can be precisely described through complex contracts. Then we are dealing with commodity exchange. We may also stipulate that one party in an exchange surrenders complete control to the other party, although this surrender is usually limited in time. Thus the economic modeling of labor relationships is simply a variant of chattel slavery.

It is as if we attempted to describe human sexuality and reproduction using only prostitution and abusive patriarchal marriage.

Which would be fine if only economists were damaged by this perverse thought process. Unfortunately a great part of the moral initiation of our business classes is entrusted to economists and, because this group already has a tendency to oversimplification and various moral deficiencies, the damage done is great.

Many of us are familiar with the various abusive employment relationships that this has fostered but Wonks Anonymous would like to point out the damage that it has done to financial markets.

Over that past decades US financial markets have fundamentally changed. Banks once knew their borrowers and held most of their own loans. This led to the maintenance of complex relationships between borrowers and lenders - possibly including discrete phone calls to borrowers who were seen to be spending too much. With the advent of financial engineering we began to pretend that loans were commodities, precisely described by credit scores and complex mathematical models.

This greatly simplified individual banking transactions. Why read and verify a loan application when you can just mail out a credit card on the advice of your infallible statistical model? It also enabled banks to mass produce securities based on individual debts. It worked really well except that it didn't.

Joe Nocera contrasts this sort of banking with the more staid practices found in India.

Guess what? India doesn't have a serious domestic financial crisis.