Tuesday, January 5, 2016

Brad DeLong — Martin Sandbu: Free Lunch: On Models and Making Policy


Models are tools. They are not religious dogmas.

Larry Summers is absolutely right that economists and policymakers need to take into account important factors that models do not adequately capture if they even consider them in the assumptions. Whether business confidence is one of them is debatable but Keynes is on the side of those arguing for it.

However, the recent financial crisis that went global that economic models did not detect developing and were unable to adequately address by way of remedy proves beyond the shadow of a doubt that the Fed was remiss owing to erroneous assumptions that only models count. 

The Fed is the chief regulators of the financial system and years before the crisis broke, the FBI (Federal Bureau of Investigation) warned of massive fraud in the mortgage markets. Alan Greenspan, who was the Fed chair at the time, disregarded the warning because his conceptual model ruled out the financial sector acting against its longterm interests for short term gain.

This has not only been exhaustively documented but Greenspan himself admitted that he had been wrong about this assumption.

As Bill Black has noted, however, the work of George Akerlof and Paul Romer was already on record.

Three Passages From Akerlof & Romer’s 1993 Article That Should Have Prevented The Crisis

Finance Refuses to Take Akerlof and Romer Seriously about Looting

From this perspective, the crisis was no "black swan" event that could not be foreseen. 

Another proof lies in conventional economists' disregarding of Hyman Minsky's work. Only a few, notably L. Randall Wray, who was Minsky's student, were saving red flags. Why os few? Minsky's model was not formalized and was "merely" conceptual. It was only after the crisis that the economics profession discovered Minsky. It remains to be seen what will come of that. So far, the results with respect to reform are not encouraging.

1 comment:

Matt Franko said...

Tom if they rely on the Minsky stuff "stability creates instability" then no one is going to listen to that kind of thing... its a non-specific statement (think Schiff or someone like that ie "its going to happen.... someday!") and not technically credible...