How the fate of the US economy rests on a Dell workstation
Quick, someone send Bernanke a supercomputer
When I asked how the Financial Accelerator was implemented, Gertler said: "The model is a set of non-linear equations. We take a loglinear approximation around the steady-state, which gives us a set of difference equations. We then solve use software such as Matlab or Dynare."
In the NPR piece, Gertler said that the Financial Accelerator model had been used in the South Korean economic crisis a decade ago, where gross domestic product for that country dropped by 12 per cent - with half of that drop coming due to the effects of feedback loops of ever-tightening credit.
Gertler added that his colleague at Boston University was using the model to look at current events, so I contacted him.
You wanted a bailout? You got a Dell dude
As it turns out, Gilchrist's version of the Financial Accelerator model is running on a Dell workstation, and it implements the 20 equations to describe the economy during a credit crunch in a programming language called Matlab from MathWorks. Graduate students run the model, which takes GDP, economic output, investment spending, employment figures, corporate bond spreads, and other economic data to figure out how these changing figures will affect the economy.
A simulation run to solve the equations takes a few hours. "Even before the September meltdown, the pressures in the market have been reducing GDP by two per cent," Gilchrist says. This is due to the financial accelerator effect, adding that this has been partially offset by relatively loose monetary policy.
Interestingly, the Financial Accelerator simulation does not model the financial services sector, and therefore Bernanke and secretary of the Treasury Hank Paulson have not been calling Gilchrist up to run some numbers.
According to Gilchrist, the Bank of England and the Bank of Canada have put together some computerised simulations of their countries' respective economies, and the Fed was talking about possibly putting together a more sophisticated model. But there are limits, apparently.
"These types of models are rarely used for economic forecasts," explains Gilchrist.
But why not? Why isn't there a supercomputer somewhere, right now, simulating the effects of a $700bn bailout?
I think right now we'd all settle for a real good spreadsheet, and maybe the thing to do is get Gilchrist a few more grad students, a baby supercomputer like he says he will need to be able to pump real-time data into the model, and get us to a point where we can do some more sophisticated "what-if" modelling on economic policy. Put an earmark for the project right into the bill as it passes from the House to the Senate.
I studied rocket science, and I can tell you - this ain't rocket science.®
Sponsored: Benefits from the lessons learned in HPC