Friday, April 19, 2013

In scientific computing, batch processing is king

Reinhart & Rogoff are out with their response to the critique by Herndon et al. Fine. Whatever.

I have never been a fan of Excel-jockeying. In fact, I do consider any study that depends on manipulating data in an Excel worksheet to be unscientific. Not only does Excel automatically change too many things behind one's back, but using a pointy-clicky interface also means you do not put together, however simple, a script that records the operations you performed on the data to get the result you got, and a log file of the actual output you generated.

Put simply, when it comes to scientific work, batch processing is king. Anything else, I have to take your word for what you did. Or, when it comes to the case of Carmen Reinhart and Kenneth Rogoff, they had to take the word of some R.A. whom they asked to do the Excel-jockeying.

I have mentioned before that I do not consider government budget deficits and debt to be the determinants of economic growth. The argument against big government does not depend on debts and deficits and it may not be possible to make the argument empirically because big government can produce high GDP numbers on paper.

So, in short, I find both the point and counter-point uninteresting and distracting from the main issue. But, something in Reinhart & Rogoff's response attracted my attention:

It is utterly misleading to speak of a 1% growth differential that lasts 10-25 years as small. If a country grows at 1% below trend for 23 years, output will be roughly 25% below trend at the end of the period, with massive cumulative effects.

For example, if the GDP grows at 1% for 25 years, at the end of the period, it will be 28% higher than the starting point. If it grows at 2% for 25 year, it will be 64% higher than the starting point, and if it grows at 3% for 25 years, it will be 109% higher than the starting point, representing more than a doubling.

So, the U.S. seems to have gone from mostly around or above 3% annual growth in real GDP, to about 2% in the recent years.

The reason for that is not current debt. It is the shadow the government sector (Federal, state, local) casts on the rest of the economy which feeds it.

No comments:

Post a Comment