# A “Picture war”

Lately Mike Kimel at Angry Bear has been jostling with Scott Sumner about how fiscal policy better explains what got the economy going in 1933. His latest post on the topic says:

While I´m a firm believer in the importance of monetary policy I don´t believe it made much of a difference in the New Deal era. As figure 3 shows, changes in nondefense spending – hiring people to build roads, dams and the like, explain subsequent changes in real GDP growth rates exceptionally well from 1934 to 1938. This simple model explains more than 90% of the change in real GDP growth rates over that period.

OK, he finds a positive relationship among second derivatives (changes in growth rates) of the relevant variables, but in this “picture war” I think my pictures, showing LEVELS of variables and associating their behavior with the monetary policy reversal in March 1933 when FDR took the US off Gold and announced a “price level target”, are much more convincing. (Note: they come from this post from early this year).

## One thought on “A “Picture war””

1. Boa tarde.

A few comments about my post, if I may.

I didn’t post a graph with levels of real GDP because I thought the fact that the growth rates were correlated so highly was even more impressive.

For what its worth, the first year for which its possible to compute the little function of nondefense spending (i.e., [(2/3) * nondefense spending t/ GDP t-1] + [(1/3) * Nondefense spending t-1 / GDP t-2] is 1931.

The correlation between that series from 1931 to 1938 and real (not nominal) GDP in the following year is above 95%. No first or second derivatives, merely levels.

Now, the lead time is important because nondefense spending at time t is determined at time t-1. That means that real GDP in time t is being explained by a function of a variable that is partly determined in time t-1 and partly in time t-2. If there is causality, it can only run one way.

Conversely, monetary policy at time t is determined in large part by the state of the economy at time t. You have M1 and M2 bottoming out at the same time as nominal GDP. Which leads?

This site uses Akismet to reduce spam. Learn how your comment data is processed.