Fed Independence and Alan Greenspan – Correction

Regular readers may recall I had a series of posts on how the Fed tries to influence elections. Bill Polley has gone through my spreadsheet, and found an error. I have corrected that error. This post shows some of the corrected information.

Table 1 shows the correlation between real M1 per capita and real GDP per capita using data going back to 1959, which is the first year for which data on the money supply is available from the Fed.

Table 2 shows the correlation between real M2 per capita and real GDP per capita.

Conclusion… real GDP per capita is affected more by M2 than by M1. Additionally, while the biggest effect occurs in real time, when a friend is running for re-election, changes in the money supply lead changes in real GDP per capita, perhaps because the Fed is pulling out all stops.

Note, however, that the Fed only influences M2 indirectly, through M1, which it controls. Thus, if it wants to affect elections, it will do so by moving M1 and hoping that this will result in changes in M2. So what do we see happening in the Greenspan years?

In 1988, when the Dukakis campaign committed seppuku, Greenspan didn’t do anything out of the ordinary. But in 1992, there was the second largest increase in real M1 per capita in the Fed’s history. (Granted, the Fed was dealing with a banking problem at the time, but a historically large increase was probably more than was called for.) In 1996, we saw the largest drop in real M1 per capita in the history of the Fed. In 2000, the third largest drop. In other words… at the time of three close elections, there were three almost unprecedented movements in the money lever the Fed controls. However, in each case, the big movements didn’t work. (in 1992, 1996, and 2000, Greenspan’s favored candidate lost the popular vote). So in 2004, he tried a different lever: 2004 saw the third lowest real Fed Funds rate (Fed Funds rate less inflation in the preceding 12 months) in history, -1.59. (By comparison, the rates in 2002 and 2003, years in which the economy was arguably in trouble, were -.86 and -.77 respectively. Historically, the only lower rates were in 1974 and 1975.)

So… either by coincidence, or for whatever other reason, Mr. Greenspan really pushed the levers at his disposal hard around election time.

My next post on the subject is something I haven’t addressed before. Taking into account another comment by Bill Polley (about there being a structural break in the 1990s), as well as further comments by PGL, I want to explain why Greenspan would have believed that moving M1 was going to cause bigger changes in real GDP per capita (and hence the elections) than they actually did. It also explains why Greenspan didn’t get the desired result, and provides another reason for him to have switched strategies in 2004.


Monthly M1
Monthly CPI
Quarterly population data which I then linearized into monthly
Quarterly real GDP per capita


As always, let me know if you want my spreadsheet.