The Fed, Presidential Elections, and Coincidence – Part 9
This may be the last post, at least for a while, in the series I’ve been working on for a while. I think it’s a good summary (using data from 1959 to 2006).
Table 1 below looks at the average one year % rate of change in the real GDP per capita (4Q), the average one year % rate of change in real m1 per capita (November to November), the average one year % rate of change in real m2 per capita (November to November), and the average real fed funds rate (November). It does so for years in which there is no Presidential election, for years in which there is a Presidential election with no incumbent, for years in which there is an election in which the incumbent is from a party different than the Fed chair, and for years in which there is an incumbent from the same party as the Fed chair.
Table 1 shows:
1. Economic growth rates are fastest in years in which there is an election and a “friend” (i.e., someone from the same party as the Fed chair) is running for re-election.
2. Economic growth rates are slowest when an “enemy” (someone from a different party than the Fed chair) is running for re-election.
3. Monetary policy (is most favorable to growth) when a friend is running for re-election.
4. Monetary policy (is least favorable to growth) when an enemy is running for re-election.
Now, if the Fed was trying to smooth out the economy, which supposedly is its purpose, presumably when growth was fastest, it would be trying to slow down the economy, and when growth was slowest, it would be trying to speed up the economy. That is not what we see. Furthermore, recall that as shown in an earlier post, in years when a friend or enemy is running for re-election (columns 3 and 4), monetary policy is correlated with lags in the economy – hence, the economy is following the dictates of the Fed.
OK. What about when the economy is moving along at a nice clip. Does the pattern still hold? Table 2 is the same, but it looks only at years for which the annual percentage growth rate in real GDP per capita is at least 2%.
In the first two columns, we see the Fed trying to dampen the economy relative to Table 1, which presumably is its job. But note… Despite the fact average growth rates are higher in columns 1 and 2, column 3 has the most restrictive monetary policy by two out of three measures (real m1 per cap, real m2 per cap, but not fed funds rate). And how restrictive? It is enough to point out that the one observation was 1996 – when Clinton was running for re-election, and negative 10.293% is the biggest decrease in real M1 per capita in any year in the sample.
Notice also… while column 4 still has the fastest growing economy, as the other columns become more comparable, the difference in the monetary policy really stands out. The fourth column, for instance, is the only one in which real M1 per capita actually increased, real M2 per capita increases by four and a half times what it does in the next fastest column, and the Fed Funds rate is only 60% as high as in the next highest column. This is no accident or margin of error. This is very deliberate, very precise.
What about years when the real GDP per capita shrinks? Table 3 below looks at that.
In the first two columns, we now see faster growth (or less shrinkage) of real Money per capita, and we see lower fed funds rates, exactly as should be the case. Overall, the Fed doesn’t seem to show any more mercy to an enemy in trouble than an enemy that isn’t, and column 4 has disappeared.
From this, we learn: When there is no Presidential election, or there is a presidential election in which no incumbent is running for office, the Fed seems to behave precisely as we expect. When the economy is running too quickly, it cuts back on the fuel, and when the economy needs help, the Fed makes money cheap to try to jumpstart things.
Things are very different when the Fed feels it has a dog in the fight. When an enemy is running for re-election, the Fed uses monetary policy to slow down the economy, regardless of whether the economy is growing quickly or slowly. This is not just to the detriment of the candidate, but to the detriment of the American public as well. Because we feel poorer, the likelihood we will vote to re-elect the incumbent decreases.
On the other hand, when a friend is running for re-election, the Fed runs the loosest economic policy possible. That makes us all feel wealthier, and thus more likely to re-elect the incumbent. But it increases the likelihood of inflation – otherwise, the Fed would behave that way all the time, not just when its friends could use the help.
—————————————————
Data:
Monthly M1 and M2
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.