Is the Fed inflating asset prices?
By Scott Sumner
I recently spoke with some very smart people who work in the financial industry, and encountered a widely held view that asset bubbles are created by easy money policies. I think that view is wrong, but first let me acknowledge that there is a great deal of evidence in favor of that hypothesis:
1. In recent years, asset prices have often been unusually high by historical standards.
2. In recent years, the Fed has often adopted a low interest rate policy, with QE.
3. Event studies show that monetary easing announcements often increase asset prices.
4. Fed officials have cited asset prices as one of the ways that monetary stimulus can help during a depressed economy.
Put all that together and it sure looks like easy money is artificially inflating asset bubbles. However, I believe there is a better explanation for this set of facts.
First, let’s assume that basic economic theory is correct, and that the long run downtrend in (long-term) real interest rates is not caused by easy money, rather it reflects changes in the global economy related to saving rates in Asia, demographics, declining investment in traditional (capital intensive) industries, less construction of infrastructure, etc. Theory suggests that monetary policy does not determine long-term real interest rates.
Second, assume that assets are more valuable in a booming economy with high earnings than in a depressed economy. Then one can describe 4 states of the world:
1. High real interest rates, weak economy (early 1980s)
2. High real interest rates, strong economy (late 1980s)
3. Low real interest rates, weak economy (2002, 2009)
4. Low real interest rates, strong economy (today)
Think of the 20th century as a high real interest rate century and the 21st century as a low real interest rate century. Low asset prices occur in state #1, such as the early 1980s, medium asset prices occur in states #2 and #3, and high asset prices occur when real interest rates are low and the economy is booming, as is the case today.
In my hypothesis, monetary stimulus announcements do not boost asset prices by reducing long-term real interest rates, rather they increase the probability that the economy will be in the prosperous state going forward, not the depressed state. Another part of my hypothesis is that money has not been as “easy” during the 21st century as widely perceived, rather the low rates and QE policies mostly reflect defensive maneuvers to deal with the low interest rate environment caused by unusually low NGDP growth rates. Actual periods of extremely easy money (i.e. 1965-81) were not associated with asset bubbles.
My hypothesis has several advantages over the standard view:
1. The standard view is that the sharp fall in asset prices during 2007-09 was a sort of “correction” from bubble levels. My view predicts that asset prices would rise again to what look like inflated levels after the economy recovered. They have.
2. The standard view is that money has been easy throughout much of the 21st century. My view explains why we have not had the price inflation normally associated with easy money. The low interest rates and QE are misleading, and the stance of monetary policy is better described by NGDP growth rates.
3. The Fed’s attempt to restrain the stock market boom during 1928-29 failed until rates got so high (in September 1929) that they drove the entire economy into depression. Real rates rose to 7%. There was no monetary policy stance capable of popping an asset bubble without tanking the economy. The Fed has never again repeated this “experiment”.
Asset markets respond positively to easy money announcements that make a “bad economy” scenario less likely, which is 100% consistent with an efficient markets world with no irrational bubbles.