Raghuram Rajan has come out swinging against U.S. monetary policy. He argues it is wrong to look to the Fed as some monetary wizard who can "revive the economy through a swish of the monetary wand." He also believes the Fed's monetary stimulus causes more problems that it solves. In particular, he views the Fed's low-interest rate polices causing excessive risk taking by investors and harm to savers. On the surface his arguments are consistent with his belief that the housing and credit boom was similarly driven by monetary policy that was too easy back in the early-to-mid 2000s. I am sympathetic to his views on the Fed's role in the housing and credit boom, but believe his current take on U.S. monetary policy is off. Let me explain why.
First, a low policy interest rate target by itself does not mean monetary policy is loose, let alone too loose. Interest rates have to be low relative to the neutral (or the natural) interest rate to be stimulative. The neutral interest rate, in turn, is closely tied to the performance of the economy. Thus, given the anemic economic conditions the neutral interest rate is most likely low now. Estimates of the real neutral interest rate show this to be the case. For example, the updated estimates for the highly cited Laubach-Williams paper puts the real neutral interest rate at 0.27% for 2010:Q4. (Aside: I know John Williams is now busy running the San Francisco Fed, but us bloggers need him to keep updating his natural rate estimates!) Another way of saying this is that interest rates would probably be low right now even if there were no Federal Reserve. Savers, therefore, are not being harmed by the Fed's low targeted policy interest rate, but by the weakened economy.
It is true that the real federal funds rate is probably still slightly lower than the real neutral rate, but that by itself still does not mean monetary policy is too loose if it is being eased to close the output gap. The Taylor Rule, for example, prescribes a federal funds rate target different than the neutral federal funds rate if there is an output gap or if inflation is running too high. Now, I will concede it is difficult, maybe even impossible for the Fed to finely tune its targeted interest rate so as to flawlessly close the output gap or reign in inflation. But that is not the issue here. The point is one has to be careful about claiming low interest rates necessarily mean excessively easy monetary policy.
So what can and should monetary policy do? First, monetary policy can make a big difference even in a 0% interest rate environment. FDR showed us that with his QE program of 1933-1936. By most accounts it was a smashing success. Unlike our current QE programs, it worked because it change nominal expectations in a meaningful manner. FDR effectively created a price level target that convinced everyone he would return the price level to its pre-crisis value. This caused money demand to fall and nominal spending to take off. Given the large amount of excess slack, this increase in nominal spending caused a sharp recovery in real economic activity. The same could be done today if the Fed would announce an explicit level target, preferably a nominal GDP level target. Such a target would create a period of rapid catch up spending--to return spending to its trend--and thereafter stabilize the growth rate of nominal spending around some target growth rate. If this happened, the neutral interest rate would rise, investors would be less interested in risky investments, savers would benefit, and there would be long-term certainty about the path of nominal spending. In short, Rajan's concerns would be eliminated. Rajan just needs more faith in the right kind of monetary policy can do.
Now if, on the other hand, the Fed were simply do another round of QE without an explicit level target I doubt it will help much. Such an approach does not shape expectations well, is a lightning rod for criticism, and ultimately could add more uncertainty to markets. We need to move toward more systematic monetary policy, one that adds plenty of monetary stimulus but does so in a predictable manner. What we need is a nominal GDP level target.
Update: Brad DeLong weighs in and Bill Woolsey responds in his comment section.
It is true that the real federal funds rate is probably still slightly lower than the real neutral rate, but that by itself still does not mean monetary policy is too loose if it is being eased to close the output gap. The Taylor Rule, for example, prescribes a federal funds rate target different than the neutral federal funds rate if there is an output gap or if inflation is running too high. Now, I will concede it is difficult, maybe even impossible for the Fed to finely tune its targeted interest rate so as to flawlessly close the output gap or reign in inflation. But that is not the issue here. The point is one has to be careful about claiming low interest rates necessarily mean excessively easy monetary policy.
So what can and should monetary policy do? First, monetary policy can make a big difference even in a 0% interest rate environment. FDR showed us that with his QE program of 1933-1936. By most accounts it was a smashing success. Unlike our current QE programs, it worked because it change nominal expectations in a meaningful manner. FDR effectively created a price level target that convinced everyone he would return the price level to its pre-crisis value. This caused money demand to fall and nominal spending to take off. Given the large amount of excess slack, this increase in nominal spending caused a sharp recovery in real economic activity. The same could be done today if the Fed would announce an explicit level target, preferably a nominal GDP level target. Such a target would create a period of rapid catch up spending--to return spending to its trend--and thereafter stabilize the growth rate of nominal spending around some target growth rate. If this happened, the neutral interest rate would rise, investors would be less interested in risky investments, savers would benefit, and there would be long-term certainty about the path of nominal spending. In short, Rajan's concerns would be eliminated. Rajan just needs more faith in the right kind of monetary policy can do.
Now if, on the other hand, the Fed were simply do another round of QE without an explicit level target I doubt it will help much. Such an approach does not shape expectations well, is a lightning rod for criticism, and ultimately could add more uncertainty to markets. We need to move toward more systematic monetary policy, one that adds plenty of monetary stimulus but does so in a predictable manner. What we need is a nominal GDP level target.
Update: Brad DeLong weighs in and Bill Woolsey responds in his comment section.
Excellent commentary.
ReplyDeleteI still contend that "bubbles" are not caused by the Fed. How does the Fed-bubble idea square with efficient market theory? Should not all asset classes balloon just a bit, rather than one asset class irrationally bubbling?
But in general, I like the NGDP targeting policy.
Side note: You have some right-wingers calling for 5 percent real GDP growth, and John Taylor saying it is doable. That would call for massive Fed accommodation.
Not sure how you would get real 5 percent GDP growth and follow the Taylor rule.
Another way of saying this is that interest rates would probably be low right now even if there were no Federal Reserve.
ReplyDeleteInterest rates would absolutely be low right now even if there were no Federal Reserve.
The Fed is an interest rate follower, not a leader. Interest rates are set by the market. Check Fed Funds and 3 mo T-bill rates. Fed Res has data back to '54. I am not making this up.
All asset classes have been part of the serial bubbles since the late 90's. This in the irrationality that puts the lie to EMT.
Pawlenty's 5 percent GDP growth sustained for a decade is beyond reason, fiction, or fantasy. And auterian Rethug policies (Pawlenty hasn't explained anything in specific) would kill whatever scant hope for recovery we might have. The only time we ever had that kind of sustained growth was during the New Deal and WWII. Taylor has taken leave of his senses.
Alas,
JzB
You are absolutely right in saying that "a low policy interest rate target by itself does not mean monetary policy is loose, let alone too loose".
ReplyDeleteA low interest rate could mean either of two things a) Both the demand and supply of money are low b) The demand for money is high and is being accommodated by a high supply of money.
To decide which it is you need a measure of money. And that is exactly what is missing in modern macroeconomics.
Take a look at http://www.philipji.com/item/2011-06-09/ben-bernanke-sorry-we-did-it-again to check out an accurate measure of money. Observe that the supply of money has been rising drastically since the third quarter of 2009.
The coming recession will make The Great Recession look like a picnic.
When that happens you can come back and see how Corrected Money Supply in the graph above is calculated.
Beckworth: "The point is one has to be careful about claiming low interest rates necessarily mean excessively easy monetary policy."
ReplyDeleteIsn't the relevant question today, then: Why is the neutral (or the natural) rate of interest so low?
And then, since "The neutral interest rate, in turn, is closely tied to the performance of the economy," the question becomes: Why is the performance of the economy so low?
I think that until this question is answered, it is too soon to fall back on the Taylor Rule or Fed targets or any other solution.
ArtS