75 Basis Points
As long as we’re talking about what I think the Fed should do rather than what I think it will do, why stop at 50 basis points? What I think the Fed should do is move – in one dramatic step – from an officially “restrictive” interest rate policy to an officially “neutral” one. As I argued in the last post, 4.75% is not yet neutral. 4.50% is still not my idea of neutral, but many people will accept it as neutral, and if the Fed describes it as “neutral”, the market will accept that description. With a financial crisis and declining employment, it makes no sense to continue with a “restrictive” policy.
I realize there is a theoretical advantage to small steps in monetary policy, but this past cycle has made me question whether that advantage exists in reality. The idea is to take a small step, observe the results, and then decide whether to take another small step. The problem, though, is that you don’t really get to observe many results until it’s too late. After the Fed started its “measured pace” tightening phase in 2004, there were almost no macroeconomic results to observe until 2006. The big result, ultimately, was that houses (at any given price) became harder to buy – and harder to hold on to if you had an ARM. That result is fairly obvious today – more than 3 years after the tightening phase began – but it was almost imperceptible as recently as the winter of 2006. Moreover it wasn’t until late summer of 2006 that we saw any indication of the effect that result is now clearly having on economic growth. So the idea of doing something and observing the results appears not to be very useful when it comes to the relevant time frames for monetary policy.
When it comes to financial results, as opposed to macroeconomic results, it’s not clear that gradualism even has a theoretical advantage, at least in an environment like today’s. To stop panicking, a market needs to see dramatic action. The Fed has made valiant attempts to undertake dramatic actions that don’t involve its interest rate target, but so far those actions apparently have not been dramatic enough. Apparently, you just can’t have a tight money policy and an easy money policy at the same time.
I think there is a more general theoretical case to be made against gradualism, at least when it comes to “first moves,” in which the direction of movement in the policy variable changes. It’s kind of like menu costs and Ss pricing. After the central bank has maintained a certain policy for a significant period of time, there is a cost to changing it. Indications that might, under other circumstances, argue for a small change in policy, are outweighed by the disruption that a “first move” in policy would cause, and by whatever other “menu costs” are associated with monetary policy. But at some point, the argument for change becomes so strong that it clearly outweighs these “menu costs”. At that point, the optimal policy change is necessarily a large one – because, if the optimal policy change were a small one, then the optimal policy (given the menu costs) would be no change at all.
To put it another way, under normal circumstances, the choice is either do something, do nothing, or sort of do something. Under those circumstances, sort of doing something (e.g., changing the funds rate target by such a small amount that the change can’t be expected to have much real impact) is often the best choice. But when the action being contemplated is a reversal in the direction of policy change, the option of sort of doing something doesn’t exist, because anything you do will be noteworthy. So if nothing is clearly the wrong thing to do – and that is surely the case for the US today – then for Heaven’s sake do something!
I realize there is a theoretical advantage to small steps in monetary policy, but this past cycle has made me question whether that advantage exists in reality. The idea is to take a small step, observe the results, and then decide whether to take another small step. The problem, though, is that you don’t really get to observe many results until it’s too late. After the Fed started its “measured pace” tightening phase in 2004, there were almost no macroeconomic results to observe until 2006. The big result, ultimately, was that houses (at any given price) became harder to buy – and harder to hold on to if you had an ARM. That result is fairly obvious today – more than 3 years after the tightening phase began – but it was almost imperceptible as recently as the winter of 2006. Moreover it wasn’t until late summer of 2006 that we saw any indication of the effect that result is now clearly having on economic growth. So the idea of doing something and observing the results appears not to be very useful when it comes to the relevant time frames for monetary policy.
When it comes to financial results, as opposed to macroeconomic results, it’s not clear that gradualism even has a theoretical advantage, at least in an environment like today’s. To stop panicking, a market needs to see dramatic action. The Fed has made valiant attempts to undertake dramatic actions that don’t involve its interest rate target, but so far those actions apparently have not been dramatic enough. Apparently, you just can’t have a tight money policy and an easy money policy at the same time.
I think there is a more general theoretical case to be made against gradualism, at least when it comes to “first moves,” in which the direction of movement in the policy variable changes. It’s kind of like menu costs and Ss pricing. After the central bank has maintained a certain policy for a significant period of time, there is a cost to changing it. Indications that might, under other circumstances, argue for a small change in policy, are outweighed by the disruption that a “first move” in policy would cause, and by whatever other “menu costs” are associated with monetary policy. But at some point, the argument for change becomes so strong that it clearly outweighs these “menu costs”. At that point, the optimal policy change is necessarily a large one – because, if the optimal policy change were a small one, then the optimal policy (given the menu costs) would be no change at all.
To put it another way, under normal circumstances, the choice is either do something, do nothing, or sort of do something. Under those circumstances, sort of doing something (e.g., changing the funds rate target by such a small amount that the change can’t be expected to have much real impact) is often the best choice. But when the action being contemplated is a reversal in the direction of policy change, the option of sort of doing something doesn’t exist, because anything you do will be noteworthy. So if nothing is clearly the wrong thing to do – and that is surely the case for the US today – then for Heaven’s sake do something!
Labels: economics, interest rates, macroeconomics, monetary policy
posted by knzn at 9/10/2007 09:09:00 AM
1 Comments:
Of course, your whole argument in this posting and before is founded on the assumption that there is a real causal relationship between Fed rate and economic activity, the business cycle, or inflation. That needed to be proven first. Recessions come, recessions go. It just is a basic phenomenon of capitalism. Whoever is in charge at the Fed, Greenspan, Bernanke, or whoever, no one can do anything to prevent it. I think too much emphasis is put in this whole discussion on the responsibility of individual persons who are in charge for the monetary policy with respect to what causes recessions or inflation and other symptoms of capitalism.
Post a Comment
<< Home