Monday, 25 April 2016
There’s no place for interest rate adjustments.
I had a discussion with George Selgin recently on this topic in the comments here.
My basic point was that when there’s a need for stimulus, there is no particular reason to impart stimulus via more borrowing, lending and investment than there is to impart it by artificially increasing the production and sale of ice-cream, lollipops, cars, education or any other narrow selection of goods. Thus interest rates should be left to market forces, as should the price of most items.
Milton Friedman argued that government should not interfere with interest rates. See para starting “Under the proposal..” here.
Warren Mosler advocated the same. See final two paras here.
In other words stimulus should come in the form of expanding the output of as wide a range of goods as possible. The only exception to that might come if a recession is caused by a dramatic and irrational fall in the output of some particular set of goods or services, in which case subsidising the output of that set of goods could be justified.
It could be argued that the recent recession was sparked off by a “dramatic and irrational fall” in bank lending. Well the fall was certainly dramatic, but it’s not clear that it was “irrational”. What happened was that banks realized that a significant proportion of their loans were no good, and cut back on such lending: entirely rational!!
To put it more bluntley, the crises was sparked off by excessive and irresponsible lending, and the solution advocated by all those clever professional economists was to cut interest rates so as to encourage more lending: the phrase "raving bonkers" springs to mind.
Another potential excuse for interest rate adjustments is that the lag between the decision to impart stimulus and stimulus actually arriving might be shorter in the case of interest adjustments than other forms of stimulus. Or the PREDICTABILITY of the effect of interest rate changes might be better. But that doesn’t seem to be the case.
So why do we have interest rate adjustments? Well probably the main reason is that we don’t want politicians having the last word on stimulus, i.e. we don’t want politicians in charge of the printing press. So we have a system under which politicians CAN IMPLEMENT SOME STIMULUS: they can borrow more and spend more. Economists are generally agreed that that’s stimulatory assuming the central bank arranges for interest rates to remain stable: i.e. arranges for the extra lending not to push up interest rates.
But note that the ultimate say on stimulus under that system rests with central banks.
Actually the evidence seems to be that if politicians DO HAVE access to the printing press (i.e. have the last word on stimulus) the results are not too bad – with Robert Mugabe being the obvious exception. That is, the evidence seems to be that there isn't much to choose between independent and non-independent central banks. See the first chart here.
However, and to repeat, the consensus is that independent central banks are better because the last word on stimulus is then in the hands of professional economists rather than politicians, thus central banks have to be given some sort of tool or weapon to actually impose their will, and what better than control over interest rates?
But the obvious problem there is the one set out at the start of this article above, namely that interest rate adjustments are a defective tool. So how do we get round that apparent contradiction? Well it’s not too difficult. In fact the solution was set out in this work a few years ago.
The solution (assuming we’re going to have some sort of independent committee of economists have the last word on stimulus), is to have such a committee determine BOTH monetary AND fiscal stimulus. Whether that committee is based at the central bank or the treasury doesn’t matter too much. Those who want the committee kept as far away from politicians as possible will want it based at the central bank of course.
As to how we actually combine monetary and fiscal stimulus, that’s easily done: just fund the extra public spending or tax cuts that make up a stimulus package from new money issued by the central bank. I.e. when there’s a need for stimulus, just have the state print extra money and spend it or cut taxes.
As to whether a stimulus package consists primarily of more public spending or more tax cuts, that’s clearly a POLITICAL decision which should stay with politicians. And that’s exactly what’s advocated in the above work: that is, the committee of economists decides on the SIZE OF stimulus package, while POLITICIANS decide whether the package consists mainly of extra public spending or mainly of tax cuts.
And finally, having poured cold water on interest rate adjustments, I wouldn't rule them out altogether in an emergency. But they're certainly not my first choice.