The Economy’s Not a Rock, and Paper Isn’t Killing It

Matt Yglesias writes that paper money kills economies because it institutes an effective lower bound on interest rates, the primary instrument of monetary policy. People like Miles Kimball have long argued that the “zero lower bound” and, hence, liquidity trap is basically a “policy choice” – a consequence of our cash economy. That is to say, under an electronic money system, we can easily implement negative nominal interest rates, giving monetary policy infinite power at all times.

The best way to think about this is imagine we all had dollar bills whose purchasing power was proportional to the length of the note, and the government can change the rate at which the note evaporates at will. If the rate of paper evaporation is increased, money demand will fall as consumers and investors across the world will throw cash for goods, services, and assets propping up a rapid recovery. In fact, the government doesn’t even have to increase the rate of evaporation, as long as it merely announces a credible intention to do so in the future.

But central banks can’t credibly announce the coming of negative interest rates, not the least because it isn’t physically possible in our current paradigm. Does that mean paper money is killing the economy…? I think not, for several reasons. Primarily, paper electronic money would be a superior choice to current monetary policy (and I, personally, have no animus towards accepting the future. Note that I’ve always hated carrying around cash, and don’t even have a wallet, so I love my electronic money, also known as Visa. But a lot of people use cash, and this would distributionally hurt them the most). But what we might call “conventionally unconventional” policies of quantitative easing are not out of steam.

Even Paul Krugman, the high priest of “monetary policy probably can’t gain traction in a liquidity trap”, agrees this has nothing to do with economic constraints of monetary policy as much as the conservatism of the central banking profession. That is to say, no one believes the central bank is sufficiently radical. In which case, we ask, what exactly is the point of militating for the abolishment of paper money – which is a far more radical experiment altogether?

Kimball – and maybe Yglesias – might argue that bringing it to the front of economic dialogue will make it easier for future crises to be handled effectively. But that would suppose getting rid of paper money is somehow fundamentally better than other unconventional options – given central banks were the right level of radical. We must make this stipulation because it makes no sense comparing a world where we are too timid to follow the policies proposed by Scott Sumner or Paul Krugman, but crazy enough to burn away paper.

And this contention is questionable at best. For example, even with negative interest rates, the central bank would need to set some sort of policy target. A lot of bloggers and forward-thinking economists like the idea of a nominal income target. I think Paul Krugman prefers a 4% inflation target (I don’t really know). If the latter – a higher inflation target – was the “optimal” solution between the two, then electronic money is most likely superior. The only point of a higher inflation target is to vastly reduce the risk of future recessions driving the equilibrium interest rate to the point of liquidity trap (and maybe reducing the effect of wage rigidities, but nominal targeting would accomplish this as well), and there’s some evidence that this is not optimal. (Edit: though still far, far better than current policy).

For reasons noted here, I don’t think a higher inflation target is superior to nominal income level targeting. In this case, it’s worth wondering whether negative interest rates really add a whole lot to the monetary policy arsenal, again, provided, markets didn’t think the central bank was skittish about its policies. The answer, I expect, is a probable “no” for reasons listed below:

• A nominal income target works as much through expectations of action as much as action. So once such a regime is credibly instituted, money velocity is unlikely to fall as much as it did even during a pretty serious crash. This doesn’t require lower interest rates.
• A proper quantitative easing plan can very easily provide above expectations with adequate firepower. What would such a plan look like? A symmetric Evans Rule. That is, for every month that the Federal Reserve misses its nominal income target, the scale of asset purchases are increased by some percent n. The exponential growth here would immediately convince investors that inflation is coming, stabilizing nominal income growth.

The only reason we would need negative interest rates under a nominal income target is if quantitative easing was somehow less welfare efficient than negative interest rates. This is likely to be the case if the central bank was forced to inject liquidity via the purchase of many private assets. However, with government debt levels where they are, this is not a significant concern, and can always be fixed with the issuance of higher maturity debt.

Another contention might be that it is easier to implement a “rules based” rather than “discretionary” monetary policy with the ability to vary interest rates below zero. While there is a sense that quantitative easing is discretionary, that only has to do with the conservative nature of the Federal Reserve, and there’s no reason why following a Taylor Rule when rates exceed zero, and an “exponential growth of asset purchases” when rates fall below zero is any more discretionary. It’s just as algorithmic, but with one extra condition. Die hard monetarists even want to institute a “nominal income prediction market” which would remove the discretionary element of maintaining the rule entirely.

Michael Woodford or Paul Krugman’s preferred monetary policy – “credibly promising to be irresponsible” – might be better achieved under negative interest rates but, even then, that’s not a clear conclusion. If central banks can’t credibly commit to keep interest rates at zero for a long time due to institutional conservatism, I’m not sure how we could a) abolish paper money, and b) credibly convince the market that we’ll bring them below zero for long enough. That the Bank of Japan increased interest rates whenever the economy was just about to recover is an example of this difficulty.

Anyway, if Bernanke could somehow promise that interest rates will be zero on January 01, 2017 there’s a very good chance we’d hit escape velocity from the liquidity trap. And that’s a less radical promise than getting rid of paper money altogether. (Can you imagine what Austrians and their ilk who dominate America’s policy minds on both sides of the aisle would say about NEGATIVE interest rates?) By the time we reach a political consensus that paper money is archaic, we’ll have a far better monetary policy to begin with by virtue of the fact that it is less radical. (Not to mention the fact that in modern markets simply holding a foreign currency won’t be a bad choice, either).

I sometimes feel arguments for a negative interest rate are substituted in place of arguments for a better policy target. But the latter is the important debate, and the former is only a tool to achieve the latter. If an optimal target necessitated negative interest rates, I’d be writing about them every day. But I’ve yet to see an argument why a nominal income level target requires negative interest rates. Importantly, the central bank can technically levy a penalty on excess reserves which has the same benefits.

Paper money is killing the economy, but so is central bank conservatism. The former is just a symptom, the latter a disease.