Tuesday, November 26, 2013

Guest Post: Ralph Musgrave — Scott Sumner, MMT, and irrational expectations

Scott Sumner, MMT, and irrational expectations
Ralph Musgrave

Sumner and MMTers don’t see eye to eye. He criticises MMT on his blog from time to time. And Randall Wray had a go at Sumner recently here and here. Anyway, I want to demolish an idea put by Sumner (and indeed many other economists). It’s that rational expectations / Ricardian nonsense. Bill Mitchell described Ricardianism as an idea from La-la land, and quite right. And Joseph Stiglitz said “Ricardian equivalence is taught in every graduate school in the country. It is also sheer nonsense.”

One nonsensical element in “rational” expectations is that economists like Sumner ascribe to households and firms expectation type ideas that are sometimes completely IRRATIONAL. Plus, more often than not, those economists don’t provide any empirical evidence that households and firms actually adhere to those ideas or expectations. A classic example of this is where Sumner claims that helicopter drops will be ineffective because everyone expects the policy to be reversed at some stage via tax increases. As he puts it, “the injections are not expected to be permanent”. Thus the private sector will supposedly hoard it’s helicopter money so that it can pay those taxes. Thus, so Sumner claims, heli-drops have no effect.

Now if government behaves in a rational manner, it won’t withdraw those “injections” at any old random point in time: the injections will be left in place as long as they’re needed. In fact the only reason to withdraw the injections, i.e. run a surplus, is when the private sector gets too confident, and inflation looks like becoming excessive. Put another way, if government behaves rationally, it will withdraw surplus monetary base from the economy only when that withdrawal controls inflation rather than actually reduces demand in real terms, or reduces incomes in real terms. So the typical private sector agent, if they’re 100% rational, and 100% clued up on central bank operations, deficits, etc will not hoard helicopter money. Quite the reverse: they’ll up their spending.

Of course the idea that the typical household or small firm is “100% clued up on central bank operations, deficits, etc is straight out of cloud cuckoo land. But it’s something of that sort that pro-Ricardian economists presumably have in mind when they refer to “rational expectations”. So let’s run with this cloud cuckoo land idea for bit. (I’ll abandon the idea shortly.)

Another point that 100% clued up private sector agents will understand is that it’s impossible for the private sector in the aggregate to get rid of monetary base until the government / central bank machine decides to withdraw it from the private sector. Thus the average or typical private sector entity will realise that when they do up their spending, they won’t actually lose their stock of monetary base because everyone else will be upping their spending as well, so the net effect is that the typical private sector entity’s stock of base remains constant.

Conclusion so far: 100% rational households and firms will not hoard their stock of helicopter money to such an extent that heli-drops are ineffective.

Of course, and to repeat, the idea that the typical household or firm is 100% clued up on central bank operations, deficits, fiscal stimulus, etc is straight out of La-la land, to use Bill’s phrase. And the empirical evidence seems to be pretty much in line with common sense, namely that when households and firms notice an increase in their incomes, they immediately up their spending by a significant amount. Certainly the evidence in relation to tax rebates is that households spend a significant proportion within a year. E.g. see here, here and here.

Of course tax rebates are not exactly the same as distributing helicopter money. But they’re FAIRLY SIMILAR. Plus at least I’ve provided SOME SORT OF evidence to back my points, which is more than most Ricardian enthusiasts do.

And finally there’s a technical point on monetary base I’d better address, as follows. I referred above to non-bank private sector’s stock of base, which conflicts with the popular belief that non-bank private sector entities don’t have access to monetary base. The reality is that if I get helicopter money in the form of a check for $X, I deposit that at my commercial bank, which in turn has it’s account at the Fed credited. And I can do whatever I want with that money. So in effect I do have access to monetary base: it’s just that some commercial bank acts as agent for me when I want my stock of base paid to someone, or if I want to withdraw it in the form of physical cash.


Brian Romanchuk said...

The monetary base (under most definitions that I have seen) includes notes and coin, and so most people do have a piece of the monetary base sitting under their couch cushions.

In mainstream models, fiscal policy is essentially non-existent. It's an exogenous variable, to use economics slang. This means that agents in the model would not expect it to be used to control inflation, like you suggest. Ricardian equivalence has to be asssumed mathematically in order for them to find a solution (otherwise, they are missing a mathematical constraint on the possible solutions).

I am working on a series of posts about fiscal policy, and I will be covering the problems DSGE models have with modelling it.

y said...

"So in effect I do have access to monetary base: it’s just that some commercial bank acts as agent for me when I want my stock of base paid to someone"

When you deposit base money in a bank you're lending it to them. They give you a promise to pay call a deposit. The relationship between bank and depositor is debtor-creditor, not fiduciary.

Ralph Musgrave said...


I’m puzzled by your claim that “agents in the model would not expect it to be used to control inflation..”. Surely the fact is that both fiscal and monetary policy HAVE BEEN used in an attempt to regulate demand and inflation. As Sumner put it at the start of his article, “A helicopter drop is economist lingo for a combined fiscal/monetary stimulus. Trillion dollar deficits combined with trillion dollar QE. We’ve done that…”

Derek said...

Please don't call it "Ricardianism", Ralph. Ricardo thought it was just as stupid an idea as you and I do. It does him a disservice to associate him with a concept which he argued against so successfully that it was seen as nonsense until the late twentieth century.

If you want a villain, Robert Barro was the man who re-introduced it to modern economics, so "Barrovianism" would be a much more suitable name for it.