Wednesday, 29 September 2010

Message to the Fed: this is the modern monetary theory (MMT) transmission mechanism.

It seems a number of Fed and ex-Fed people have at last tumbled to the fact that there are serious problems with traditional ideas on money supply transmission mechanisms.

This Fed paper to which Warren Mosler draws attention points to the fact that there has been a 2,173% rise in bank reserves in the last two years, with (contrary to text book predictions) almost no effect. Yes you read that right: 2,173%.

The authors of the Fed paper also say “if the quantity of reserves is relevant for the transmission of monetary policy, a different mechanism must be found.”

And then there is Arnold Kling, who worked for a time as an economist for the Fed. He says “I am having an equally hard time understanding modern monetary theory.”

In view of the above, it might be helpful to set out the MMT transmission mechanism. It is very simple and it’s thus (as I see it).

1. The government-central bank machine net spends in a recession. Assuming government wants the relative sizes of public and private sectors to remain constant, some of the money will be spent on hiring extra public sector workers and/or making sure that tax shortfalls don’t result in public sector workers being sacked. That creates employment.

2. There is a multiplier effect from “1”, that is, part of the above additional payroll costs will be spent, which in turn employs more people.

3. The above additional public sector workers will probably save some of their income. That boosts private sector savings. Those savings will not expand for ever. The point will come where the private sector thinks it has enough by way of savings, at which point it will cease saving and will spend, or try to spend all its income. That employs yet more people.

4. While the above boost for the public sector boosts the private sector INDIRECTLY, it is probably desirable to give the private sector a DIRECT boost as well. That can be done by, for example, reducing payroll taxes or income tax. That boosts employees’ take home pay, which in turn will boost their spending, which employs more people.

Doubtless some of the above private sector increase in take home pay, will be saved. The consequences are exactly the same as with the above public sector employees. That is, savings will rise to the point where no more savings are required, etc, etc.


  1. Where's the evidence for the fiscal multiplier effect?

    John Maudlin states “I can’t think of a more tired old theory than the Keynesian notion that $1 of additional government spending will generate $1.5 of real GDP. This ‘multiplier effect’ is taught in every introductory macroeconomic textbook. Yet, it is both theoretically and empirically questionable.”

    He then states "the supporting evidence for my view [is] in an OECD working paper by Roberto Perotti titled “Estimating the Effects of Fiscal Policy in OECD Countries,”

    Is there opposing evidence?

  2. Neil: the explanation for the difference of opinion on the multiplier could be thus. My points above were based on the assumption that government “net spends” to use my actual phrase. This was supposed to imply what you might call an Abba Lerner scenario where government just prints money and spends it (as opposed to what you might call a Keynes scenario, where government borrows money and spends it).

    Assuming a Lerner scenario, I don’t see how the multiplier can be denied. That is, extra people employed are bound to spend a portion of their extra income, which in turn employs more people.

    In contrast, in a Keynes scenario, it is possible there is no multiplier because the borrowing cancels out not just the multiplier, but all the stimulatory effects of “borrow and spend”.

    On this subject I’ve just sent an article off to Munich RepEc attacking Keynes for the above sort of reasons. The article is called “The Flaws in Keynesian Borrow and Spend” and should appear there in the next week or so. And the above anti-stimulatory effects of government borrowing is one of the potential weaknesses in “borrow and spend” which I point out.


  3. As Mosler says, it doesn't matter if the fiscal multiplier is smaller. That just means you can cut taxes more. but many have found evidence of the fiscal multiplier, even of late. See the paper by Zandi in 2008, for instance, or even CBO. Basically, as long as the multiplier is greater than 0, MMT's transmission mechanism works.

  4. Anon: strikes me the MMT transmission mechanism works even if the multiplier WAS zero. By “multiplier”, I mean the fact that those who get jobs increase their weekly spending, which in turn employs yet more people. As I pointed out above, the transmission works via several channels. I.e. even if the multiplier is zero, you’ve still got channels 1 and 3 plus “4 less the multiplier element in 4”.


Post a comment.