Recent Posts

Wednesday, January 23, 2019

If You Want To Understand MMT...

Brad DeLong has written a primer on Modern Monetary Theory (MMT), and just gave an important lesson about MMT: if you want to understand MMT, you need to read a primer written by a MMTer. I am not going to hold myself out as an expert on DeLong's thinking, but I would argue that a fear of bond market disruptions is a constant theme in his writings. The MMT analysis of bond markets is superior to his rather vague description. (H/T Mike Norman Economics for the link.)

To be blunt, I largely skipped over his article and I am only focused on this passage:
So what can go wrong with MMT?
Three things can go wrong:
1. MMT implicitly assumes that the debt market is efficient—that if the government debt gets on an unduly burdensome and unsustainable path, we will see that immediately in high interest rates. If that is not true, the government and the economy can face one hell of a mess should a bubble in government bond prices develop and then collapse. Cf. Greece. (Note: he lists a couple of other points.)
This is a rather unusual characterisation of MMT: one of the possible policy recommendations under the MMT umbrella is the suggestion that nominal interest rates be locked at 0%, and bond issuance suspended. (See technical appendix below.) The reasoning behind this proposal is that interest rate policy is weak, and its effects unknown, so there is no real cost to locking the nominal rate at 0%. The advantage of this policy is that eliminates the belief that fiscal policymakers need to care even the slightest what the "bond vigilantes" think.

To be clear, the belief that interest rate policy is ineffective is not a complete consensus within MMT (unlike the Job Guarantee), so it is not completely safe to say that a permanent 0% interest rate policy is a key MMT policy stance. That said, the analysis of the pros and cons of the policy is as I described above, which is distinctly different than neoclassical economics -- where interest rate policy is assumed to be extremely effective for steering the economy.

Under a 0% nominal interest rate policy, there is no real way to have an "unsustainable debt trajectory": debt will just grow in line with the fiscal deficit. (The existing stock of debt will have a fixed interest cost (modulo inflation-linked bonds), which will be run off as that debt stock matures.) The only way that a high real rate can be created is by the private sector undergoing severe deflation. That's right: in order to thwart allegedly loose fiscal policy, the private sector will massively cut nominal prices! That will show those Keynesians!

Admittedly, the yields on long-dated bonds can rise, which would inconvenience private sector borrowers who want to borrow long-term fixed. Tough luck; they will need to borrow at short/floating rates, which will be crowded near the 0% policy rate. Sooner or later, the unbalanced demand for duration need to match actuarial liabilities will win out, and the curve will return to normality. In any event, by stopping the issuance of benchmark bonds, the government is saying that term interest rates are the private sector's problem.

When can problems occur? The MMT analysis is straightforward: we can only have debt problems in systems where the authorities deliberately allow themselves to blackmailed by bond markets. DeLong uses Greece as an example, which is appropriate: the entire euro project was designed to make fiscal policy subservient to "the markets." This is a policy choice, which DeLong appears to agree with.

In other words, we might need to be somewhat cautious in the context of current monetary operating procedures (which is a signature area of study of MMT). However, as Japan shows us, those constraints are not particularly tight. Furthermore, there is nothing stopping us from changing those operating procedures. This insight makes DeLong's statements about the risk of bond market collapse totally unlike MMT's views on the matter.

One of the standard putdowns of MMT I see on Twitter is that it offers nothing new. Brad DeLong's text provides yet another example of statements by a mainstream economist which are "out of paradigm" with MMT.

Concluding Remarks

The debate between MMT and the mainstream is mainly a battle of assumptions. The advantage of MMT is that it is clear what the assumptions are.

Technical Appendix: The 0% Policy Target

Saying that the nominal rate would be locked at 0% is a slightly simplification. In practice, it would probably be around 0.25%.

The government would issue three types of liabilities:
  1. notes and coin (with a 0% interest rate);
  2. deposits at the central bank ("reserves" in American parlance), with a statutory 0% interest rate;
  3. Treasury bills that are issued at fixed price, unlimited size auctions, with the rate (price) capped at 0.25% annualised.
The private sector would allocate between base money (currency and reserves) and the 0.25% bills based on portfolio needs.

As for debt sustainability, it just means that interest costs will be at most 0.25% of GDP. This is a rounding error if policy aims to deliver nominal growth of at least 4% per annum.

(c) Brian Romanchuk 2019


  1. I'm curious what the effect of he The 0% Policy Target would be on bank lending. They would find many qualified customers at that rate and one would imagine that this would be expansionary and inflationary. Would MMT then require that this be addressed (as DeLong suggests) by reducing the govt deficit ? If not, how does the 0% Policy Target play our for the banking sector ?

    1. (1) The 0% is the t-bill rate, not the borrowing rate for the private sector. Borrowers will face a credit spread.

      (2) Japan and the U.S. ran ZIRP for years, and lending was hardly exuberant. The MMT argument is straightforward: the effects of a 0% policy rate is mixed. Although borrowing desires presumably increase, this is offset by the loss of interest income. Based on the Japan/US experience, the income loss may be more important. The mainstream just *assumes* that a low a policy rate is expansionary; I'd love to see some plausible evidence for that assumption.

      Arguably, this is a point of debate. But that is the real debate, not assertions about mythical bond vigilantes.

    2. I assume that MMTers would reject it but the obvious reason would be the Wicksellian one that at a 0% risk free rate there would be more borrowers than lenders and the difference would be made up by banks creating new money to - which (the argument goes) leads to an ever rising price level unless fiscal policy is used to tax the new money back again.

    3. lol MMT rejects the natural rate. This is a major part of the literature.

  2. I also wonder about the impact of a true permanent zero rate policy. Intuitively I would think this would result in risk free yields dropping close to zero across the yield curve. My thinking is that high long-term risk free yields would lead to an arbitrage profit. We know that forward prices can be derived from the spot rate and a chain linking of forward rates. For example, it should cost the same to borrow for 5 years as it would to borrow for 1 year and agree to borrow for 4 years 1 year from now. If traders KNEW the short-term spot rate would ALWAYS be zero, the rate on one-day forwards should be zero at any point in the future. This means that if long-term risk free yields increased substantially, a trader could obtain a risk free profit by borrowing short and chain linking one day forwards to some point in the future.

  3. This is a good analysis

    superior to DeLong's

    in fact

    superior to MMT's

    because of this :

    "To be clear, the belief that interest rate policy is ineffective is not a complete consensus within MMT (unlike the Job Guarantee), so it is not completely safe to say that a permanent 0% interest rate policy is a key MMT policy stance."

    MMT would be a materially stronger force in actual policy circles if its key people closed ranks and adopted 0 rates no bonds as an unambiguously united policy position

    as it is, the policy and its analysis is diluted, and pretty much wasted as a real world consideration

  4. My answer to DeLong's unsubstantiated three implicit assumptions which MMTers allegedly make, is that De Long implicitly assumes the Earth is flat, that the Sun revolves round the Earth and that the Moon is made of cheeze....:-)

  5. This comment has been removed by the author.


Note: Posts may be moderated, and there may be a considerable delay before they appear.

Although I welcome people who disagree with me, please be civil.

Please note that my spam comment filter appears to dislike long "anonymous" posts.

Note: if you want to post comments from Apple devices (iPhone, iPad), you apparently need to turn off "prevent cross-site tracking" in Safari privacy settings. (The reason presumably is that another URL handles comments, and so the user session needs to be preserved when redirected to that site. I don't like this, but this is not enough to make me switch my hosting service.)