Before Xmas, I published a two-part reply to Gregory Mankiw’s paper on Modern Monetary Theory (MMT) – A Skeptic’s Guide to Modern Monetary Theory (December 12, 2019). I was trying to get the response finished before the break and Part 2 had already become too long. So I decided to leave one issue that I didn’t get to address for a shorter third response once service resumed. I think this part of the response is necessary to set right on the public record. It exemplifies how critics need to work harder to actually understand what MMT is about. And while they try to claim that MMT is opaque and difficult to get to terms with, thereby sheeting the blame for their misguided renditions of our work back onto us, the issue I discuss today is very easy to come to terms with. It is front and centre and there have been many scholarly and other articles written about it. I refer, of course, to the Job Guarantee as MMTs response to the mainstream Phillips curve. The failure to appreciate where this sits in the MMT framework is not confined to mainstream economists. But this group know all about the Phillips curve literature and the place it holds in their macroeconomics. So there is no excuse not to understand it within a buffer stock framework and how MMT responds.
1. A response to Greg Mankiw – Part 1 (December 23, 2019).
2. A response to Greg Mankiw – Part 2 (December 24, 2019).
Since Part 2 was published last week, there has been further correspondence with Greg (on his initial instigation), which clarified some of the unknowns laid out in Part 1.
He assured us that he was acting in good faith (which I accept) and had “spent several months reading … [our] … book carefully”. He also said his initial intention when he wrote to us initially was different to what eventually was produced in the “Skeptic’s Guide”.
The shift was driven by deadlines.
While I have no reason to doubt any of that, I still think it is beholden on an academic critic to render the object of criticism fairly.
I don’t think his “Skeptic’s Guide” is a fair rendition for reasons I outlined in Part 2 and will extend in this part today.
In fact, his readers will come away from reading his paper with a very weird version of our work. I hope they are curious and come back to source to learn how they have been mislead – intentionally or otherwise.
So here is one way his readers were totally mislead about MMT.
Somehow Greg Mankiw missed a major departure of MMT from the mainstream he loves
In his critical ‘guide’ (or non-guide as it happens), Greg Mankiw writes:
MMT proponents advance a very different approach to inflation. They write, “Conflict theory situates the problem of inflation as being intrinsic to the power relations between workers and capital (class conflict), which are mediated by government within a capitalist system.” (MW&W, p. 255) That is, inflation gets out of control when workers and capitalists each struggle to claim a larger share of national income. According to this view, incomes policies, such as government guidelines for wages and prices, are a solution to high inflation. MMT advocates see these guidelines, and even government controls on wages and prices, as a kind of arbitration in the ongoing class struggle. (MW&W, pp. 264-265)
This is a totally inaccurate account of the MMT approach to inflation control and reflects badly on his claim to have read our – Macroeconomics – textbook faithfully before writing about it.
In Chapter 17 on ‘Unemployment and Inflation’, we do consider incomes policies (Section 17.5) in the context of inflation being the outcome of the distributional struggle over real income shares by labour and capital, fought out by each claimant pushing nominal wages and/or prices up.
We say that:
Governments facing a wage price spiral and who are reluctant to introduce a sharp contraction in the economy, which might otherwise discipline the combatants in the distributional struggle, have from time to time considered the use of so called incomes policies.
Progressive economists often advocate the use of incomes policies to rein in cost pressures to avoid the need to reduce overall spending, which creates higher involuntary unemployment.
We document historical episodes in various countries of attempts to use incomes policies to control inflation.
We conclude that, in general, that historical trends (institutions etc) have made the “operation of incomes policies difficult” despite this approach have been successful in some nations (for example, Scandinavian economies).
Nowhere in that discussion do we say that the MMT approach to inflation is to use incomes policy.
In fact, we closed the discussion with this:
In Chapter 19, we will introduce the concept of employment and unemployment buffer stocks in a macroeconomy and analyse how they can be manipulated by policy to maintain price stability.
And in Chapter 19, at the outset, we reflect on the material in Chapter 17 before pursuing the “main focus” which is to analyse:
… two approaches to achieve sustained low and stable inflation (inflation proofing). We construct the discussion in terms of a comparison between two types of buffer stocks, both of which are created by government policy aimed at avoiding aggregate demand pressures that might fuel an inflationary spiral.
The two buffer stocks that we will compare are:
- Unemployment Buffer Stocks: Under a Natural Rate of Unemployment (NRU) also referred to a Non Accelerating Inflation Rate of Unemployment (NAIRU) regime, inflation is controlled using tight monetary and fiscal policy, which leads to a buffer stock of unemployment. This is a very costly and unreliable target for policy makers who are trying to achieve price stability.
- Employment Buffer Stocks: The national government exploits the fiscal power embodied in a fiat currency issuing system to introduce full employment based on an employment buffer stock. The Job Guarantee (JG) model is an example of this type of policy approach.
Both buffer stock approaches to inflation control introduce so called inflation anchors. In the NAIRU case, the anchor is unemployment, which serves to discipline the labour market and prevent inflationary wage demands from being pursued. Under a Job Guarantee, the inflation anchor is provided in the form of an unconditional, fixed wage employment guarantee provided by the government.
And, in stating our objective for the rest of the Chapter we conclude that:
Finally we outline and contrast the two buffer stock schemes, which are designed to control inflation. We show that only a Job Guarantee approach provides an employed buffer stock that promotes both full employment and price stability.
That could not have been stated more clearly than that.
A central organising concept in New Keynesian (Mankiw) macroeconomics is the NAIRU.
A central organising concept in MMT is the Job Guarantee.
The latter is a specifically designed departure from the former.
Later on, as we introduce each ‘buffer stock’ framework specifically, we note:
As we demonstrate, a Job Guarantee (JG) is at the centrepiece of MMT reasoning. It is neither an emergency policy nor a substitute for private employment, but rather would become a permanent complement to private sector employment. A direct job creation program can provide employment at a basic wage for those who cannot otherwise find work. No other program can guarantee access to jobs at decent wages. Further, the JG approach has the advantage that it simultaneously deals with the main objection to full employment: the Phillips curve argument that the maintenance of full employment causes unsustainable rates of inflation.
In Section 19.6, we show the way the introduction of a Job Guarantee, which we had earlier analysed in detail, replaces the traditional Phillips curve.
Given the centrality of the Phillips curve, the relationship between inflation and unemployment, in the mainstream macroeconomics framework and the embedded concept of the NAIRU in that framework, one would have thought that a section on this topic where the assertion was that employed buffer stocks replace the NAIRU would have piqued the interest of a New Keyesian seeking to understand our work.
Figure 19.3 in the textbook explained how “the introduction of the JG eliminates the Phillips curve”. Which is quite a statement and hardly one that a mainstream reviewer intent on producing a ‘skeptics guide’ would allow to pass without comment.
The diagram (produced below) contrasts the dynamics of a Phillips curve world, where lower unemployment is associated with higher inflation rates to an economy where an employment buffer stock is used.
The text says that we might start with a Phillips curve world where the unemployment rate is currently at URA, the inflation rate was IA, and the full employment unemployment rate is URFULL, which denotes frictional unemployment.
We posit that the government wants to reduce unemployment and knows that if it increases aggregate spending it will set off cost pressures that drive the inflation rate up to IB – which in the Phillips curve world takes us from Point A to Point B.
We get full employment but a higher inflation rate.
In the NAIRU model, inflation would not be stable at B because bargaining agents (workers and firms) would incorporate the new higher inflation rate into their expectations and the Phillips curve would start moving out.
We considered that issue in detail in Chapter 18 of the textbook where we provided a faithful rendering of the mainstream theoretical approach.
Now consider what would happen in Figure 19.3 if the government introduced a Job Guarantee when the economy was at A.
The Job Guarantee could absorb workers in jobs commensurate with the difference between URA and URFULL, although in reality, as more work was available, workers from outside the labour force (the hidden unemployed) would also take Job Guarantee jobs in preference to remaining without income.
But whatever the quantum of workers that would initially be absorbed in the Job Guarantee pool, the economy would move from A to AJG rather than from A to B.
So we find that when there is a Job Guarantee operating, full employment and price stability go hand in hand.
This is why MMT considers the Job Guarantee replaces the Phillips Curve as a major conceptual device and policy application.
This is no small issue.
I hope that Greg Mankiw changes his “Skeptic’s guide” when he presents it in the coming days at a major conference in the US.
He has clearly misrepresented our work in significant ways and that should not be tolerated in an academic milieu.
That is enough for today!
(c) Copyright 2019 William Mitchell. All Rights Reserved.