MMT Primer: Mosler's White Paper

Warren Mosler, one of the founders of Modern Monetary Theory (MMT) has a white paper (link) that acts as a summary of MMT. It is brief (1300 words), and does meet the objective of offering an overview of the question "What is MMT?" for more advanced readers. The brief nature of the paper means that not everything is covered, so I would point out that it does not represent all of MMT.

In this article, I will summarise the contents of the white paper, discuss what I see as the main legitimate critiques of MMT that arise, and note some of my reservations about the paper. I will conclude with comments aimed at readers looking for an academic discussion. I expect to cover certain topics in follow up articles, such as a longer discussion of critiques, as well the implications of the discussion of the price level theory.

Another point to note about the white paper is that it is extremely terse. There may be a lot of information hidden under the use of catch phrases. Since I am familiar with the arguments, I can easily follow the logic, but others might be tripped up. All I can really do is add some background information.

Summary

I will be quoting heavily from Warren Mosler's white paper, but my assumption is that the reader has read the white paper itself.

Mosler answers the question "What is MMT?" as follows.

MMT began largely a description of monetary operations, which are best thought of as debits and credits to accounts kept by banks, businesses, and individuals.

Warren Mosler independently originated what has been popularized as MMT in 1992.  And while subsequent research has revealed writings of authors who had similar thoughts on some of MMT’s understanding and insights, including Abba Lerner, George Knapp, Mitchell Innes, Adam Smith, and former NY Fed chief Beardsley Ruml, MMT is unique in its analysis of monetary economies, and therefore best considered as its own school of thought.

To give further background, Warren Mosler is successful fixed income investor who developed the ideas around MMT independently of the other founders, William Mitchell and Randall Wray, who were academics, within the broad post-Keynesian school of thought. The white paper follows Mosler's thinking, and is less reliant on post-Keynesian theory. The white paper manages to be brief by reducing the emphasis on post-Keynesian analysis -- but that understates the scope of the broader academic MMT project.

According to the white paper, the key difference between MMT and other schools of the thought relies upon the following observations. This is what I will refer to as the "operational core" of MMT.

MMT alone recognizes that the US Government and its agents are the only supplier of that which it demands for payment of taxes
That is, the currency itself is a simple public monopoly.
The US government levies taxes payable in US dollars.
The US dollars to pay those taxes or purchase US Treasury securities can only originate from
      the US government and its agents.
The economy has to sell goods, services or assets to the US government (or borrow from the US government) or it will not be able to pay its taxes or purchase US Treasury securities. 

(The reader may note the use of the "US government" in the above. I will return to this point later.)

Most online discussion of MMT revolves around the operational core of MMT, and its implications. However, this obscures what is probably more important from a theoretical point of view: the discussion of the Job Guarantee and the determination of the price level.

Only MMT recognizes the source of the price level.  The currency itself is a public monopoly.  Monopolists are necessarily “price setters”.
Therefore: The price level is necessarily a function of prices paid by the government’s agents when it spends, or collateral demanded when it lends. [emphasis in original]
In a market economy the government need only set one price, letting market forces continuously determine all other prices as expressions of relative value, as further influenced by institutional structure.   

If one wants to discuss "what is new in MMT?", this is where one really needs to start. However, critiques of MMT typically stay focused on the operational aspects.

(Update: I have added a small technical appendix that discusses an observation that is easily missed. If we look at neoclassical models, the fiscal arm of the government is assumed to be a price taker, and that assumption is carried over into almost all analysis of fiscal policy. That is an assumption that needs to be questioned.)

"Where Does MMT Apply?"

The white paper specifically refers to the "United States," which is unfortunate. One standard criticism of MMT is that it only applies to the United States, allegedly because the U.S. dollar is "the reserve currency." Most proponents of MMT (such as myself) reject that argument. Instead, MMTers would normally state that countries with some form of currency peg -- which includes euro area countries -- are constrained, and lack the full freedom for fiscal policy. The preferred phrasing is to use "currency sovereigns."

At the minimum, most people recognise that many of the developed countries (outside the euro area) are in a similar boat as the United States; too many people have lost too much money betting on a default by the Japanese government. To what extent there is a debate, it is for the developing countries. Since I have no expertise in development economics, I prefer to avoid that debate.

Monetary Operations: Implications and Debate

The key implication of monetary operations analysis is that currency sovereigns cannot be forced to involuntarily default for financial reasons. Instead, the constraint on fiscal policy is the inflationary consequences -- which will include the cost of imported goods.

The White paper lists implications of operations analysis as follows.

1.  The US government and its agents, from inception, necessarily spend (or lend) first, only then can taxes be paid or US Treasury securities purchased.

This is in direct contrast to the rhetoric that states the US government must tax to get US dollars to spend, and what it doesn't tax it must borrow from the likes of China and leave the debt to our grandchildren.

MMT therefore recognizes that it's not the US government that needs to get dollars to spend, but instead, the driving force is that taxpayers need the US government’s dollars to be able to pay taxes and purchase US Treasury securities.

2.  Crowding out private spending or private borrowing, driving up interest rates, federal funding requirements and solvency issues are not applicable for a government that spends first, and then borrows.   

One alternate way to summarise the implications of operations analysis is that the debt-to-GDP ratio is a meaningless statistic. (I believe that I first heard that statement in a verbal comment by Warren Mosler.)  If we look at "conventional" analyses of fiscal policy (at least historically), we see the opposite -- there is considerable alarm about the level of the debt-to-GDP ratio, and the objective of fiscal policy is to steer that ratio. UPDATE: This post offers a lengthier discussion of the use of the debt-to-GDP ratio to clarify debates about MMT.

Debates about the Implications of Operations: Yawn

If and when I write my MMT primer, I expect to include reprints of articles where I discuss online MMT debates. Most of the articles will involve disputes about monetary operations, and their implications.

However, I would argue that most of the debates end up being pointless semantic tussles. Critics ultimately recognise that the only true limitation on fiscal policy is the fall in the value of the domestic currency unit -- either versus domestic goods and services ("inflation") or an increase in the prices of imported goods and services ("import inflation"). That is exactly what MMT says is the constraint -- not some arbitrary level of the debt-to-GDP ratio.

I write popularisations for economics. My job is to explain in plain English what these debates are about. If we put both sides arguments' into plain English, and they are saying the same bloody thing, I get rather testy. As a result, I will not pursue this argument further herein.

Debates about Novelty?

Another standard criticism of MMT is that there is nothing new there. If we look at the operational core of MMT, it overlaps Functional Finance (link to primer), heavily associated with Abba Lerner.

One may note the MMTers themselves cite Lerner and Functional Finance (as in the introduction to the white paper, quoted above). (This is in contrast to neoclassical authors, that largely airbrushed functional finance out of citations.) So we would need to look at the actual articles written by MMTers to see where they assert the novelty lies. One of the distinguishing characteristics of criticisms of MMT is a general unwillingness to cite journal articles written by MMTers, so there is no actual content within this debate to discuss. 

(As an ex-academic, the discussion of academic novelty is a topic I would prefer to avoid, since my views would be unwelcome if not hurtful to the vast majority of the professoriate.)

Role of Interest Rates

From a fixed income practitioner's perspective, the most eye-catching part of Mosler's work is the discussion of interest rates.

Mosler in the white paper:

MMT recognizes that a positive policy rate results in a payment of interest that can be understood as “basic income for those who already have money.”

MMT recognizes that with government a net payer of interest, higher interest rates can impart an expansionary, inflationary (and regressive) bias through two types of channels -- interest income channels and forward pricing channels.  This means that what’s called “Fed tightening” by increasing rates may increase total spending and foster price increases, contrary to the advertised intended effects of reducing demand and bringing down inflation.  Likewise, lowering rates removes interest income from the economy which works to reduce demand and bring down inflation, again contrary to advertised intended effects.

MMT understands that a permanent 0% policy rate is the base case for analysis for a floating exchange rate policy. 

The role of interest rates in steering the economy is a topic that should be of far more interest. Why did the economy no accelerate, even though there was a multi-year period of negative real interest rates? Those of us with long memories are well aware of the argument to the effect that sustained negative real rates were the cause of the 1970s inflation.

The analysis of a permanent 0% interest rate policy ("PZIRP") is another topic that receives not yet enough attention. For example, critics will discuss an out-of-control spiral due to surging interest rates. If the nominal interest rate on government liabilities is locked at 0%, how does such a spiral happen?

The Price Level

One argument popular among some trolls on the internet is that there is no MMT theory of inflation. The quoted passage earlier shows this to be incorrect. The problem is that it is not entirely clear that all MMTers agree on what all the details of the theory should be.

I will repeat this statement from Mosler:

In a market economy the government need only set one price, letting market forces continuously determine all other prices as expressions of relative value, as further influenced by institutional structure.   

There is actually a lot hidden within that statement, but it puts Mosler's thinking somewhat closer to neoclassical theory than post-Keynesians. However, the theoretical split is papered over by the fact that the role of the Job Guarantee is the same: it acts as an anchor for the price level.

I will return to this topic in a later article. But I will summarise what I see as the key point. Mosler's complaint (which I arrived at independently) was that all prices that matter in neoclassical models are relative prices: unit wages/price of a unit of goods, price of unit of goods/price of a unit of money, etc. There is nothing to pin down the level of prices; we can scale all spot prices by a scalar factor, and all that happens is the endowment of initial financial wealth changes in real terms.

The neoclassicals largely dodge this issue by changing the subject. One cannot find the worked out full solutions to models, rather there is a leap to discussion of the derivatives of an unknown solution. Only the Fiscal Theory of the Price Level grasps that nettle.

One may note that both Mosler's view and Fiscal Theory of the Price Level come up with a similar-sounding solution: the spot price level is based on fiscal policy. The difference is that the Fiscal Theory of the Price Level relies on rather silly assumptions: the price level now is determined by the constraints on buying goods in forward markets -- including forwards 20 billion years in the future. Conversely, Mosler's argument is that the price level can be set by transactions in spot markets -- and the best way to do that is to set a floor price for labour wages.

Unfortunately, this is a somewhat theoretical discussion. What happens when a Job Guarantee does not exist (which is pretty much the case everywhere)? This concern is why the white paper summary is too terse for many readers' needs.

Concluding Remarks

Many people who might be interested in MMT are under severe reading time constraints, and need terse high level summaries. Warren Mosler's white paper does fit that description. That said, it is missing a discussion of the post-Keynesian theory wing of MMT, which obviously matters for those with an interest in new approaches to macro.

My focus in this article was to comment on the high level content of the white paper, and the main legitimate criticisms one might make of the contents. The perplexing thing about online discussions of MMT is that most revolve around criticisms that make little sense, but are ultimately linguistic disputes ("what is a financial constraint?"). The interesting bits of MMT are ignored -- possibly because that would require some reading of journal articles.

For Academics

Warren Mosler's white paper is not a document aimed at academics, although they might find it useful. Based on various attempts to critique MMT that I have seen over the years, academics have had a difficult time coming to grips with MMT, which I find rather surprising.

• One potential starting point is the textbook Macroeconomics, by William Mitchell, L. Randall Wray, and Martin Watts (my initial comments). However, what one will discover is that the textbook is aimed at undergraduates, with rather limited bibliographic information. If one wants to work with this textbook, one may need to look through the index to find  discussion of background information.
• There are various article search engines, such as Google Scholar. These services did not exist when I was an academic, and so I am not an expert on their use. The obvious starting point would be to look for promising publications by William Mitchell and L. Randall Wray, as they are considered the founders of MMT (along with Warren Mosler). There are other foundational articles written by other MMT scholars, but I will dodge giving a list here (to avoid offending anyone by omitting them).
• One could go various centres for study or website, and consult their resources. The Gower Initiative for Modern Money Studies (URL: https://gimms.org.uk/) announced that it will be developing a list of key journal article. There is a database of articles at https://www.zotero.org/groups/2251544/mmt_academic_resources/items. Another major website is New Economic Perspectives (URL: http://neweconomicperspectives.org/).

I expect that I will write a MMT primer of some sort, and will have a discussion of key articles.

Update: Link to the next article in this sequence, on price level determination.

Appendix: Government as a Price Taker?

Based on a comment by Ralph Musgrave on the Mike Norman Economics website (URL: http://mikenormaneconomics.blogspot.com/2020/01/mmt-primer-moslers-white-paper-brian.html), I realise that there is a lot of hidden information in Mosler's statements about prices being a function of what the government pays.

The embedded assumption of neoclassical models is that the government is a price taker: it has to buy goods at the market price. Real-world governments sort-of follow that (although wages of government employees are administered in concert with union negotiations).

The standard assumption is that the government buys a fixed amount of goods in real terms, and if the price rises, the nominal expenditure rises. This fuels inflation. Mosler's argument points out that the governments do not have to follow that framework.

If the government refuses to pay increased prices, the volume of goods and services purchased will necessarily fall. This tightens fiscal policy relative to plans. Since taxes are levied based on nominal incomes or transactions (if a value-added tax is imposed), rising private sector prices and wages will be met with a higher tax burden. At some point, the contractionary fiscal policy would reverse inflationary pressures.

This is a topic that will be returned to when I discuss the price level aspects in a later article, but it underlines how relying on conventional assumptions results in missing aspects of what MMTers are saying.

(c) Brian Romanchuk 2019 2020