Mainstream Macroeconomics and Modern Monetary Theory: What Really Divides Them? A Very Great Deal Indeed!
I have been asked by Economic Reform Australia for a response to a very recent John Jay College, CUNY working paper, entitled Mainstream Macroeconomics and Modern Monetary Theory: What Really Divides Them? (Jayadev & Mason 2018).
The authors of the paper, who identify themselves as being sympathetic to modern monetary theory, although they are themselves mainstream economists, argue that the answer to this question is ‘not very much’.
Their view is that neither mainstream macroeconomists nor modern monetary theorists are correct to see modern monetary theory as a radical challenge to macroeconomic thought, and that instead the essential difference between the modern mainstream and modern monetary theorists is a debate about the relative effectiveness of monetary policy and fiscal policy as tools for stabilising the economy across the business cycle.
In their abstract, they claim that ‘while MMT’s policy proposals are unorthodox, the analysis underlying them is entirely orthodox’.
This would be correct if modern monetary theory equated to the old-fashioned 1950s and 1960s Neo-Keynesian economics of what is often described as ‘the first neoclassical synthesis’. But MMT does not fit that equation, and so the authors of this paper are incorrect.
In the early 1970s, James Tobin argued that the debate between the (Neo-) Keynesians, including Tobin himself, and Milton Friedman’s monetarists boiled down to the same issues the authors of this paper have chosen to raise in 2018 (Tobin 1971). It was all about whether changes in interest rates or changes in the fiscal balance were a more effective way of influencing total spending, output and employment, and this depended on the interest elasticity of investment spending by firms, and the interest elasticity of the demand for money. At the time, Friedman correctly argued that there was more to it than that. The idea that this is the key issue between mainstream economists, as the modern-day Friedmanites, and modern monetary theorists, as though we are channelling James Tobin, is deeply mistaken.
Space precludes me from discussing this in full in this piece. That will have to wait for another occasion, or more correctly a series of other occasions. But we can deal here with a few misconceptions and errors of emphasis in Jayadev and Mason’s paper.
They claim their goal is ‘not to make an assessment of MMT as a whole’ and they admit to making ‘only limited references to MMT literature’. How you are supposed to answer the question they set themselves without assessing MMT as a whole, and how you can imagine yourself able to answer this question without referring in depth to the relevant MMT literature I will leave to one side. Perhaps we should be grateful that they are engaging with MMT at all, and doing so in a way which is not entirely dismissive.
They choose to ignore the chartal or state theories of money on which MMT is founded, and do not discuss at all the role of an employment guarantee within MMT. Indeed they state, ‘It is unfortunate, in our view, that many MMT texts begin with discussions of endogenous money, chartalism, and the mechanics of government fiscal operations. These arguments are intended to make the case that modern states have the capacity to borrow without limit at an interest rate of their choosing, but there is no need to establish this. It is already implicit in the orthodox view that the central bank can set the interest rate.’
Without an understanding of fiscal operations within a modern monetary system, it is not clear that government financial liabilities are properly not viewed as debt in the conventional sense at all, but as net financial assets for the non-government (private) sector; it is not clear that the ‘money multiplier’ theory is based on misconceptions about how the monetary system works, so that the issuance of debt securities by a monetary sovereign running fiscal deficits is optional; and it is not clear that the fiscal balance and any government debt to GDP ratio you choose to define are never appropriate targets for policy makers.
Jayadev and Mason are right to say that modern monetary theorists ‘do not see the debt ratio as an important target for policy’. They are wrong, or at least misleading, to say ‘that most MMT advocates would probably agree that the debt ratio should not be allowed to rise without limit’. The debt ratio WILL not rise without limit, and should only rise if there is a demand from the non-government sector to net save domestic currency, while the economy operates at the full employment level of income. It is not that the debt ratio is not an important target for policy. It is NEVER a suitable target variable – important or not – and should evolve as necessary in response to the evolution of the fiscal balance, to maintain non-inflationary full employment.
They argue that modern monetary theorists argue interest rates should be kept low in order to prevent the government debt ratio rising. This also is misleading. Modern monetary theorists argue for low, or even zero, official interest rates, because the link between interest rates and total spending is unreliable, shifts over time as balance sheets evolve, and even has an uncertain sign; and because non-zero risk-free interest rates have adverse implications for the distribution of income. This has nothing to do with stabilising an essentially irrelevant debt ratio. The fact that a low or zero interest rate DOES prevent that ratio rising without limit is true, but the idea that this is the reason we advocate for low or zero and stable interest rates is false.
They claim ‘the limits of a central bank’s ability to control interest rates remains an open question’. Once again, it is a pity they have no interest in the ‘mechanics’ of fiscal and monetary operations, or it would be obvious that a central bank can always use its balance sheet to set risk-free interest rates across the yield curve. This is a matter of fact, and there are no limits to it. It should not be an open question.
Lower, or zero, interest rates do not necessarily imply higher private spending, and so do not imply a larger primary fiscal surplus (or smaller primary fiscal deficit) for non-inflationary full employment. Zero official interest rates reduce the flow of interest payments from the government to the private sector, for example, which depresses demand; interest payments within the private sector are transfers of purchasing power; and any impact of lower interest rates on private borrowing has balance sheet effects, which means that any stimulus is limited in duration.
This of course is why modern monetary theorists favour more extensive bank regulation, and in some cases bank nationalisation, to influence the direction of new credit creation, and to limit the amount of new lending over time. Higher interest rates are not the only available mechanism for limiting private sector indebtedness.
They are right to say that MMTists ‘pursue the fiscal balance consistent with price stability’, and that we accept the basic logic behind the Phillips Curve relationship between inflation and unemployment. They are wrong to confuse our definition of full employment, which is a situation where there is no involuntary unemployment, with the mainstream definition, of a so called ‘natural rate of unemployment’ (or NAIRU). They are locked inside the mainstream, general equilibrium view of capitalism, as are other mainstream economists. This is a view which both modern monetary theorists, and Post-Keynesians more generally, reject as unrealistic, unhelpful, misleading, and biased.
This is where their failure to consider a job guarantee is so unfortunate. The job guarantee is the automatic fiscal stabiliser which eliminates the Phillips Curve trade off and does away with NAIRU, because it acts an institutional mechanism to allow us to attain genuine full employment without inflationary consequences. Mainstream economists use a buffer stock of unemployed labour to control inflation. We advocate a buffer stock of equitably and productively employed labour as a superior tool of stabilisation.
Because Jayadev and Mason are locked into natural rate thinking and the associated Phillips Curve as immutable facts of life, their definition of potential output is different to ours. They define potential output as the level of output consistent with the NAIRU rate of unemployment, and see the task of the authorities as stabilising output at or close to this level. We define full employment output as the level of output consistent with non-inflationary full employment, when a job guarantee is in place. Just as the size of the job guarantee automatically sets the appropriate fiscal balance, so the level of output will be whatever it needs to be for there to be equitable and non-inflationary full employment.
It is unfortunate that they repeat the oft-stated claim that democracy imparts a bias towards deficits and inflation. It is surely increasingly difficult to argue that this is the case, given the history of recent years. The problems of the great democracies of the USA, Europe and Japan, and so many others, in recent times have hardly been indicative of a bias towards inflation. Democracy ought to impart a bias over time to effective government, in the sense that ineffective governments ought to lose elections, and be replaced. Where this is not the case, the model of democracy being used is at fault.
They conclude by saying, ‘What reason do we have to believe that an elected government that was free to set the budget balance at whatever level was consistent with price stability and full employment, would actually do so? This is where the real resistance lies.’ How unfortunate that they have no idea of the central significance of the job guarantee in modern monetary theory. It is also unfortunate that they do not ask, ‘What reason do we have to believe that a government irresponsibly pursuing an inappropriate policy target of a fiscal surplus will not drive the economy into increasing private indebtedness, financial fragility and a severe recession?’
I am grateful to the authors of this paper for their attempt to engage with modern monetary theory. That they have not been able to do so effectively is because they have not delved deeply into the inadequacy of mainstream (neoclassical) macroeconomics as a useful description of how monetary economies actually function. To do this, they would need to acquaint themselves with the work of a variety of Post-Keynesian economists, including Paul Davidson, Hyman Minsky, and the great Michal Kalecki, and then go back and re-read Keynes himself. They ought then to re-read Lerner and to study carefully Wynne Godley’s stock-flow consistent monetary models of the economy. They refer to a paper by Godley’s co-author, Mark Lavoie, but they ignore the much more important work Lavoie did with Godley on stock-flow consistent modelling, which is an important tool within MMT.
Actually, they could just read my book, Economics for Sustainable Prosperity (Hail 2018), which explains the philosophical and scientific differences between mainstream macroeconomics and modern monetary theory, and builds connections from MMT to both behavioural economics and ecological economics.
I am planning to summarise the chapters of this book for readers of the ERA Review over the next year or so, so that readers can fully answer the question Mainstream Macroeconomics and Modern Monetary Theory: What Really Divides Them?
The brief answer is ‘a very great deal indeed’.