Money, Fiscal Policy, and Interest Rates: A Critique of Modern Monetary Theory

Abstract

This paper examines modern monetary theory (MMT). MMT is a restatement of established Keynesian monetary macroeconomics and so there is nothing new warranting a separate nomenclature. MMT over-simplifies the challenges of attaining non-inflationary full employment by ignoring dilemmas posed by the Phillips curve, maintaining real and financial sector stability, and an open economy. Its policy recommendations take little account of political economy difficulties, while its interest rate policy recommendation would likely generate instability. On the plus side, MMT's advocacy of expansionary fiscal policy is useful at a time when too many policymakers are being drawn toward mistaken fiscal austerity.

Notes

¹Mitchell's work has an uneasy relationship because of his emphasis on the Phillips curve, which is in contradiction to Wray's position.

²Wray (1998, chapter 2) provides a thorough and concise discussion of Chartalist theory.

³Wray (1998) cites this quote but it is buried in the last footnote (number 16) of his chapter on Chartalism.

⁴For instance, Michael Hoexter (2013), a leading contributor to the MMT website New Economic Perspectives, writes: ‘One of the components of macroeconomic management recommended by Keynes but theorized only later by Modern Money Theorists was ending the gold standard and transitioning to a fiat currency … . Abba Lerner's functional finance and later Modern Money Theory (MMT) have been the theories of fiat currency which have as yet not been self-consciously utilized within government policy or integrated into the mainstream economic teaching which still views money as a commodity among other commodities and not what it has been now for many years, a fiat money monopoly.’ MMT claims in this vein are self-promoting and false.

⁵Neo-Keynesians were fully aware of this difference. For instance, Haliossos & Tobin (1990, p. 899) describe the language of ‘budget restraint’ as a ‘misnomer’ and they prefer to call it a ‘budget flow identity’, reflecting its purely accounting character.

⁶Judging by the absence of citation references (see for instance, Wray, 1998), MMT-ers appear to be unaware of this earlier neo-Keynesian literature.

⁷Lavoie (2011) and Fiebiger (2012) focus on the fiscal authority–central bank relationship but there is also the issue of monetary policy. Even when the central bank and fiscal authority are not consolidated, it can look as if they are. The effect of budget deficits on the money supply and interest rates depends critically on the stance of monetary policy. Thus, if an independent central bank decides to fully accommodate fiscal policy then it will produce an outcome that looks as if the central bank and fiscal authority are consolidated. MMT therefore implicitly assumes both consolidation and full accommodation by monetary policy.

⁸My suggested reform of the Federal Reserve System is described in Palley (2012b) and would give the President greater influence over the Federal Reserve.

⁹In a growing economy, the fiscal authority can run persistent money-financed deficits and still maintain price stability if the implied growth of the money supply equals the rate of growth of real output.

¹⁰The expenditure multiplier for a standard income expenditure model is given by m = 1 – α – β[1 – t] where α = accelerator investment coefficient, β = marginal propensity to consume, and t = income tax rate >0. Expenditure multiplier stability requires 1 – α – β > 0. It can then be shown that ΔD < 0 requires 0 < 1 – α – β. The deficit can only fall if the economy is unstable.

¹¹In terms of the Blinder & Solow (1973) and Tobin & Buiter (1976) models, the MMT experiment is identical to a money financed budget deficit with a constant interest rate. Blinder & Solow (1973) and Tobin & Buiter (1976) actually analyzed fiscal policy under more arduous conditions when the monetary authority holds the high-powered money supply constant. Their analysis shows fiscal policy is still expansionary (although Post Keynesians would also argue this is a fictional experiment as monetary authorities do not target the high-powered money supply). In both cases the economy settles at a stable equilibrium characterized by a balanced budget (D = 0). The MMT experiment adds the additional requirement that the budget deficit only be balanced when the economy reaches full employment output (y = y*). However, there is nothing theoretically new.

¹²MMT's failure to acknowledge this neo-Keynesian IS-LM literature reflects a complicated sociology. First, the IS-LM stock-flow consistent literature is quite mathematical and MMT-ers have tended to exhibit an aversion to mathematical modeling on grounds that such modeling is an exercise in fiction. A second reason is that MMT is an extreme wing of Post Keynesian economics and many Post Keynesians have an allergy to IS-LM analysis. Rejection of IS-LM analysis has become a near-litmus test among many Post Keynesians. Such thinking is misinformed. Stock-flow consistent IS-LM analysis is temporary equilibrium analysis (like the income–expenditure model) conducted in output–interest rate space. IS-LM's architecture is fine but its detailed specification is subject to meaningful critique, particularly as regards the omission of endogenous money, inside debt, and inside debt effects on AD. This theological rejection of IS-LM has done great damage to PK economics by creating an unnecessary schism with neo-Keynesians and blinding Post Keynesians to the merits of the IS-LM model as a base on which they could have built. Indeed, Post Keynesian economics is now unwittingly re-inventing the stock-flow consistent IS-LM model as evidenced by the much cited work of Godley & Lavoie (2007). The principal innovation in their framework is the extension of the stock-flow consistent IS-LM model to include endogenous money, inside debt, and inside debt effects on AD.

¹³The language of ‘taxes drive money’ is misleading. What MMT means is that taxes positively affect the demand for sovereign money. However, taxes are just one factor, not the only factor. Moreover, in an endogenous money system in which the monetary authority targets the overnight interest rate, any demand for sovereign money to pay taxes will be provided by the monetary authority. Consequently, there can never be a shortage of sovereign money to pay taxes.

¹⁴The money demand function could also include T as a separate argument with H_T  > 0.

¹⁵The Federal Reserve's QE programs have directly monetized part of the budget deficit via government bonds purchases. The Federal Reserve has also purchased agency-backed mortgage backed securities (MBS). Since such securities are very close portfolio substitutes for Treasury bonds, their purchase is tantamount to deficit monetization.

¹⁶Phillips curve analysis imposes a trade-off between inflation and unemployment. My experience of MMT-ers is that they apply an extreme discount on inflation and view it as essentially costless economically. In this regard, I have heard a leading proponent of MMT claim inflation below 40 per cent is costless.

¹⁷My thanks to Esteban Perez for pointing out the balance of payments constraint on expansionary fiscal policy and Wynne Godley's concern with it.

¹⁸That is because current generations own the bonds and will therefore receive payment made from some combination of future taxes and future money issue.

¹⁹Ensuring AD = y* and D = 0 may requires a bit of fiscal fine tuning with both T and G going up. Increased T closes the budget gap, while further increased G offsets any negative AD impact of increased T.

²⁰Arguments in this section are based on an early critique (Palley, 2001). Sawyer (2003) also argues that ELR is neither the macro nor micro economically efficient way to reach full employment, and it does not solve the problem of inflation at full employment. Seccarecia (2004) has also offered a critique of ELR that focuses on the failure of ELR to address the distributive problems of the existing system.

²¹See ‘Million jobless may face six months’ unpaid work or have unemployment benefits stopped,’ The Guardian, Sunday 29 July, 2012. The Conservative's proposed scheme requires unemployed workers to take ELR jobs in order to receive unemployment benefits so that the ELR weekly wage effectively equals unemployment benefit. The Conservative proposal differs from the MMT proposal in that taking ELR jobs is mandatory. In the MMT proposal accepting an ELR job is a voluntary choice and workers continue to receive unemployment benefits even if they decline taking ELR jobs.