Why MMT can’t work


Using a stock-flow consistent ISLM open-economy model, this article shows that, unless very specific country circumstances hold, modern money theory (MMT) cannot work as an effective and sustainable macroeconomic policy program aimed to achieve and maintain full-employment output through persistent money-financed fiscal deficits in economies suffering from Keynesian unemployment or underemployment. Specific country circumstances include cases where the economy enjoys very high policy credibility in the eyes of the international financial markets or issues an international reserve currency; under such circumstances, the adverse outcomes of MMT policy can be prevented and expansionary demand shocks can be effective. Short of such features, an open and internationally highly financially integrated economy that implements MMT policy would either see its money stock grow unsustainably large or would have to set domestic interest rates to levels that would be inconsistent with the policy objective of resource full employment and that would cause instead economic and financial instability. The article explains why ISLM analysis is used to support the arguments developed in it.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4


  1. 1.

    Credibility lies at the core of the Portfolio Theory of Inflation (PTI), which shows it to be a critical determinant of the (in)effectiveness of monetary and fiscal policies [4, 5]. In a nutshell, according to the PTI (whose results inspire this article), international financial markets determine the value of all public sector liabilities of open and internationally highly financially integrated economies, including domestically denominated debts and currency. While anti-recessionary policies undertaken by credible governments are effective, those implemented by poorly credible governments cause markets to short their bonds and currencies, thus causing their value to fall. In response, policy makers may either correct nominal interest rates, though at the cost of dampening the expansion, or push still on the money lever in an attempt to support bond prices and sustain demand, yet only fueling capital outflows and depressing the (external and internal) value of the currency.

  2. 2.

    Notice the distinction between the terminology "Modern Money Theory," which will be used here throughout, and "Modern Monetary Theory," often referred to in the literature and in commentaries on the subject. In the words of one of the leading MMT theorists and proponents, L. Randall Wray, «I will note here that I use the terminology Modern Money Theory—and have seen that usage as following on from the title of my 1998 book (Understanding Modern Money). We also used this terminology in our textbook. However, many of my colleagues had used the other terminology Modern Monetary Theory. I object to that as it draws attention to the word “monetary” and leads many to believe it is all about monetary policy—while in reality much of the focus is on fiscal policy (and we argue that the traditional division between the two is highly misleading in any case). Further, it seems to conjure in some minds a similarity to Monetarism.» [32].

  3. 3.

    This is precisely the point made by Aspromourgos [1]: «in a world of inconvertible fiat currencies, public investment as a driver of aggregate demand faces little financial constraint. In the first instance, such a constraint would exist only to the extent that the suppliers of goods and services that government wishes to purchase are resistant to accepting payment in outside money or “cash” (including electronic or “book entry” outside money). But the willingness of private sector agents to accept payment in cash is one thing,their willingness to then hold money, as a desired asset, is another. If there results excess money balances for the private sector as a whole, then it is possible that the excess can be drained from the private sector via its purchasing government securities of various maturities…[I]f, at prevailing yields on government securities, the private sector as a whole is unwilling to substitute government securities for the entirety of any such excess money balances (net of taxation), then that money will find its way into other channels (expenditure on other assets or on goods and services) until it ceases to be an excess—unless government yields become more attractive.» (p. 510).

  4. 4.

    This issue has been recently discussed by Jayadev and Mason [13], which holds that MMT and orthodox macroeconomics rely on many of the same theoretical foundations, with the only difference that mainstream economists think that monetary policy should be privileged to look after full employment and price stability, while MMT economists think fiscal policy should have that role. This conclusion (and the methodological approach applied to arrive at it) has been strongly opposed by Mitchell [19].

  5. 5.

    The fact that the aggregate expenses of one sector of the economy is the income of other sectors, and aggregate saving is always a residual of aggregate investment, does not necessarily imply that savers are always willing to hold any asset stocks that are supplied to the economy, unless the prices of these asset stocks adjust adequately (and are expected to be sustainable). Failure of equilibrium prices to attain, at which asset stocks are fully demanded, leads savers not to absorb any addition to those stocks in their portfolio and possibly even to reallocate their portfolios away from those stocks, causing their value to decline. On the other hand, equilibrium asset prices, even if achieved, might not be consistent with full employment and, with imperfect markets, there might not be mechanisms to ensure such consistency. This is not an issue of aggregate saving’s adequacy, but one of optimal portfolio composition vis-à-vis wealth-holder preferences, with equilibrium prices being those that equate the supply of and demand for money vis-à-vis other assets into which wealth holdings can be placed, based on wealth-holders’ preferences to hold them in liquid and/or less liquid forms.

  6. 6.

    Mitchell (2018b) rejects DSGE modeling as irreconcilable with MMT.

  7. 7.

    Thus, with nonzero tax and public debt, the complete expression for the money supply would be \(\frac{\Delta M}{P}=G+{B}_{-1}{i}_{{B}_{-1}}-T-\frac{\Delta B}{P}\).

  8. 8.

    See my works on the Portfolio Theory of Inflation, cited earlier. While even in closed and financially unintegrated economies, large wealth holders always find ways to move money abroad (not infrequently illegally), governments can rely on captive (if not repressed) markets to place their liabilities at subsidized conditions, without having to confront the judgment of the markets and the constraints imposed by them, though at the cost of significant economic distortions and limited or no access to valuable investment resources from abroad.

  9. 9.

    In a typical Keynesian fashion, the marginal efficiency of capital, \(\mu\), obtains as the value that equates the supply price of capital with the present value of the expected profit stream over a relevant future time horizon, \({P}_{K}={\sum }_{t=1}^{n}{\pi }_{t}{(1+\mu )}^{-1}\) (Chick, 1983). As indicated by MMT proponents Bill Mitchell (2012) and Warren Mosler (see Rowe, cit.), MMT accepts Michal Kalecki’s consideration that the risk of increasing indebtedness ensures that the marginal efficiency of capital is downward sloping with respect to the market rate of interest and that, all else equal, higher interest rates render unprofitable many projects that would otherwise be profitable.

  10. 10.

    As Palley [24] observes, «In a no growth economy, having the fiscal authority run persistent money financed deficits will cause the money supply to increase relative to GDP…» (p. 8), and dynamically the same would hold under MMT as the stock of money would grow at a rate not lower than GDP (all else being equal).

  11. 11.

    In an economy with alternative assets in addition to FX, wealth holders would purchase also speculative assets (typically featuring very low output elasticity), with money shifting hands at an increasing velocity within select groups of (domestic and foreign) wealth holders, who would be acting in the expectation of extracting surpluses from further asset price increases, limiting money circulation in the economy, and eventually raising the risk of bubble bursts. In the model above, this chain of events would neutralize the wealth effect on consumption. At disequilibrium interest rates, while some individual wealth holders would be able to get rid of their own "excess" money balances, not all of them would be able to do so at the same time. If an equilibrium price is not reached, some wealth holders will ultimately be forced to stick temporarily to (at least part of) their excess balances, until new opportunities would again be available to restore normal (optimal) holdings.

  12. 12.

    This was noted by Palley yyy[24], which specifically refers to the Pigou effect. Also, in a model where taxation would be included in the form of income taxes (instead of, or in addition to, lump sum taxes), the adjustment of the budget deficit would occur not only through deliberate reductions in public expenses (mirroring the absorption of the private sector surplus) but also, or even exclusively, through higher government income. The same private sector saving surplus would be reduced not only by the wealth effect, but also by reductions in household disposable incomes.

  13. 13.

    For a review of the empirical literature on policy credibility, the exchange rate, and inflation, see Bossone (2019). On the relevance of the exchange-rate pass-through effect in particular for developing economies, see Vernengo and Pérez Caldentey (2019).

  14. 14.

    Notice that it is always the case that \({m}^{*}={m}^{opt}\), while the reverse is not necessarily true.


  1. 1.

    Aspromourgos, T. (2018). Keynes, public debt, and the complex of interest rates. Journal of the History of Economic Thought, 40(4), 493–512.

    Article  Google Scholar 

  2. 2.

    Backus, D., & Driffill, E. J. (1985). Inflation and reputation. American Economic Review, 75(June), 530–538.

    Google Scholar 

  3. 3.

    Backus, D., & Driffill, E. J. (1985). Expectation and policy credibility following a change in regime. Review of Economic Studies, 52(April), 211–221.

    Article  Google Scholar 

  4. 4.

    Bossone, B. (2020). Global finance and effectiveness of macro-policies. World Economics, 21(2) 135–164.

  5. 5.

    Bossone, B. (2019). The portfolio theory of inflation and policy (in)effectiveness. Economics, 13(2019–33) 1–25.

  6. 6.

    Chick, V. (1983). Macroeconomics After Keynes, The MIT Press Cambridge, Massachusetts.

  7. 7.

    Cochrane, J. H. (2015). A new structure for U.S. Federal Debt. No 15108, Economics Working Papers, Hoover Institution, Stanford University.

  8. 8.

    Congressional Research Service. (2019). Deficit financing, the debt, and ‘modern monetary theory’. CRS Report by G. A. Driessen and J. G. Gravelle, prepared for Member and Committees of Congress, October 21, U.S.

  9. 9.

    Davidson, P. (2019), "What is modern about MMT? A concise note. In: RWER (2019) (pp. 72–74).

  10. 10.

    Forstater, M., & W. Mosler. (2005). The natural rate of interest is zero. Journal of Economic Issues, XXXIX(2) 535–542.

  11. 11.

    Hviid, S. J., & A. Kuchler. (2017). Consumption and savings in a low interest-rate environment. Denmark National Bank working paper no. 116, June.

  12. 12.

    IGM Forum. (2019). Modern monetary theory. Chicago Booth, 13 March.

  13. 13.

    Jayadev, A., & Mason J. W. (2018). Mainstream macroeconomics and modern monetary theory: What really divides them? Institute for New Economic Thinking, 6 September.

  14. 14.

    Kelton, S. (2020). The deficit myth: Modern monetary theory and the birth of the people’s economy, Public Affairs.

  15. 15.

    Kreps, D., & Wilson, R. (1982). Reputation and imperfect information. Journal of Economic Theory, 27, 253–279.

    Article  Google Scholar 

  16. 16.

    Krugman, P. (2019). Running on MMT (Wonkish). The New York Times, 25 February.

  17. 17.

    Mitchell, W., Wray, L. R, Watts, M. (2019). Macroeconomics. Macmillan Publishers, London.

  18. 18.

    Mitchell, B. (2018a). MMT and the external sector—redux. Bill Mitchell—modern monetary theory, 26 September.

  19. 19.

    Mitchell, B. (2018b). The divide between mainstream macro and MMT is irreconcilable—part 2. Bill Mitchell—modern monetary theory, 11 September.

  20. 20.

    Mitchell, B. (2013). The IS-LM Framework—parts 1–7. Bill Mitchell—modern monetary theory, 26 July–6 September.

  21. 21.

    Mitchell, B. (2012). Investment and interest rates. Bill Mitchell—modern monetary theory, 26 September, August 10.

  22. 22.

    Mosler, W. B. (2018). Soft currency economics. Mimeo. https://web.archive.org/web/20060904234833/http://epicoalition.org:80/docs/soft0004.htm. Accessed 27 Mar.

  23. 23.

    Palley, T. (2020). What’s wrong with modern money theory (MMT): Macro and political economic restraints on deficit financed fiscal policy, Review of Keynesian Economics, 8(4) 472–493.

  24. 24.

    Palley, T. (2013). "Money, fiscal policy, and interest rates: A critique of Modern Monetary Theory," Mimeo.

  25. 25.

    PERI. (2012). Modern monetary theory: A debate. Political Economy Research Institute, Working Paper Series No. 279, University of Massachusetts Amherst, January.

  26. 26.

    Rowe, N. (2011). Reverse-engineering the MMT model. Worthwhile Canadian Initiative, 15 April.

  27. 27.

    RWER (2019). Modern monetary theory and its critics, Real-World Economics Review, Issue no. 89, 1 October

  28. 28.

    Sawyer, M. (2019). Modern monetary theory: Is there any added value? In: RWER (2019) (pp. 167–178).

  29. 29.

    Summers, L. H (2019), "The left’s embrace of modern monetary theory is a recipe for disster," The Washington Post, 4 March.

  30. 30.

    Tymoigne, E., & L. R. Wray. (2013). Modern money theory 101: A reply to critics. Working Paper No. 778, Levy Economics Institute of Bard College, November.

  31. 31.

    Vernengo, M., & E. Pérez Caldentey. (2019). Modern money theory (MMT) in the tropics: Functional finance in developing countries. Working paper, Political Economy Research Institute, University of Massachusetts Amherst, 25 September.

  32. 32.

    Wray, L. R. (2019), MMT: Report from the front (part 2), new economic perspectives, 14 October.

  33. 33.

    Wray, L. R. (2015), Modern money theory—a primer on macroeconomics for sovereign monetary systems (2nd edition). Palgrave Macmillan.

  34. 34.

    Wray, L. R. (1998). Understanding modern money: The key to full employment and price stability. Cheltenham: Edward Elgar.

    Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Biagio Bossone.

Ethics declarations

Conflict of interest

Not applicable.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

I wish to thank Massimo Costa and Simon Wren-Lewis for their feedback, and Thomas Palley for his encouragement. I also wish to thank Keith Kuester and David VanHoose for their critical and very constructive remarks. Obviously, I am the only responsible for the views and opinions expressed in the article and for any remaining errors. I am immensely grateful to my wife Ornella for her unremitting support.

Appendix 1: the optimal demand for money

Appendix 1: the optimal demand for money

This appendix develops a modified version of the Allais–Baumol–Tobin (ABT) inventory model for transaction money demand used in the article. The original ABT model is here modified to show that, as discussed in the text, equilibrium in the money market with a growing stock of money requires the government to pay interest to money holders, and that, all else equal, the required interest rate conditional on the government’s policy credibility as perceived by the financial markets. The ABT approach was selected as a simple and yet effective method to incorporate an optimal demand-for-money framework in the economy’s model used in this article. The objective was not to pick the best possible methodology available to derive optimal money but to recognize the role of optimal money demand as instrumental to assess MMT policy tendency toward the creation of "excess" money balances, in a context where money is demanded for its transaction services and based on its opportunity cost.

Suppose households receive nominal income PY at the beginning of each period and spend it evenly during the period. Average wealth is \(PY/2\), and, according to the model assumptions, it is held in the form of money balances, M, earning interest \({i}_{M}\), and foreign asset balances, FX, earning interest \({i}_{RW}\). To finance transactions, households must first hold M balances; thus, before wealth held in the form of FX can be spent, it has to be converted into M at transaction cost f per transaction. Suppose each household divides the period into n subperiods initially placing \(PY/n\) in money balances and the rest in FX. At the end of each subperiod, FX balances are converted into M balances in n − 1 transactions of equal size \(PY/n\). Thus, average money holdings over n subperiods will be \(M=\frac{1}{n}\frac{PY}{2}\) and average foreign asset holdings will be \(FX=\frac{n-1}{n}\frac{PY}{2}\).

Thus, the net gain, \(\Gamma\), from holding wealth in both assets, taking into account the need to finance transactions, is given by:

$$\Gamma =\frac{1}{n}\frac{PY{i}_{M}}{2}+\frac{(n-1)}{n}\frac{PY{i}_{RW}}{2}-(n-1)f$$

Maximizing \(\Gamma\) with respect to n requires:

$$\frac{\partial\Gamma }{\partial n}=\frac{-2{n}^{2}f+PY{i}_{RW}}{2{n}^{2}}-\frac{PY{i}_{M}}{2{n}^{2}}=0.$$

Therefore, the optimal choice for n is


Replacing Eq. (12) into the equation for M, the optimal demand-for-money equation is

$$\frac{{M}^{opt}}{P}\equiv {m}^{opt}=Y\sqrt{\frac{f}{2\left({i}_{RW}-{i}_{M}\right)}},$$

which shows that the demand for money varies positively with real income, the transaction cost, and the interest paid on money balances, and negatively with the rate of return on alternative assets (in this case, foreign exchange).

Consider now that in an LC economy (i.e., characterized by a low \(\beta\)) the expectations that the government mismanages the money supply over the relevant future time horizon induce wealth holders, all else equal, to hold larger shares of their wealth held in FX, as they factor into their portfolio choices the future expected (internal and external) value of money. In particular, they may fear the risk of excess money supply creation and government’s inadequacy or unwillingness to react to threats of currency depreciation and inflation through appropriate policy action. This implies that \({n}_{LC}^{opt}>{n}_{HC}^{opt}\) and, hence, \({m}_{LC}^{opt}<{m}_{HC}^{opt}\). Thus, in order to induce wealth holders in the LC economy and HC economy, respectively, to hold the same level of real money \(m\), all else equal, it must be that \({i}_{M}^{LC}(m-{m}_{LC}^{opt})|{\beta }_{LC}>{i}_{M}^{HC}(m-{m}_{HC}^{opt})|{\beta }_{HC}\).

Similarly, the change in the interest rate required by a given rise of excess money balances will be larger in an LC than in an HC economy, and the difference will grow larger with the rise of excess money balances, that is, \({i}_{M}^{{^{\prime}}LC}(\bullet )|{\beta }_{LC}>{i}_{M}^{{^{\prime}}HC}(\bullet )|{\beta }_{HC}\). These relative differences in the interest rate adjustment will vary inversely with the difference in the level of policy credibility of the economies being compared, that is, the policy space available would be higher, the higher the level of policy credibility of the economy concerned (see Sect. 6).

Formally, all these features are formally captured by the following equation for the interest rate on money:

$${i}_{M,j}={i}_{M}\left({m}_{j}-{m}_{j}^{opt}|{\beta }_{j}\right),$$

with \(j=LC,HC\) and \({i}_{M}\left|{\beta }_{LC}>{i}_{M}\right|{\beta }_{HC}>0;\,{i}_{M}^{{^{\prime}}{^{\prime}}}\left|{\beta }_{LC}>{i}_{M}^{{^{\prime}}{^{\prime}}}\right|{\beta }_{HC}>0;\)\({i}_{M}^{{^{\prime}}{^{\prime}}{^{\prime}}}\left|{\beta }_{LC}>{i}_{M}^{{^{\prime}}{^{\prime}}{^{\prime}}}\right|{\beta }_{HC}>0\), where \({m}_{j}-{m}_{j}^{opt}\) measures "excess" money in country j.

Equation (14) defines the position and shape of the optimal demand-for-money function and indicates that (1) the required real interest rate on money rises with excess money balances and (2) the height and steepness of the demand schedule in the \(({i}_{M},m|\beta )\) space are conditional on the economy’s level of policy credibility, all else being equal. All this graphically represented in Fig. 5 below, which appears as Fig. 1 in the text.

Fig. 5

Stylized demand-for-money schedules in economies with different policy credibility

In light of Eq. (14), and dropping the country index, the optimal demand-for-money function given by Eq. (13) can be written in implicit form as:

$$\frac{{M}^{D}}{P}=m\left(Y, {i}_{M}|\beta ,f,{i}_{RW}\right).$$

With f and \({i}_{RW}\) set exogenously and assumed to be constant (for reasons of simplicity but at no loss of generality), Eq. (15) reduces to:

$$\frac{{M}^{D}}{P}=m\left(Y,{i}_{M}|\beta \right),\mathrm{ with }{m}_{Y }{{^{\prime}}},{m}_{{i}_{M}}{{^{\prime}}}>0,$$

which enters the money market equilibrium MM schedule of Eq. (3) in the text.

Unlike the conventional LM schedule in ISLM analysis, the MM schedule has a negative slope: for a given stock of M and all else being equal, a higher transaction demand for M driven by Y requires a decrease in interest rate paid on M balances, \({i}_{M}\), required to keep the money market in equilibrium.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Bossone, B. Why MMT can’t work. IJEPS (2021). https://doi.org/10.1007/s42495-020-00055-w

Download citation


  • Aggregate demand and output
  • Equilibrium prices
  • Fiscal deficits
  • Interest rate
  • Money
  • Policy credibility
  • Stocks and flows

JEL Classification

  • E12
  • E2
  • E4
  • E5
  • E6