Book club: Scott Sumner’s ‘Alternative Approaches to Monetary Policy’
Part 2, Chapters 6-8: New Keynesian, MMT and libertarian approaches to monetary policy
In Part Two of AAMP, Sumner critiques interest rate-oriented monetary economics, modern monetary theory (MMT) and libertarian approaches to monetary policy, contrasting them with his own price of money approach.
Chapter 6 reproduces his Mercatus working paper ‘A Critique of Interest Rate-Oriented Monetary Economics,’ which has featured in this space previously. The main line of argument is one already established in previous chapters: changes in interest rates are generally the effect rather than the cause of monetary policy and there is no necessary correlation between changes in interest rates and the stance of monetary policy.
Sumner’s argument derives unintended support from debates within mainstream macroeconomics between New Keynesians and NeoFisherians, who cannot agree on the sign on the relationship between monetary policy, interest rates and inflation. As Sumner rightly notes, this disagreement is an embarrassment to modern macro.
It is also strongly suggested by the way in which modern macro has struggled to identify the role of monetary policy when proxied by changes in official interest rates. Workhorse vector autoregression models have been plagued by interest rate and price ‘puzzles’ that seemingly yield perverse or contrary-to-theory outcomes from changes in official interest rates. As Sumner notes, ‘the Fed can raise interest rates with either expansionary or a contractionary monetary policy.’ Sumner cleverly uses exchange rate dynamics to illustrate the nature of the identification problem and its resolution through the price of money approach.
In my view, this material flows more naturally from Chapters 1 and 4, which taken together would work better as Part 1 of the book. Part 2 would then work better as applications of the price of monetary approach to the specific issues and problems raised in Chapters 2, 3, 7 and 8.
Modern monetary theory
In Chapter 7, Sumner takes on modern monetary theory, or MMT. The core proposition of MMT is to deny that monetary policy determines the path of aggregate demand. It is an institutionally-contingent special case of the more general fiscal theory of the price level (FTPL), in which fiscal policy determines the price level, an idea well established in mainstream macro. Indeed, John Cochrane has recently published a book on the FTPL which in its core propositions is little different from MMT, although he draws very different implications for public policy. In both Cochrane’s version and MMT, money and bonds are viewed as government liabilities and therefore close substitutes, whereas monetarists have always rejected the latter proposition, highlighting the unique properties of money, as well as the very different institutional arrangements governing both types of liability.
As Sumner notes, MMTers share with mainstream macro a preoccupation with interest rates as the instrument of monetary policy, in fact, even more so. They treat variables as either inherently endogenous or exogenous without recognising that these are not just theoretical but also policy and institutional choices. Many of the key propositions of MMT are tautologically or arbitrarily true, rendering them devoid of economic content.
My favourite joke about MMT is that it is analogous to the question of whether you can microwave a fork. Yes, a fork will fit inside a microwave. And, yes, you can switch the microwave on with a fork in it. MMTers talk about the logic behind all the steps leading up to the pressing of the start button. The question is not whether you can do MMT, but whether you would want to.
My other favourite joke about MMT is that it is the theory that the 1970s never happened. Prior to 1982-83, when the RBA used a tap rather tender system for bond issuance and the exchange rate was an administered price, Australia arguably lived in a MMT world. Fiscal policy largely determined the price level, the government used price and wage controls to manage inflation and credit rationing was used to allocate saving and investment. We junked all of those institutional arrangements for very good reasons.
MMT has fallen out of favour in the last few years because the large fiscal expansions associated with the pandemic, accommodated by monetary policy, proved to be inflationary in ways that MMTers assumed could be easily managed by the political process, despite all evidence to the contrary. Instead, the Fed has carried the burden of inflation control. Sumner also points to the Great Monetary Policy experiment of 2013, in which the US engaged in the biggest fiscal consolidation since World War Two while the economy continued to expand, thanks to accommodative monetary policy, all of which goes against the expectations of MMT.
MMT gained much more traction in the pre-pandemic period when inflation and interest rates were low and questions were being raised about the effectiveness of monetary policy, not least by central banks themselves. As I noted when RBA Governor Philip Lowe argued against MMT, if central bankers fail to deliver on their mandates, politicians will step in to do their job for them. It is no coincidence that during the pre-pandemic period, we saw widespread calls for more activist fiscal policy and even mainstream calls for a return to centralised wage fixing in Australia. By failing to meet its inflation target for years on end, the RBA not only undermined confidence in the inflation target and monetary policy, but also the broader institutional foundations of post-reform era economic policy (a point I made to the RBA review).
Libertarian approaches
Chapter 8 is a critique of libertarian monetary economics. There is no single libertarian approach to monetary policy. Indeed, monetary economics has been a massive intellectual fault line within the Mont Pelerin Society (as both Sumner and I can attest!) Libertarians not surprisingly view discretionary fiat money regimes with suspicion and highlight the knowledge problem facing central bankers in setting monetary policy. Their proposed solutions are typically based around removing discretion (e.g, a gold standard, or Friedman’s k-percent rule) or endogenising the money supply through private money (e.g, Selgin-White free banking, or the Black-Fama-Hall system separating the medium of exchange from the unit of account).
I have found much of the libertarian discourse around monetary economics to be misdirected, a product of the high inflation 1970s, in which many of the proponents of that discourse came of age. Since then, central banks have been given greater independence and made greater use of rules-based approaches to monetary policy, including inflation targeting. Far from running inflationary policies, central banks have, at least until recently, erred mainly on the side of undershooting their inflation targets. Libertarians assume central banks have an inflationary bias, but the historical record suggests they are just as prone to presiding over disinflationary or deflationary episodes.
Libertarians are also very concerned with the fiscal-monetary nexus. In particular, they fear fiat monetary regimes will be used to monetise government debts. That is always a possibility, but typically also requires extensive financial repression, which should be the bigger concern for libertarians. The FTPL in its non-MMT form seeks to establish the conditions under which fiscal policy might dominate monetary policy in the determination of the price level. But this concern is really about fiscal not monetary institutions. Afterall, the FTPL is about what happens when existing monetary institutions and fiscal-monetary interactions break-down and no longer discipline the fiscal authority. Some libertarians are advocates of the gold standard or fixed exchange rate regimes because they see these regimes as a stronger discipline on the fiscal authority. That sometimes works, but only serves to relocate rather than fundamentally solve the problem of disciplining the fiscal authority. If given monetary institutions are indeed a hard constraint on fiscal irresponsibility, they are unlikely to be robust to a government determined to be fiscally irresponsible.
Sumner often frames his advocacy of nominal GDP targeting as a choice between a little more inflation or a lot more socialism. This trade-off is greatly underappreciated by most libertarians. When central banks undershoot their inflation targets or nominal GDP comes in lower than expectations over extended periods, the disappointment of expectations for nominal or real income growth will often be channelled into demands for more government intervention through the political process. Bad monetary policy is often blamed on the operation of free markets. As noted above, this is exactly what happened during Australia’s leaning against the wind episode between 2016 and 2019, when there were growing demands for more activist fiscal policy and a return to centralised wage fixing to address weak wages growth.
Much of the libertarian discourse around monetary economics is not only misplaced, but actually harmful to its own cause. Libertarians’ preferred monetary regimes are more often than not an attempt to impose a technical fix on more fundamental problems of political economy. The Bitcoin white paper is perhaps the ultimate example of this, even though its proposed fix is not very different from earlier libertarian proposals to freeze the stock of base money and impose a productivity norm for the price level (which is not to deny that distributed ledger technology is a potentially useful innovation).
By contrast, market monetarism can be seen as something of a half-way house between free and central banking, in which monetary policy is endogenised through market mechanisms, but in the context of existing central bank institutions, the modern ubiquity of which argues for their presumptive efficiency. From a macro and financial stability perspective, there is a lot to be said for a government-supplied reserve medium the supply of which is perfectly elastic when needed. Perhaps the principal advantage of market monetarism is that its proposed monetary regime (nominal income futures targeting) can be readily implemented within the framework of existing monetary and fiscal institutions, while addressing the key issue identified by libertarians: the need for a market-based solution to the knowledge problem faced by central bankers in maintaining monetary equilibrium.
It is perhaps also worth noting that market monetarism serves as a bridge between libertarians and progressives. Indeed, many market monetarist insights have been absorbed into supply-side progressivism (aka neo-liberalism). Employ America’s advocacy of gross labour income targeting is an obvious example. Nominal income targeting reconciles the price stability and full employment mandates that have traditionally been a point of division between libertarians and conservatives on the one hand and social democrats/progressives on the other.
That concludes our book club. Hopefully this discussion inspires you to read Scott’s book and his other work as well. For me, it has been great to watch Scott’s idea progress from blog posts to systematic, book-length treatments and to have played a small role in that process by publishing one of the first long form treatments of Scott’s thesis 10 years ago.
ICYMI
US credit default swaps are more expensive than contracts on the bonds of Greece, Mexico and Brazil.
A handy primer on the 14th Amendment to the US constitution, arguing that ‘the debt limit is unenforceable.’ But as the authors note, the constitutionality of the debt limit is beside the point: ‘even if our argument — or a cleverer constitutional one — is a slam-dunk case, any course of action that leaves the status of a $25tn global market and the livelihoods of millions of people to the vicissitudes of constitutional litigation would be enormously risky…This risk, and the prospect of annual recurrence, makes imperative a new political settlement that includes eliminating the debt limit once and for all.’
Jim Overdahl on why short selling bans are counter-productive.
Economics is one of the least upwardly mobile academic fields. Compared with political science or sociology programs, an economist’s doctoral program is more likely to determine their career path.
Did the RBA perform better than its peers? No, it came stone-cold, motherless last.
The Bid to Dethrone the Dollar. The greenback’s dominance is here to stay. Here’s why.
Immigrants & Their Kids Were 70% of U.S. Labor Force Growth Since 1995.
Machines of mind: The case for an AI-powered productivity boom.
Vanguard’s new superannuation offering has 10,000 members and $500 million under management (I am one of them). Small beginnings, but they say they have a ten year horizon and are looking at acquisitions.