It is customary for Brian to write a post every year explaining the work of the Nobel Prize winner(s) in economics. Brian does a great job with these and thus I should feel no obligation to weigh in. Nonetheless, I occasionally can’t resist (see here and here). Because I have pretty idiosyncratic preferences, I especially like to write posts highlighting the work of the winners that didn’t get recognized by the committee. Today, I’d like to highlight a paper by Bob Hall in which he thinks carefully about how to design a purely fiduciary (what many call “fiat”) monetary system. Initially, I wanted to write this post because so much of what I saw written about Hall stressed his work on consumption and investment. But the more I thought about the paper, the more I thought the paper might be a good way to illustrate the connection between Hall and fellow winner Michael Woodford.
Money and Price Theory
There is a weird dichotomy in monetary economics. In fact, it is this dichotomy that originally attracted me to monetary economics in the first place. It’s even evident when I write about money (which I seem to do a lot) for the newsletter. Occasionally, people will ask me why I write about money for a price theory newsletter. I understand the perspective they are communicating. When people think price theory, they think microeconomics. When people think money, they think macroeconomics. What therefore does money have to do with price theory? Even within the monetary literature, there is generally a distinction between optimal monetary policy, using some type of efficiency criteria, and optimal monetary policy, using some type of macroeconomic stabilization criteria. One can stretch and say these are both about efficiency, but they are about two distinct types of efficiency.
In fact, what initially sparked my interest in monetary economics (although I probably didn’t know enough to articulate it at the time) is what appeared to be a disconnect. When we think about policy in almost any other area of economics, the general approach is to think about how the market would typically function. If there is some friction that prevents the market from getting the efficient allocation, then there is potentially a role for policy. This is the approach taken by Milton Friedman, for example, in his “Optimum Quantity of Money.” However, much of the rest of the literature focused on monetary policy as optimal stabilization policy. According to this view, the primary purpose of monetary policy is to minimize fluctuations in the economy. (If you believe that business cycles are inefficient, this is sort of similar, but there was another debate about how much of business cycles were inefficient and how much were efficient responses to random shocks. I’d rather not discuss that here.)
Friedman was one of the rare examples of people who not only did both, but integrated the two approaches together. For example, his k-percent rule (which called for the money supply to grow at a fixed rate of k percent per year) was based on the idea that monetary policy largely did more harm than good when conducting stabilization policy. In his “Optimum Quantity of Money,” he pointed out that his two proposed rules for monetary policy were perfectly consistent with one another. This consistency is now an implicit characteristic of a lot of modern models that focus on efficiency from the price theoretic perspective.
Another economist who did both was Bob Hall. In his paper, “Optimal Fiduciary Monetary Systems,” he attempts to address both the price theoretic approach to policy that mitigates the deadweight loss and the macroeconomic stabilization approach. In doing so, Hall contributed to a literature that was then called “New Monetary Economics.” Although many lament that this literature largely went stagnant over time, I am going to argue that Hall, in particular, had a significant impact in how we think about monetary policy today with his contribution.
Optimal Fiduciary Monetary Systems
As I have written about many times by now, many people fail to appreciate the rarity of the current monetary system. Although there are many exceptions, for a large amount of recorded human history, people traded physical objects (like gold and/or silver) as a medium of exchange that had positive value distinct from the value associated with the role as a medium of exchange. Even when paper money was introduced, the paper was often redeemable for these commodities.
When the Nixon administration closed the gold window, it did so as a temporary measure. The closing of the gold window was meant to bring other members of the Bretton Woods system to the negotiating table. Of course, the gold window never re-opened. The rarity of the current monetary system was thus much more appreciated in the years that followed. As time when by, many economists began to speculate about how to think about this system in which paper dollars circulated and were not redeemable for anything. This was a period of big contributions to monetary theory (at least in my view). People like Ben Klein, Earl Thompson, and Bob Hall wrote important contributions that helped one think about the uniqueness of our present monetary regime.
Bob Hall was careful to frame his work around “fiduciary” money. He did not call it fiat money. His stated reason for this is that U.S. law doesn’t require anything to be priced in terms of dollars or that contracts be written in dollars, but the dollar is nonetheless used for these purposes.
Hall identified several problems with our system of fiduciary money as he saw it at the time. The first is that fluctuations the demand for such money leads to instability in the price level as well as corresponding fluctuations in real economic activity. At the same time, the existing regulatory environment led to a lot of inefficiencies at the micro level. Meanwhile, although deregulation efforts could reduce or eliminate these inefficiencies, such deregulation would seem to exacerbate the macroeconomic instability of the system.
Given those challenges, how could one design an optimal system?
Hall begins by arguing that the critical functioning of this system is that there is a central bank that produces what he calls “reserve certificates.” Everything else is essentially redeemable for reserve certificates. He begins with issues that are of particular interest to price theorists. He argues that for such a system to minimize the deadweight loss inherent to the system, one would need to create higher-denomination reserve certificates or account-based reserve certificates coupled with deregulation sufficient to allow private banks the ability to supply both deposits and currency to the market. Similar to both Earl Thompson before him and David Glasner after him, Hall argues that this would lead to a perfectly elastic supply curve for claims to reserve certificates. As a result, the supply of claims to reserve certificates would be entirely demand determined thereby eliminating an inefficiency and also mitigating a source of macroeconomic fluctuations. The other element of tackling the inefficiency is for reserve certificates to earn interest. This eliminates the inefficiency typical of irredeemable paper money that pays no interest.
Nonetheless, this raises the question of how to set the interest paid on reserve certificates in order to promote macroeconomic goals, such as price stability and minimizing fluctuations in output.
Suppose that we normalize the price level to 100. Hall proposed having the reserve certificates pay the market rate of interest (on bonds) plus a premium that would be a function of the percent to which the expected future price level was above 100. Thus, when the price level was expected to rise (relative to 100), the interest rate on reserve certificates would automatically rise. The extent to which the expected price level fell below 100, the interest rate on reserve certificates would decline below the market rate. This created an automatic policy rule. What Hall was able to show with a couple of simple models is that such a rule would ensure that the price level only deviates from that initial price level when there are random shocks that hit the economy. Otherwise, the price level would be constant. Furthermore, his models imply that the more responsive that the interest rate on reserve certificates is to deviations in the price level, the smaller the effect random shocks will have on the price level and the faster that output adjusts back to its equilibrium level.
As a result, the model demonstrates how one can think about an optimal fiduciary monetary system both in terms of the efficiency concerns of price theorists and in terms of common macroeconomic objectives of central banks.
The Connection with Woodford and the Nobel Prize
Michael Woodford, who shared the award with Hall, is mostly known for his work on the so-called New Keynesian model. For better or worse, this has become the workhorse model for monetary policy analysis. Meanwhile, if you were to ask someone what Bob Hall is mostly known for, you would probably get nearly as many answers as the number of people you asked. In linking the two together, a lot of people have linked Hall’s work on consumption with the household component of the New Keynesian model. I think that there is more to it than that.
Not only have I seen a lot of people remark that the Hall and Michael Woodford are an odd pairing for the award, but I have also seen a number of people remark that it is odd that the award specifically mentions “macroeconomic stabilization policy.” However, I think that the paper I have highlighted in this post represents a link to the work they’ve done. The emphasis on feedback rules for interest on reserves that target the forecast, written in 1983, was clearly ahead of its time. And the central themes of that paper are themes throughout Woodford’s own work encapsulated in his book, Interest and Prices. While it is true that the paper only has 141 of Hall’s 77,000+ citations, it’s also important to note that three of those citations come from Michael Woodford and, in particular, the paper is cited in Woodford’s paper “Monetary Policy in a World Without Money,” which is the paper that kicked off this branch of Woodford’s work. Woodford references Hall specifically (and almost exclusively) as a precursor in thinking about the monetary system in the same way. In fact, where Woodford departs from Hall is that while Hall focused on keeping the price level constant, the feedback rule can be much more generalizable to Taylor-type rules for monetary policy. (And thus a shame that John Taylor wasn’t included in this award.)