This policy brief is part of a Mercatus Symposium titled “What Would Milton Friedman Say?” The symposium explores what the late Nobel laureate, economist Milton Friedman, might say about monetary policy today, as the Federal Reserve grapples with increasing inflation in the wake of the COVID-19 pandemic.
In response to the COVID-19 pandemic, the United States saw a simultaneous expansion of both fiscal policy and monetary policy. Although it is unclear what Milton Friedman would have said about these particular pandemic-related policies, Friedman’s views on monetary and fiscal policy coordination were quite clear. Throughout his career, he advocated rules-based policy and thought that coordination between monetary and fiscal policy was unnecessary.
Those who are familiar with the totality of Friedman’s work might want to quarrel with the certainty of that statement. After all, in his early days as an economist, Milton Friedman wrote a policy proposal explicitly calling for coordination between monetary and fiscal policy. On the surface, his view that fiscal policy is largely ineffective and his overall skepticism of countercyclical policy seem to run counter to that work in his early career. However, a careful reading of Friedman’s work reveals that the consistent theme throughout his career is a commitment to rules-based policy based on the belief that discretionary policy tended to result in more harm than good. What did change over the course of his career is that his work on money convinced him of both the primacy of monetary policy and the irrelevance of fiscal policy due to what we would now call the monetary offset.
Friedman is perhaps best known for things like the permanent income hypothesis, his work on monetary history, the Friedman rule (the idea that the opportunity cost of holding cash should be equal to zero), and the k-percent rule for money growth. However, early in Friedman’s career he wrote a proposal for a coordinated rules-based approach to monetary and fiscal policy.
This proposal called for the government to (1) require that banks hold 100 percent reserves, (2) determine its spending by the preferences of the public, (3) maintain a rules-based system of government transfer payments, (4) collect taxes primarily through a progressive income tax, (5) maintain a balanced budget during “normal” times, and (6) eliminate the use of public debt. It is the last two characteristics that created a link between monetary and fiscal policy.
Under Friedman’s proposal, deviations between government revenue and government spending would primarily be driven by fluctuations in economic activity. If nominal income fell below trend, the government would tend to run a budget deficit since tax revenue would tend to decline and transfer payments would increase. Because the government would not use debt, the deficit would be financed entirely by money creation. Conversely, when nominal income increased above its trend, tax revenue would tend to be higher and government expenditures would decline due to the reduction in transfer payments. The government’s budget surplus would then be used to remove money from circulation.
The proposal created a clear coordination between monetary and fiscal policy. On average, the government would maintain a balanced budget. During downturns in the business cycle, this proposal would not require reductions in government expenditures in response to declining tax revenues. Instead, the money supply would automatically increase to cover the deficit. This expansion of the money supply would push nominal income higher and back toward its trend. As nominal income returned to its trend, tax revenue would rise, transfer payments would decline, budget balance would be restored, and the monetary expansion would cease.
During economic booms, when nominal income was above its trend, tax revenue would rise and transfer payments would decline. The resulting government surplus would result in a monetary contraction that would tend to reduce nominal income toward its trend. In the process, tax revenue and transfer payments would return to a normal level and the government’s budget would be balanced.
This rules-based proposal created an automatic coordination between monetary and fiscal policy that, in theory, would create greater long-run economic stability by reducing or even eliminating short-run fluctuations in economic activity. Given Friedman’s later views, those who are unfamiliar with his early work might be surprised to find out about this particular proposal. For most of his career, Friedman was critical not only of fiscal policy but of countercyclical policy more generally. In fact, just 12 years after he published this proposal, Friedman published A Program for Monetary Stability in which he advocated a rule for monetary policy that would require that the central bank commit to growing the money supply at a constant rate. One might therefore wonder how Friedman’s views changed so significantly.
In a sense, Friedman’s views did change. However, the change in his views is more subtle than a simple juxtaposition between this early proposal and the k-percent rule might suggest. What changed is that Friedman’s interpretation of the empirical evidence convinced him of the ineffectiveness of fiscal policy due to the primacy of monetary policy. What remained the same was that both Friedman’s proposal for monetary and fiscal policy coordination and his k-percent rule were motivated by his goal of long-run economic stability and his belief that discretionary policy tended to be a significant source of economic fluctuations.
It is important to note that although Friedman’s 1948 proposal called for monetary and fiscal policy coordination, the role of countercyclical policy was entirely filled by monetary policy. Nonetheless, fiscal policy did play a role through automatic stabilizers and transfer payments. Thus, what is to be explained is twofold. First, why did Friedman subsequently view fiscal policy as largely irrelevant? Second, why did Friedman abandon countercyclical monetary policy in his later advocacy of monetary rules?
The answer to the first question is that Friedman was motivated by his empirical work on money and then key events in the 1960s demonstrating the primacy of monetary policy. According to Friedman, the difficulty in evaluating the effectiveness of either monetary policy or fiscal policy is that the two rarely operate independent from each other. If monetary and fiscal policy are both expansionary and there is an increase in output and prices, to what extent can these effects be attributed to monetary policy and to what extent are they due to fiscal policy? Perhaps some credit should go to both types of policy, or perhaps one type of policy was dominant.
In a speech in London in 1970, Friedman proposed a way of evaluating the relative importance of monetary and fiscal policy. He argued that one could examine periods when monetary policy was expansionary and fiscal policy was contractionary and periods when monetary policy was contractionary and fiscal policy was expansionary. A comparison between such periods might help to determine which type of policy dominated. He then provided two examples.
The first example was from 1966. Fiscal policy was expansionary, whereas the money supply was effectively constant for most of the year. Keynesians and Monetarists disagreed on what the likely outcome would be, with Keynesians seeing fiscal policy as dominant and Monetarists seeing monetary policy as dominant. The Monetarists were proven correct when real economic activity declined in 1967 and only began to recover when the Federal Reserve reversed course.
The second example was from 1968 when the federal government instituted a 10 percent surtax on income. Many, including the Federal Reserve, expected this to produce a drag on economic activity. As a result, monetary policy was expansionary and the expected decline in economic activity was avoided.
Friedman’s interpretation of these examples was that they illustrated the dominance of monetary policy. What Friedman was articulating in these examples is an early version of what Scott Sumner has more recently called the monetary offset. According to this view, monetary policy effectively offsets any fiscal action. When fiscal policy is expansionary, this leads to expectations of higher output and prices, which tends to make monetary policy tighter than it otherwise would have been. As a result, expansionary fiscal policy has no effect on economic activity. On the other hand, when fiscal policy contracts, this is expected to result in a reduction in economic activity and disinflation. The central bank responds to these expectations by engaging in more expansionary policy. Again, fiscal policy appears as though it has no impact on economic activity. Monetary policy is dominant. If one believes that the monetary offset is supported by the data, it is not surprising that one would draw the conclusion that fiscal policy is largely irrelevant.
As a corollary to this argument, Friedman came to think of his original proposal as a purely monetary proposal. He acknowledged that fiscal policy that was accommodated with money creation would be expansionary. However, he explained that “if the government prints money to meet its bills, that is monetary policy.” Thus, for Friedman, monetary accommodation of fiscal policy was by definition monetary policy. This follows naturally from his view of monetary policy dominance.
That this shift in Friedman’s thinking was driven by his own empirical work is something that he was explicit about. In an interview with John Taylor, he explained:
"In the earlier paper, I was at the point where I would say money is important but the quantity of money should vary countercyclically—increase when there was a recession and, the opposite, decrease when there was an expansion. Rules for taxes and spending that would give budget balance on average but have deficits and surpluses over the cycle could automatically impart the right movement to the quantity of money."
"Then I got involved in the statistical analysis of the role of money, and the relationship between money and money income. I came to the conclusion that this policy rule was more complicated than necessary and that you really didn’t need to worry too much about what was happening on the fiscal end, that you should concentrate on just keeping the money supply rising at a constant rate. That conclusion was, I’m sure, the result of the empirical evidence."
Thus, it is not so much that Friedman’s views on the effects of monetary and fiscal policy coordination had changed but rather that monetary policy could accomplish the same outcome on its own. Monetary policy was dominant and therefore fiscal policy was irrelevant. Perhaps more important in this quote is that Friedman viewed the k-percent rule as the natural evolution of his thinking over time. This might seem odd. His original 1948 proposal was not just about monetary and fiscal policy coordination; it was about countercyclical policy. He had called for monetary expansion during downturns and monetary contraction during booms. How could a rule that called for a constant growth rate in the money supply be a natural extension of a rule that called for countercyclical changes in the growth of the money supply?
The answer is found from a careful reading of Friedman’s original proposal. In that paper, Friedman notes that the desirability of the proposal is that “it largely eliminates the uncertainty and undesirable political implications of discretionary action by governmental authorities.” Furthermore, he notes that the proposal is not one that people “would consider optimum if our knowledge of the fundamental causes of cyclical fluctuations were considerably greater than I, for one, think it to be; it is a proposal that involves minimum reliance on uncertain and untested knowledge.”
In this description of the proposal, one hears echoes of Friedman’s case for the k-percent rule for monetary policy. His advocacy of a constant growth rate for the money supply was not motivated by his belief that a constant rate of money growth was the optimal policy but rather by his desire to minimize significant policy mistakes due to discretionary policy. Thus, given that his original policy proposal was designed to limit discretionary policy and given his interpretation of the evidence on the monetary offset, it becomes clear why he believed that the k-percent rule was a natural extension of his thinking.
Overall, what this reveals is that Friedman’s view on the coordination of monetary and fiscal policy was that it was unnecessary. Since monetary policy was dominant, the stance of fiscal policy was largely irrelevant. Why resort to the coordination of monetary and fiscal policy when monetary policy alone would do just fine? Furthermore, the one consistent theme in Friedman’s policy advocacy over the years was the need for rules-based policy. These rules need not be complicated. In fact, the simpler, the better. Friedman viewed the costs of mistakes made through discretionary policy as a significant source of economic fluctuations. As a result, policy rules did not need to be optimal in the sense that they minimize all economic fluctuations. Instead, the policy rules should be designed to eliminate policy-induced fluctuations that came from the discretionary actions of policymakers.