At the Cato Institute’s 40th Annual Monetary Conference, held virtually on September 8, Fed Chair Jerome Powell stated:

We’ve got a dual mandate—maximum employment and price stability—and it comes down to: Is nominal income [GDP] targeting the best way to promote that? We don’t think it is; I don’t think it is. And part of that is that it would be very difficult, I think, to explain to the public the relationship between a nominal income target (nominal GDP target) to those goals. It’s just … a level of complexity that even some economists and policymakers struggle with, let alone the general public. So, it seems like it would be a reach to have that be our fundamental framework.

In other words, Powell is concerned that estimates of the trend growth of output are “highly uncertain” and that a decline in trend growth would require higher inflation. Thus, under a nominal gross domestic product (NGDP) target, he thinks there would be “communication issues” as well as “big chances of policy error because we just don’t know any of the starred variables” (e.g., the equilibrium values of real output, employment, and interest rates).

While Powell’s concern is not only that it may be hard to relate NGDP targeting to the goals of maximum employment and price stability, but also that NGDP targeting is complex and hard to explain. Yet, in June, at a conference held by the European Central Bank, Powell stated: “We now understand better how little we understand about inflation.” If that is true, then the alleged complexity of NGDP targeting should not be a reason to not pursue it given the current monetary system is already so complicated that even the Fed struggles to understand it.

This article counters Fed Chair Powell’s contention that NGDP targeting, although “interesting,” is not suitable to serve as the Fed’s “fundamental framework” for the conduct of monetary policy. He prefers to stay with the current floor system, with its administratively set interest rates on reserves and reverse repos, a large balance sheet, and flexible average inflation targeting (FAIT). Yet, that system itself is complex and hasn’t worked to prevent inflation from rising to a 40-year high. A close examination of NGDP targeting shows that it is not as complex as Powell contends, that the information requirements to implement it are less onerous than under current operating procedures, and that it is superior to either a price-level rule or inflation targeting.

The Complexity Issue

While Powell thinks NGDP targeting is “interesting and works very well in models,” he is reluctant to adopt that monetary framework—primarily because “it seems difficult to implement from a practical standpoint.” His position echoes that of former Fed Chair Ben Bernanke, who in his 2015 memoir wrote: “Nominal GDP targeting is complicated and would be very difficult to communicate to the public (as well as to Congress, which would have to be consulted).”

However, it is not rocket science to tell the public that total spending is going to be kept on a stable growth path of, say, 4 percent—and that such a rule will also be consistent with the Fed’s price stability mandate over the long run. The public should welcome the goal of stabilizing the growth of nominal GDP as a long-run policy objective. If the trend rate of growth of real output is 2 percent, then long-run inflation will end up being about 2 percent. The benefit of an NGDP target regime—preferably a level target that makes up for misses—is that it increases public confidence that the Fed will not move in the wrong direction when supply shocks occur under either a price-level target or an inflation target. Such a regime does not depend on interest rate manipulation or knowledge of potential output. It only depends on the willingness of the Fed to allow the money supply to respond to changes in the demand for money balances, in order to keep NGDP on track. When output growth falls below its potential, inflation would increase until output returns to its long-run, potential growth path.

Clive Crook, former deputy editor of The Economist and now with Bloomberg, has argued that “the simplest way to improve monetary policy” is to adopt an NGDP targeting rule. Drawing on the work of Scott Sumner and David Beckworth, Crook makes a strong case that such a rule would not only make monetary policy more transparent in normal times but even more so in turbulent times, especially now with significant negative supply-side shocks.

The Fed’s ability to create money out of thin air means it can directly affect the money value of total output (i.e., NGDP or aggregate demand). Yet, as Crook notes,

The Fed can’t control how much growth in NGDP is from higher real output and how much is from higher prices, the quantities it cares about. Using NGDP as an intermediate target to guide monetary policy recognizes this. Unlike the ‘dual mandate’ of full employment and stable prices, it acknowledges the limits of what the Fed can do.

The Fed cannot determine real variables—like output and employment—in the long run, but it can set a target path for NGDP. With a 4 percent NGDP growth path, a negative supply shock that reduces output growth by 2 percent would lead to inflation of 6 percent until actual output growth normalized to its 2 percent trend rate. Consequently, Crook recognizes that, under a credible NGDP targeting regime, a firm commitment to maintaining steady growth of NGDP would put guardrails up to fence in longer-run inflation, and the Fed would not have “to take account of every deviation in output and inflation—something that, try as it might, it can’t actually do.” The current inflation fighting episode is a great example of this problem—as in, they can’t account for the spike and haven’t been able to control it.

Like Crook, Carola Binder, an economist at Haverford College, also points to the relative simplicity of adopting a level growth target for NGDP. Writing for the Cato Journal, Binder (2020) argues:

The single, explicit, numerical target associated with NGDPLT [LT indicates “level targeting”] should make it easier for the public to verify whether the central bank is doing what it has promised to do, easier for the central bank to justify the policy stance at any point in time, improving transparency and accountability. The single target will make it harder for politicians to argue that the stance of policy is too easy or too tight.

In sum, there is a strong case to be made for NGDP targeting based on its simplicity and its consistency with long-run monetary stability.

The Information Issue

A major benefit of NGDP targeting is that the information requirements to operate such a regime are less onerous than alternative approaches that require assumptions about the real equilibrium interest rate (known as r*) and the natural rate of unemployment (known as u*). Feedback rules, such as the McCallum Rule, offer superior alternatives to the Fed’s macro-models, which have proven to have large forecast errors.

In a 1987 article, Bennett McCallum, an economist at Carnegie-Mellon University, noted that “the basic idea” underlying the McCallum Rule “is that, since economists do not understand how nominal demand changes are divided between inflation and output growth, the most useful thing that monetary policy can accomplish is to keep nominal demand growing smoothly at a noninflationary rate.” To do so, he selected the monetary base (i.e., currency plus bank reserves) as the primary instrument of monetary policy because the Fed has direct control over that variable. In his specification, the Fed would increase the velocity-adjusted base money supply whenever NGDP is below its long-run (noninflationary) trend, and vice versa. The monetary authority would have to estimate the correct value of the response/​adjustment parameter to keep NGDP on a smooth growth path. However, the information requirement of his feedback rule is less demanding than in the case of Taylor-type rules relying on estimates of potential output and r* (see McCallum 2000). Indeed, as Lawrence H. White (2022) points out, the McCallum Rule (as opposed to the Taylor Rule) “is agnostic about the role of interest rate changes in monetary policy transmission.”

Finally, Beckworth and Henderson (2016) find that “nominal GDP targeting reduces the volatility of the output gap [i.e., the difference between actual and potential output] and inflation in comparison to the case in which the central bank uses a Taylor rule with imperfect information about the output gap.” Beckworth (2019) elaborates by showing that with the Taylor Rule, the Fed needs “to know both real GDP and potential real GDP in real time.” However, under an NGDP targeting regime, the central bank only needs to know “NGDP in real time.” Thus, “the information requirements are … greater for the standard Taylor Rule than for an NGDP target.”

In sum, by eliminating the necessity of estimating unobservable real variables, like r* and u*, an NGDP targeting regime reduces the information needed to conduct monetary policy and keep the economy on a steady growth path.

The Comparative Advantage Issue

In selecting a monetary rule, Niskanen (1992) pointed out three options: (1) “maintain a path of the price of some specific commodity such as gold or some broader price index”; (2) “maintain a path of some monetary aggregate such as the monetary base or M2”; and (3) “maintain a path of some measure of total demand in the economy such as nominal GNP or domestic final sales.” He argues that “any one of these rules would be better than guidance based on interest rates or exchange rates, or on any real variable such as the growth of output or the level of the unemployment rate.”

His preference, however, is for a demand rule that would target domestic final sales (rather than NGDP). According to Niskanen:

A demand rule is superior to a price rule because it does not lead to adverse monetary policy in response to unexpected—either favorable or unfavorable—changes in supply conditions. Similarly, a demand rule is superior to a money rule because it accommodates unexpected changes in the demand for money [and hence, in monetary velocity].

Likewise, Peter Ireland (2022), a member of the Shadow Open Market Committee and professor of economics at Boston College, notes that an NGDP level targeting rule would improve upon flexible average inflation targeting (FAIT) as practiced since 2019. Persistent inflation below the Fed’s desired rate of 2 percent led the Federal Open Market Committee (FOMC) to adopt FAIT in order to boost inflation expectations and circumvent the zero lower bound—namely, the problem of using conventional monetary policy to counter recessions when nominal interest rates reach zero.

The idea was to make up for shortfalls in hitting the Fed’s policy rate by allowing inflation to move above 2 percent, until average inflation was consistent with the Fed’s longer-run target of 2 percent. However, the Fed was silent on how high, and for how long, it would allow rates to rise before reversing course. Moreover, FAIT was not symmetric: the Fed did not say anything about how to proceed if it overshot the 2 percent target. With today’s high inflation, which the Fed no longer sees as transitory, the ambiguity of FAIT is obvious. Ireland therefore argues that NGDPLT would be a much better guide for sound monetary policy:

Nominal GDP targeting will restore balance and symmetry to the FOMC’s strategic framework. Focusing on a nominal GDP target will help FOMC members remind the public—and themselves—that monetary policy creates the most favorable environment for economic growth and stability when it aims to stabilize a nominal variable first.

Although NGDP targeting is not a forecast-free endeavor, it stands a better chance of mitigating business cycles and achieving long-run price stability than alternative rules in a fiat money system. Indeed, the Fed implicitly followed a demand rule from early 1992 through early 1998, by keeping aggregate demand, as measured by final sales to domestic purchasers (FSDP), on a steady growth path of 5.5 percent per year, despite supply-side shocks. The Fed departed from that path from the first quarter of 1998 through the second quarter of 2000 by allowing demand to increase at an annual rate of 7.3 percent, resulting in CPI inflation increasing by about 2 percentage points relative to 1998 (Niskanen 2001).

Niskanen’s test of the demand rule strengthens the case for NGDP (or FSDP) targeting. However, if such a rule is to be adopted, Niskanen favors a McCallum-type feedback/​recursive rule that minimizes the need for relying on estimates for r* and an output gap. Moreover, he does not fully support NGDP level targeting, because returning to the prior path of demand after a sharp increase would be costly:

My position is that the Fed should not try to offset the high demand growth of the period from early 1998 to mid-2000. This would require unusually low demand growth for a period until demand was back on its prior trend; the future output costs of such a policy seem higher than any value of reversing the prior price-level effects of high demand growth. Maintaining a demand rule, I suggest, requires offsetting small differences from the target path but not as large a difference as has developed since early 1998. Instead, the Federal Reserve should try to establish and maintain a new 5.5 percent demand growth path based on the current level of demand [Niskanen 2001].

Adoption today of a level target for NGDP would not require the Fed to put the economy through a difficult adjustment process by reversing all of the excess nominal GDP growth we’ve had over the last two years. But, adopting the level target today would still make clear that nominal GDP growth will have to come back down to trend going forward, and with it, inflation would return to more normal levels.

Conclusion

Shifting from a floor system with a dual mandate to an orthodox corridor system with NGDP targeting, downsizing the Fed’s balance sheet, and eliminating interest on reserves would improve monetary policy—making it simpler, more transparent, and thus more credible. Keeping NGDP on a steady path has been shown to improve economic performance and help mitigate business fluctuations.

The Fed’s forecasts, based on complex economic models of the economy, have been error-prone for years. A McCallum-type monetary rule would minimize those errors and improve the probability of obtaining a smoother path for NGDP growth.

Even so, NGDP targeting does face the problem that, in the post-2008 operating system, the link between changes in base money, broader monetary aggregates, and nominal GDP have weakened, because the Fed can pay interest on reserves (IOR). If that interest rate increases the demand for holding excess reserves at the Fed rather than lending them out, the money multiplier will weaken.

In 2012, John C. Williams, then president of the San Francisco Fed, recognized that change in the monetary framework when he wrote:

In a world where the Fed pays interest on bank reserves, traditional theories that tell of a mechanical link between reserves, money supply, and, ultimately, inflation are no longer valid. In particular, the world changes if the Fed is willing to pay a high enough interest rate on reserves. In that case, the quantity of reserves held by U.S. banks could be extremely large and have only small effects on, say, the money stock, bank lending, or inflation.

Hence, to be more effective as a tool of monetary policy, NGDP targeting needs to be preceded by changing the Fed’s operating system from a leaky floor system to an orthodox corridor system, as recommended by George Selgin (2020). Eliminating interest on reserves and making those reserves scarce again would strengthen the money multiplier. The link between changes in base money and NGDP would significantly improve the operation of a McCallum feedback rule. Keeping NGDP on a level path would improve economic performance and restore credibility to the Fed’s goal of achieving long-run price stability.

In the meantime, the Fed should take the advice of Harvard economist Jeffrey Frankel (2019) and include NGDP in its Summary of Economic Projections (SEP). As Frankel notes:

This seems a useful idea even if the governors and district presidents who fill out the SEP table simply were to derive their projected NGDP growth numbers by taking the sum of the real growth row and the inflation rate row of the table (though inflation in the SEP table is currently the PCE deflator, not the GDP deflator).

One thing is certain: the Fed needs to listen to those arguing for fundamental reform and not just tinker with the flawed floor system, embedded in a discretionary government fiat money regime (see Dorn 2017). Nominal GDP targeting should be a prominent part of that discussion.