The famous R-star (r*), also known as the neutral, the natural or the equilibrium real rate of interest, is the real policy rate that is construed as neither restrictive nor accommodative in a full-employment economy. Given the economy’s resilience to the Fed’s aggressive hiking campaign, many have recently argued that the r* has materially risen. We are not convinced.
The US economy’s resilience and the persistent above-target inflation in the face of higher interest rates have led many to argue that the r* has risen materially from pre-COVID-19 levels. In a related context, we had earlier argued that the current economic resilience in the United States (US) is artificially augmented by forces that have already started to fade. The implication of this, in our view, is that the resultant economic slowdown will cool down the speculation around a much higher r*.
Ironically, a permanently higher term premium—which global geoeconomic fragmentation and sharply higher debt levels may induce—could, in fact, argue for a decline in r*, all else being equal.
Jerome Powell, Chair of the Federal Reserve, recently noted that “we cannot identify with certainty the neutral rate of interest, and thus there is always uncertainty about the precise level of monetary policy restraint.” Frustrating as it may be, this is indeed the case. Just as the equally famous non-accelerating inflation rate of unemployment (NAIRU), the neutral rate is a theoretical concept that cannot be measured in real life. It can only be assessed retroactively in light of an economy’s performance.
If growth slows sharply, absent other exogenous factors or shocks, it is concluded that the neutral rate is below the prevailing policy rate, and vice versa. This is what Chair Powell meant when, referring to r*, he stated “we know it by its works.” The question is: Over what time horizon should those “works” be assessed?
Given the long and variable lags to monetary policy, extended periods of deviations from expected policy outcomes are required before we should even consider the possibility that the neutral rate has changed. Indeed, post COVID-19, deviations from equilibrium levels of growth and inflation have persisted for more than two years—not just in the US but also elsewhere. This means, at first glace, the durability test will appear to be satisfied.
However, there have been too many exogenous factors skewing macroeconomic outcomes for us to be able to credibly claim that the deviations are—whether entirely or even primarily—a function of interest rates. Supply chain dynamics, fiscal policy, sharp swings in immigration, labor market churn, and opportunistic pricing behaviors are but a few of the exogenous forces at play distorting the assessment of the impact of monetary policy.
If even judging r* by its “works” is not an accurate exercise at the moment, how are we to settle the question about possible shifts in the r* level? Approaching the question directionally, rather than quantitatively, is helpful, but ultimately, we are of the view that it is simply too early to settle the question. As a real rate of interest, r* reflects an economy’s potential growth rate. In turn, this reflects demographic and productivity trends. In order for potential growth to sustainably increase, either one of these two components must sustainably increase as well.
Demographic trends, however, have actually worsened in the aftermath of the pandemic. According to the Conference Board’s mid-year population estimates data, the US population has grown at an average of 0.34% per year over the past three years (2021-23), which is exactly at half the rate that prevailed between 2010 to 2019. As of October 2023, the overall labor force participation rate remained roughly half a percentage point below the rates that prevailed just before COVID-19.
As for labor productivity, the best that can be said at this point is that the gentle downtrend that began in the 2000s appears to have flattened out. It is far too early to make high conviction predictions about potential productivity enhancements associated with either reshoring (via increased capex) or regenerative AI.
Directionally, both should help boost labor productivity, but it is extremely difficult to state over what time horizon this would become visible in macro data and what the magnitude of the impact would be. Moreover, we would need some productivity gains from these sources simply to offset the intensifying demographic drag. In other words, from a potential growth standpoint, we need to “run” on productivity simply to “walk” on growth.
It is because of the considerable inertia in these fundamental drivers of real growth that we believe that not more than an incremental uptick in r* is likely to have occurred.
New York Fed President John Williams (co-creator of the Holston-Laubach-Williams r* estimation model) noted in May 2023 that: “according to the model estimates, the main longer-term consequence from the pandemic period is a reduction in potential output, but the imprint on r-star appears to be relatively modest. Importantly, there is no evidence that the era of very low natural rates of interest has ended.”1
Maurice Obstfeld, previously chief economist at the International Monetary Fund (IMF), also argued that: “equilibrium long-term real interest rates have risen recently, according to market indicators. However, the main underlying factors that have pushed real interest rates down since the 1980s and 1990s – notably demographic shifts, lower productivity growth, corporate market power, and safe asset demand relative to supply – do not appear poised to reverse and drive a durable rise in global real interest rates over the coming years. Low equilibrium interest rates probably will continue to bedevil monetary policy and financial stability.”2
And in its April 2023 World Economic Outlook, the IMF predicted that “once the current inflationary episode has passed, interest rates are likely to revert toward pre-pandemic levels in advanced economies. How close interest rates get to those levels will depend on whether alternative scenarios involving persistently higher government debt and deficit or financial fragmentation materialize.”3
Assessments of r* are implicitly embedded in estimations of the long-run neutral federal funds rate. On the default assumption that the Fed is able to deliver on the 2.0% inflation target, we can infer FOMC members’ own estimates of r* by subtracting 2.0% from their longer-run fed funds rate estimates. The median estimate for the longer-run fed funds rate had drifted gradually lower over time before steadying at 2.5% in recent year. Nevertheless, a gentle updrift in the weighted average estimate is clear in Figure 2.
What is not clear is whether these upgrades to the longer run rate exclusively reflect higher estimates of r* or whether they also reflect some marginal creep up in inflation as well. Chair Powell has stated in no uncertain terms that the inflation target has been and remains at 2.0%. As such, any changes in the longer-run fed funds rate estimate should be viewed as a change in underlying expectations for r*.
Still, given the multitude of shocks battering the global economy since COVID-19 and given major shifts including deglobalization and the energy transition, it would not be surprising if confidence in the committee’s ability to deliver on the 2.0% inflation target has been eroded at the margin.
Our colleagues recently discussed the possibility that the term premium may be sustainably shifting higher. In our view, any such move higher argues for a compensatory decline in r* insofar as the effective real long-term borrowing costs corresponding to any level of the policy interest rate would increase. That higher cost of capital should, all other things being equal, reduce growth and increase unemployment. The combination would by default imply a lower r* since r* reflects full employment conditions.
In conclusion, while several forces could push the neutral rate higher in coming years, we believe shifts so far have been very modest. Chair Powell wisely noted in his speech at the IMF on November 9, 2023 that “it is too soon to say whether the monetary policy challenges of the effective lower bound will ultimately turn out to be a thing of the past.”