Policy Has Tightened a Lot. How Tight Is It? (An Update)

Over the past few years, I have published a series of essays assessing where we are in our inflation fight and highlighting some important questions policymakers are facing. My most recent essay was in February of this year, where I questioned how much monetary policy was actually restraining demand. This essay is an update to that commentary, and I now examine the current stance of monetary policy in more detail.¹

I will argue that the Federal Open Market Committee (FOMC) has tightened policy significantly, compared with prior cycles, both in absolute terms and relative to the market’s understanding of neutral. But I will also observe that the housing market is proving more resilient to that tight policy than it generally has in the past. Given that housing is a key channel through which monetary policy affects the economy, its resilience raises questions about whether policymakers and the market are misperceiving neutral, at least in the near term. It is possible that once the reopening dynamics of the post-COVID economy have concluded, the macro forces that drove the low-rate environment that existed before the pandemic will reemerge, pulling neutral back down. But the FOMC must set policy based on where neutral is in the short run to achieve our dual mandate goals in a reasonable period of time. The uncertainty about where neutral is today creates a challenge for policymakers.

Economic Update

Since my last update in February, two significant economic developments have occurred simultaneously: Inflation appears to have stopped falling and economic activity has proven resilient, continuing the robust activity we saw in the latter half of 2023.

The FOMC targets 12-month headline inflation of 2 percent. While we saw rapid disinflation in the second half of 2023, that progress appears to have stalled in the most recent quarter (see Figure 1). The question we now face is whether the disinflationary process is in fact still underway, merely taking longer than expected, or if inflation is instead settling to around a 3 percent level, suggesting that the FOMC has more work to do to achieve our dual mandate goals.

During this time, economic activity has continued to show remarkable strength, as shown in Figure 2. While the most recent headline GDP appears somewhat weaker than prior quarters, that slowdown was driven largely by inventories and net exports. Underlying domestic demand remained strong.

The labor market, the other half of our dual mandate, has also remained strong with the unemployment rate at a historically low 3.9 percent.

Policy Is Much Tighter than the Pre-pandemic Period

In prior essays I wrote that the single best proxy for the overall stance of monetary policy is the long-term real rate, specifically the 10-year Treasury inflation-protected securities (TIPS) yield. Focusing on a long-term rate incorporates the expected path of both the federal funds rate and balance sheet, not just the current level of the federal funds rate. Moreover, it adjusts the expected path of policy by expected future inflation—the relevant comparison—rather than by recently realized inflation.

As I noted in earlier essays, prior to the pandemic the 10-year real yield was about zero, which I estimate was roughly a neutral policy stance at that time. In response to the pandemic, the FOMC acted aggressively to support the economy by driving the federal funds rate to the effective lower bound and massively expanding our balance sheet. Those combined effects drove the 10-year real yield to roughly -1 percent, as shown in Figure 3. Since we began our tightening cycle two years ago, 10-year real yields have more than fully retraced their pandemic decline and are now around 2.2 percent. Thus, we are clearly in a tighter stance now than immediately before the pandemic.

This Tightening Cycle Appears to Be as Aggressive as the 1994 Cycle

Data from the TIPS market only go back 20 years or so; thus, to evaluate earlier tightening cycles one must make a number of assumptions about the neutral rate and about inflation expectations.

I do not think comparisons to the 1970s and early 1980s are particularly relevant to us today because in those decades, the FOMC had to establish its inflation-fighting credibility, so the required monetary tightening was very large.

The tightening cycle in 1994 might be a better benchmark because at that time, as is true now, the FOMC had a lot of credibility with the public. Inflation was not as high in 1994, however, as it was in this episode. So it is not a perfect comparison either.

Minneapolis Fed staff’s best estimate is that when the FOMC raised the policy rate by 300 basis points in the 1994 tightening cycle, this translated into an increase of about 200 basis points in the 10-year real rate (Figure 4), which was coincidentally also about 200 basis points above the then-neutral 10-year real rate. So that is the key: It seems as though policy drove the 10-year real rate about 200 basis points above neutral.

How does that compare to our current tightening cycle? (See Figure 3.) If my estimate of neutral being zero before the pandemic still holds, then we have accomplished similar or a bit more tightening in this cycle than was achieved in the 1994 tightening cycle.

Yield Curve Suggests Policy Is Tight

As I stated earlier, the underlying inflationary dynamics are quite different today than in 1994, so simply repeating the 1994 tightening might not be enough. And perhaps the unique dynamics of the post-COVID reopening economy have caused neutral to increase.

Another indicator I look at to assess the stance of monetary policy is the shape of the yield curve. Specifically, if the yield curve is inverted, it might indicate that monetary policy is in a contractionary stance. A lot of public attention is given to the yield curve, and there is a robust debate about whether an inverted yield curve is a reliable recession indicator. I wrote about this in 2018. Setting aside its usefulness as a predictor of recessions, the yield curve does seem to give some indication of the stance of monetary policy. The long end of the yield curve should offer some signal of where market participants believe interest rates will settle once current economic and policy shocks have run their course; if markets understand that neutral has moved, it should be reflected in the long end of the curve. If current short rates are higher than long rates, then that might signal an overall tight stance of policy today. Figure 5 shows the history of the (nominal) yield curve with a number of inversions over the past 50 years, including the current inversion.²

Depending on the specific measures chosen, the yield curve has now been inverted for more than 20 months, which is a relatively long and somewhat deeper inversion than most prior cycles, the Volcker disinflation period being the exception. The current inverted yield curve suggests that policy is in fact tight relative to the market’s understanding of neutral.

The Resilience in the Housing Market Nonetheless Raises Questions

Housing is traditionally the most interest-rate-sensitive sector of the economy. Prior yield curve inversions also coincided with a marked slowdown or even contraction in residential investment, as shown in Figure 6. Curiously, while residential investment fell in the first part of our tightening cycle, it has since reversed and has grown 5 percent over the past year.

What could explain this apparent resilience in residential real estate given monetary policy that has led to an inverted yield curve? We know that following the Global Financial Crisis, the country built far fewer housing units than were needed to keep up with population growth and household formation. Thus, there appears to be a significant shortage of housing that will take a long time to close. In addition, responses to COVID have led to an increase in people working from home, and that has led to increased demand for housing. In recent years there has also been a significant increase in immigration. While the long-run effect of increased immigration on inflation is unclear, immigrants nonetheless need a place to live, and their arrival in the U.S. has likely also increased demand for housing. Policy actions by the FOMC have driven 30-year mortgage rates from around 4.0 percent prior to the pandemic to around 7.5 percent today. Perhaps that level of mortgage rates is not as contractionary for residential investment as it would have been absent these unique factors which are driving housing demand higher. In other words, perhaps a neutral rate for the housing market is higher than before the pandemic.

Other Signals from the Real Economy Are Mixed

Monetary policy is a blunt instrument that eventually affects virtually the entire economy, not only housing. While high interest rates may not be slowing housing as much as in prior tightening cycles, it is nonetheless having an impact on other sectors of the economy. For example, auto loan and credit card delinquencies have increased from very low levels and are now at rates higher than existed before the pandemic, indicating that some consumers are feeling stress from increased borrowing costs. Overall, however, economic activity, consumer spending and the labor market have proven surprisingly resilient.

Conclusion

The FOMC has undeniably tightened policy meaningfully, both relative to the pre-pandemic period and to some prior tightening cycles. Nonetheless, it is hard for me to explain the robust economic activity that has persisted during this cycle. My colleagues and I are of course very happy that the labor market has proven resilient, but, with inflation in the most recent quarter moving sideways, it raises questions about how restrictive policy really is. If policymakers and market participants are misperceiving the neutral policy rate, that could explain the constellation of data we are observing. This is also a communication challenge for policymakers. In my own Summary of Economic Projections (SEP) submission, I have only modestly increased my longer-run nominal neutral funds rate level from 2 percent to 2.5 percent. The SEP does not provide a simple way to communicate the possibility that the neutral rate might be at least temporarily elevated.

Endnotes

1 These comments reflect my own views and may not necessarily represent the views of others in the Federal Reserve System or of the Federal Open Market Committee.

2 Conceptually, the real yield curve is a better measure of the stance of monetary policy than the nominal yield curve. That said, because nominal and real spreads have been quite similar over the past 18 months, the nominal spread in Figure 5 currently provides an equivalent measure of the stance of monetary policy. I plot nominal yields in Figure 5 simply because they are available for a longer time period than TIPS yields.