Regional President Speech

The Long Return to Normal Remarks by Esther L. George President and Chief Executive Officer Federal Reserve Bank of Kansas City September 24, 2021 American Enterprise Institute Remarks delivered virtually The views expressed by the author are her own and do not necessarily reflect those of the Federal Reserve System, its governors, officers or representatives. Thank you, Steve, for the invitation to participate in this forum. I look forward to this morning’s conversation with your panelists. When the pandemic took hold and widespread shutdowns were adopted in the spring of 2020, it was immediately apparent that the hit to economic activity would be enormous. Less certain was how long the shock would persist. Many workplaces sent staff home with the assumption that things would return to normal within a few weeks. Now a year-and-a-half later, normal remains on the horizon. With the 2020 recession being both the deepest and shortest on record, the economy has bounced back sharply, while the return to normal continues to be the key dynamic shaping the economic outlook. In discussing the shocks to which the economy is now adjusting, I would include both the closures and disruptions associated with the pandemic as well as the enormous fiscal and monetary policy response that followed. The pandemic shock disrupted the pattern of consumption by favoring goods over services. It also disrupted production, introducing frictions in labor markets and scrambling global supply chains. The policy response led to a spike in household savings—as fiscal deficits funded transfer payments—and to a doubling of the Fed’s balance sheet. In many cases, these shocks—the pandemic and the policy reaction—were offsetting as evidenced by the rapid recovery of output and employment. However, in other cases, the pandemic and the policy response conspired to push the economy away from its equilibrium, disrupting long-run trends. For example, the ratio of goods consumption to services consumption is now far above its long-run historical trend. Similarly, the household saving rate jumped to an all-time high early in the pandemic, and, although it has come down, remains considerably above its average in recent decades. In both these examples, as well as others, it is reasonable to expect that historical patterns will reemerge, with consequences for demand, employment, and prices. As both the pandemic and the policy response fade, the pull toward normal will shape the outlook. That said, while I am comfortable suggesting that the economy has not returned to normal, I would not be comfortable suggesting that I know what normal will end up looking like. “Normal” is likely to be elusive for some time. The pandemic was a tremendous disruption that has affected every aspect of how and where people work, where they live, and what activities they engage in. Some of the changes coming out of the pandemic will persist and alter the 1 structure of the economy. Of course, separating transitory from persistent is one of the central challenges of economic policymaking. It might seem odd to be discussing a return to normal when the Delta variant has pushed new infections uncomfortably close to all-time highs. However, my sense is that, notwithstanding the tragic human toll, the primary effect of the Delta variant will be to prolong the return to normal, rather than upend the process altogether. In the remainder of my time, I would like to discuss some of the key normalization dynamics I will be watching in the economy. Overall, I view these dynamics as supporting a continued strong recovery and further gains in employment, while simultaneously relieving some (though not all) of the price pressures that have elevated inflation in recent months. I will then turn to the outlook for monetary policy, including some thoughts on what policy normalization might look like, and how the decisions of the Federal Open Market Committee (FOMC) might shape that process. The Economic Outlook One dynamic likely to shape the outlook is the reemergence of long-standing patterns of consumption, particularly spending on goods relative to services. With the onset of the pandemic, restrictions and consumer caution led to sharp declines in the consumption of contact- intensive services, such as hotel accommodations, concerts, and restaurants, even as homebound consumers ramped up spending on goods. Though services remain a far larger component of consumption than goods, their relative share has declined. For example, in the second quarter, consumers purchased $1.80 of services for every $1 of goods, far less than the $2.20 they spent on average in the years leading up to the pandemic. Over time as the pandemic’s grip wanes, I expect old patterns to reemerge as audiences return to theaters, conferencegoers resume travel, and the Jackson Hole Symposium returns to Jackson Hole, for example. This return to normal will pivot consumption back towards services and away from goods. The rotation towards services consumption will be important for further improvements in the labor market. Service sector jobs, primarily in leisure and hospitality and health, represent the overwhelming majority of the over 5 million jobs that have yet to return following the dramatic loss of jobs in March and April of last year. I expect a return of services consumption to lead to further strong job gains, with the caveat that the Delta variant could delay this rebound. 2 A pivot towards services will also take some pressure off the red-hot goods market. Strong demand for durable goods, exacerbated by supply chain difficulties, has been an important contributor to the overall rise in inflation. Over the past year, durable goods prices have increased 7 percent, the first sustained break in an over 20-year trend of declining prices. Automobiles, both new and used, have been an important part of this increase, but the prices for other goods, such as household appliances and recreational equipment have also moved up sharply. As consumption rotates from goods to services, some of the pressure should ease for overstretched goods demand, even as services growth continues to support the overall recovery in the economy. Although I expect an easing in the demand for goods, and a loosening of bottlenecks, to relieve some of the pressure on prices, the data offer a note of caution. In particular, recent months have seen a rebound in previously depressed services prices even as goods prices remain elevated. For example, prices for hotel accommodations in the index of personal consumption expenditures (PCE) are now 9 percent above pre-pandemic levels, even as spending on hotel rooms remains 16 percent below where it was prior to the pandemic. Earlier this year, PCE inflation was recording record levels of dispersion across categories of consumption. For example, in February, categories representing 16 percent of PCE expenditures were reporting inflation far in excess of average, mostly goods, whereas 26 percent of expenditures recorded extraordinarily low inflation, mostly services.1 By June, the percent of expenditures with well- below average inflation had fallen to 5 percent, while 51 percent of expenditures recorded price increases far above recent trend. This is to say, that a shift in demand towards services could help ease inflation pressures, but that a sustained step down in inflation is likely to also require a stabilization in services prices. Why are prices for some services increasing rapidly even as demand remains weak? Speaking to contacts in my district, a common explanation is the lack of available labor. Without workers, supply is curtailed and prices rise. The labor shortage has developed as over 5 million fewer people are employed relative to before the pandemic. In large part, the current tightness of labor markets reflects a dramatic decline in labor force participation that developed at the start of the pandemic and has shown little improvement since. Despite the persistence of the decline in labor force participation, I believe the pandemic shock remains behind a large part of the decline, 1 Federal Reserve Bank of San Francisco | PCE Inflation Dispersion (frbsf.org) 3 with a high likelihood of workers returning as pandemic-induced frictions, including concern over the safety of returning to work as well as disruptions to childcare arrangements, continue to ease. The pandemic shock has led many women to leave the labor force, disproportionately women of color without college degrees.2 The absence of these workers from the labor force is contributing to labor shortages in some services industries. Difficulty in obtaining child-care likely explains some of the decline in participation. Tellingly, employment in daycare centers remains 10 percent below pre-pandemic levels, suggesting that capacity in this important industry remains impaired. As the pandemic fades, a normalization of childcare could bring these workers back into the workforce, supporting growth, increasing employment, and alleviating some of the labor shortages that could be contributing to price increases. Another dynamic contributing to the fall in labor force participation has been an increase in the proportion of the population reporting being retired. Interestingly, research by staff at the Kansas City Fed has shown that the rise in reported retirement does not reflect an increase in the transition of workers into retirement, but rather a sharp decrease in the typical flow of retirees back into the workforce.3 Before the pandemic, every month a substantial number of retirees reported returning to work. It is possible increased health concerns during the pandemic disrupted this flow. Here again, as the pandemic fades, it might be reasonable for this flow to once again reemerge, adding workers to the labor force and further alleviating worker shortages. All in all, my expectation is that a return to historical patterns of consumption, with a renewed emphasis on services, and a recovery in labor force participation, with the return of childcare and decreased health concerns, will contribute to an outlook for continued growth, less-binding constraints, and moderating inflation. There are risks around this outlook, including the continued presence of the pandemic. Importantly, the effects of the upsurge in new cases could be as pronounced for supply as for demand, prolonging the tightness of the economy and maintaining upward pressure on prices. Renewed concern over the virus could impede the recovery in services consumption such that demand remains directed towards sectors of the economy that are near capacity and away from 2 Women without a College Degree, Especially Minority Mothers, Face a Steeper Road to Recovery (kansascityfed.org) 3 What Has Driven the Recent Increase in Retirements? - Federal Reserve Bank of Kansas City (kansascityfed.org) 4 those sectors that have available slack. The surge in the virus could also delay the normalization of the labor market, particularly if schooling and childcare are once again disrupted. Aside from the economic threats associated with the pandemic, the high level of household savings contributes an additional element of uncertainty to the outlook. Estimates suggest that the stockpile of household savings has increased by close to $2.5 trillion. Of course, healthy household balance sheets are positive for the economy, and, if the savings are spent out over time, this spending could support steady economic growth for some time. However, if households instead choose to spend rapidly, a rush of demand could keep the economy at capacity, reinforcing bottlenecks and putting continued upward pressure on prices. Households have a lot of firepower, and if and how quickly they choose to spend will be an important factor in how tight the economy remains. Of course, whether consumers choose to spend or save is a decision directly impacted by monetary policy, so the speed at which any excess household saving is spent down is not independent of our policy decisions. The Outlook for Monetary Policy Since last December, the Committee has stated that it expects to keep the policy rate near zero until the labor market has reached levels consistent with maximum employment and inflation has risen to 2 percent and is on track to moderately exceed 2 percent for some time. The FOMC also expects to maintain its purchases of Treasuries and mortgage-backed securities until substantial further progress has been made towards these employment and inflation goals. In my view, the criteria for substantial further progress have been met, with inflation running well above our target and the unemployment rate at 5.2 percent, down 1½ percentage points relative to December. Under these conditions, the rationale for continuing to add to our asset holdings each month has waned, and signaling that we will soon consider bringing our asset purchases to an end is appropriate. While recent public focus has been on the timeline for tapering asset purchases, the cumulative effect of these purchases is arguably the more substantive force acting on the economy. Since March of last year, the Federal Reserve has purchased more than $4 trillion of securities, pushing our total asset holdings to nearly $8.5 trillion dollars. These asset holdings are depressing longer-term interest rates most relevant for households and businesses and thereby are 5 providing a significant amount of accommodation. And, importantly, this accommodation will persist even when tapering is complete. As the adjustment of our asset purchases gets underway, the focus of attention will naturally shift to the path of the policy rate. The ongoing effects of these asset purchases, alongside the presence of both upside and downside risks, will complicate the task of judging the achievement of criteria for raising rates. One might argue that today’s inflation dynamics are likely to keep inflation moderately above 2 percent for some time and align with the Committee’s threshold criteria. On the other hand, the criteria for judging maximum employment are murkier. While it is clear that we remain far from the historic low levels of unemployment achieved pre-pandemic, it is less clear to me that such a benchmark will be the best guidepost in the current expansion. As I discussed, the pandemic introduced a number of frictions into the labor market. Barring further intensification of the virus, I would expect these frictions to fade, promoting strong job gains and a relatively fast approach to maximum employment. However, I am open to the possibility that the pandemic has resulted in a number of structural changes in the labor market. These changes could affect the assessment of maximum employment in ways that are not yet clear. Policy Normalization Beyond interpreting the evolving economic landscape, policymakers also must consider the implications of the multiple policy tools at play as we contemplate removing policy accommodation. While the shift from extraordinary policy accommodation to policy normalization remains some way off, the linkages between balance sheet policies and the path for policy rates will be factors in future policy deliberations. All else equal, maintaining a larger “normal” balance sheet should imply a higher “normal” terminal policy rate, as higher policy rates are needed to offset the stimulative effect of the balance sheet’s continued downward pressure on longer-term interest rates. Given the likely trade-offs between the policy rate and the size and composition of the balance sheet as policy shifts towards equilibrium, several factors could affect policymakers’ decisions. For example, one question that might be considered is where along the yield curve would we prefer the most policy space: at the long-end or at the short-end? If the zero lower 6 bound is thought to be a costly constraint on policy, there might be an advantage in pushing towards a higher neutral policy rate, arguing for maintaining a relatively large balance sheet weighted towards longer-maturity assets. The accommodation provided by our balance sheet in this case will have to be offset by higher policy rates, which in turn would provide us with more space to cut rates in a downturn. We might also want to consider the rationale for a smaller balance sheet, or at least shifting toward one with shorter-maturity assets, with a lower neutral policy rate. One rationale might be a desire to decrease our footprint in financial markets. Another is that we might be concerned about the slope of the yield curve. Raising rates while maintaining a large, long- maturity balance sheet is a recipe for inverting the yield curve, with potential negative effects for traditional banking models. Commentators often discount the effect of an inverted yield curve on banks, arguing that net interest margins are only weakly correlated with bank profitability. While this might be true for the largest banks, it is not true for small banks. An inverted yield curve is a particular headwind for community banks, promoting further consolidation in the sector and undermining the critical role that community banks provide in terms of local access to credit and economic development. As the economy recovers from this pandemic shock, its path is likely to confound our assumptions about what a return to normal might look like. The same is true for the monetary policy normalization process. Both point to a long and difficult process ahead as the economy heals and the stance of monetary policy responds. 7

Esther L. George (2021, September 23). Regional President Speech. Speeches, Federal Reserve. https://whenthefedspeaks.com/doc/regional_speeche_20210924_esther_l_george

@misc{wtfs_regional_speeche_20210924_esther_l_george,
  author = {Esther L. George},
  title = {Regional President Speech},
  year = {2021},
  month = {Sep},
  howpublished = {Speeches, Federal Reserve},
  url = {https://whenthefedspeaks.com/doc/regional_speeche_20210924_esther_l_george},
  note = {Retrieved via When the Fed Speaks corpus}
}