SUMMARY

- At its September meeting, the FOMC will likely conclude its strategy framework review and introduce new stimulus to aid the recovery.

- Look for a commitment to keep the policy rate at zero until inflation is sustainably at two percent. The monthly pace of Treasury and agency MBS purchases will likely be upsized, with the program’s eventual end also tied to inflation outcomes.

— The forward guidance on interest rates implies that the FOMC will not just tolerate, but will seek, inflation above two percent in the years ahead.

- The risks around the September meeting are towards more rather than less, stimulus, i.e., even stronger forward guidance on interest rates and a larger-than-anticipated increase in asset purchases.

- Whether these actions will provide meaningful stimulus to the real economy remains to be seen. But the impact on financial markets could be significant on a go-forward basis. Real interest rates could continue to fall, an outcome that would be supportive of risk assets, and equities in particular.

- Even with these steps, the Federal Reserve is not out of bullets. Should the health and economic crisis worsen, there is more the FOMC could do. Yield curve control, corporate quantitative easing (QE) and bank lending schemes would all be on the table.

BACKGROUND

For over twenty years, and prompted in no small measure by Japan’s early experience with deflation and near-zero interest rates, Federal Reserve policy makers have debated how best to achieve its policy goals in a low-rate environment. Even in the late 1990s, policy makers were concerned that in a future recession, a policy rate of zero still might not be low enough to provide sufficient stimulus. This Zero Lower Bound (“ZLB”) quandary became a reality in the Great Recession, and led the Federal Reserve to embark on progressively more aggressive forms of quantitative easing and interest rate commitments (“forward guidance”) aimed at returning the economy to full employment and achieving the two percent inflation objective as quickly as possible.

With the economy performing well and the labor market back at full employment by late 2018, the FOMC began a process of reviewing how best to pursue its employment and price stability objectives in a world where interest rates looked set to stay low for years and years to come. The FOMC saw the issue as follows: the neutral real interest rate – the real policy rate that was neither restrictive nor stimulative ─ had fallen over recent decades due to demographic and productivity trends. This was viewed as problematic because in order to ease monetary policy in a recession, the FOMC takes the real policy rate (the federal funds rate adjusted for inflation) far below the neutral rate. But if the neutral rate is quite low, the room to ease policy is constrained. This can be seen in the charts below. The neutral real interest rate (often referred to as r* or “r-star”) is shown in light blue. Estimates of r* fell from the 2.0 to 3.5 percent range in the 1990s and 2000s to less than one percent in the post-Great Recession period. The real fed funds rate, r, is shown in light blue. The right-hand chart shows the difference between the two, to make the policy stance explicit. Even with the nominal policy rate now essentially at zero, relative to r* the policy stance is not particularly accommodative. For example, in the aftermath of the 2001 recession, the real policy rate was about 300 to 400 basis points below a neutral setting. Today, the real policy rate is only around 150 basis points below r*. In sum, a depressed r* limits the ability of the Federal Reserve to provide stimulus.


The FOMC was focused on the ZLB quandary before the Covid crisis not because they feared an imminent recession, but because they were concerned that the ZLB made it difficult for them to achieve their two percent inflation objective even during good times. Their thinking went as follows:

1. If the public knows that the ZLB will limit the ability to provide stimulus in the next recession, then it will assume that the central bank will struggle in that recession to prevent deflation.

2. If the public expects deflation (or even very weak inflation) in the next recession, that will impact long-term inflation expectations even during an expansion.

3. If long-term inflation expectations are low in an expansion, then actual inflation will be low as well. Achieving the inflation objective becomes challenging, and persistently low inflation undermines the credibility of the commitment to the inflation objective.

The FOMC saw validation of this line of reasoning in survey-based measures of inflation expectations, which were generally on the low side during the prior expansion, and more importantly, in core inflation itself. Over the prior expansion, core PCE inflation averaged only 1.6 percent. From the FOMC’s perspective, this increasingly represented a policy failure that would risk deflation in a future recession.

In light of the ZLB quandary, the FOMC formally launched a review of its monetary policy strategy in the fall of 2018. Even well in advance of the Covid crisis, the direction of travel for the strategy review was quite clear ─the FOMC would tolerate inflation above two percent for the rest of the expansion, on the assumption that inflation would be below two percent during and in the immediate aftermath of the next recession. With higher inflation during good times balancing weak inflation during the bad, the FOMC hoped that inflation would more or less average two percent – its definition of price stability – over the entire business cycle.


THE LIKELY WAY FORWARD

The strategy review was set to wrap up around the middle of this year. But once the Covid crisis began, the FOMC shifted its focus to crisis response measures, and has only recently returned to the review. But there are other reasons for a delayed rollout of a new policy strategy:

- The initial context for the policy strategy review was how to manage policy late in an economic expansion; now the Committee must consider the strategy review in the context of a recession and an economy in need of ongoing monetary stimulus. This can be seen in the chart at right ─ according to a basic Taylor Rule that uses the FOMC’s own economic projections as inputs, the optimal policy setting is a deeply negative policy rate for the foreseeable future.

- The virus has created a tremendous amount of economic uncertainty; waiting a bit longer to finalize the new policy strategy allows the FOMC to calibrate additional stimulus with greater confidence.

- New stimulus can take into account any additional fiscal relief measures that go beyond the immediate crisis response of the CARES Act.

Chair Powell’s comments in his last post-FOMC press briefing imply that the strategy review will be completed at the September meeting, with an announcement at that time of additional measures to support the economy. The Committee has been very tightlipped about its preferences, beyond a few FOMC participants commenting that they support linking an eventual rate increase to achieving the inflation mandate. Based on these comments and the extensive literature on optimal policy at the ZLB, I expect that the measures announced in September will consist of:

- At least an implicit commitment to achieve an overshoot of the two percent inflation objective. More specifically, the FOMC will likely commit to keeping the policy rate at zero until core inflation is sustainably at two percent. How this threshold would be defined is unclear. Perhaps year-over-year core inflation would need to be at or above two percent for three or six months before the FOMC would begin to raise rates. This is an implicit overshoot commitment because policy affects the economy with a lag. By the time the Committee lifts the real policy rate above r*, inflation would presumably have been over two percent for some time.

- An increased pace of Treasury and MBS purchases, perhaps with a pledge to continue with purchases until progress towards the inflation objective is well in train. The Committee will likely look to end asset purchases before it begins raising the policy rate. Thus asset purchases might not be tapered until, for example, core inflation rises to two percent on a year over year basis and Committee participants project it to remain at that level or higher.

— As for the monthly pace of asset purchases, in order to provide additional stimulus, the Committee will most likely increase purchases from their current pace of $80 billion per month in Treasury securities and $40 billion per month of agency MBS. I would expect the pace to be increased to at least $100 and $50 billion, respectively. In addition, the FOMC is likely to shift its purchases toward longer-dated securities in order to lower the term structure of rates. With the policy rate at the ZLB, the FOMC will opt to provide additional stimulus not only by increasing the pace of purchases, but by actively seeking to lower term premia further out the curve.

Note that the risks around the base case outlined above are skewed towards the FOMC doing more, not less. Chair Powell and his colleagues are quite concerned that this crisis could do lasting damage to the economy, in the form of lower trend growth and a higher natural rate of unemployment. That is, mounting business closings could limit employment opportunities in the years ahead, resulting in a higher equilibrium unemployment rate than existed pre-crisis. This would lead to a permanently lower level of national income. What might more aggressive policy look like? It could take the form of an even higher rate of monthly asset purchases, or an explicit pledge to keep the policy rate at zero until inflation is sustainably above two percent.

The FOMC might also decide to tie its interest rate guidance and asset purchases to both inflation and labor market outcomes.For example, the Committee might commit to keeping the policy rate at zero until inflation is sustainably back at two percent and the economy is back at full employment. Judging by public comments and recent FOMC meeting minutes, there is only limited support for linking the interest rate commitment to employment outcomes, but this may change. For one, Committee members might feel that since this crisis manifests itself most painfully as a health and labor market shock, linking policy to labor market outcomes would resonate with the public (the prior crisis also revealed public and political discomfort with policy stimulus framed as seeking higher inflation). But the approach is not without challenges:

- First, how would the Committee define the full employment threshold? Even before this crisis the Committee was highly uncertain about the “equilibrium unemployment rate,” i.e., the level of the unemployment rate that is consistent with full employment. In fact, their estimate of the equilibrium unemployment rate fell from five percent to almost four percent over the five years ending in December 2019. Their uncertainty about full employment conditions must be even higher now.

- Second, the FOMC would have to arrive at a consensus about how to handle situations where inflation and unemployment do not move in complementary directions. For example, what if inflation were to move decisively above two percent, perhaps even to three percent, but the unemployment rate remained stubbornly high or even rose? In such a scenario would the Committee put more weight on high inflation and begin raising rates, or focus on sticky unemployment and keep rates at zero? It could be challenging for the Committee to reach agreement on relative weights for inflation and unemployment and to communicate these tradeoffs to the public, especially since these weights could vary across different economic outcomes.

DOES ANY OF THIS MATTER FOR THE ECONOMY? AND FOR MARKETS?

A common observation of recent years has been that monetary policy has diminished effectiveness. There is certainly a great deal of truth to this observation at present, at least when it comes to interest rate policy and QE. After all, it was the ZLB quandary that initially kicked off the Fed’s policy review, i.e., an admission of less policy effectiveness at the lower bound. In addition, while the Federal Reserve can impact short- and long-term interest rates and overall financial conditions, it can only create a conducive environment for borrowing and spending. Still, policy is far from entirely ineffective. For example if the policy rate was 150 basis points higher (i.e., where it was in January), and the FOMC did not increase its asset purchases at the start of this crisis, interest rates would undoubtedly be higher, the dollar would be stronger, stocks would be lower, and credit spreads would be wider. And it is evident that the Fed’s emergency facilities halted a brewing financial crisis dead in its tracks. Absent these measures, theeconomy would likely be in worse shape, not just today but for the foreseeable future.


Monetary policy is also a powerful complement to fiscal policy in this crisis. Most importantly, QE is siphoning up new Treasury supply, mitigating an increase in rates that would damage prospects for the economy. This effect can be seen in the chart at right. The grey bars represent the budgetary impact of all four Covid fiscal packages enacted to date, per the CBO’s estimates. The offsetting blue bars represent QE purchases to date and my projections going forward. If these projections prove accurate, ongoing QE would also likely offset the Treasury supply impact associated with the fiscal package currently under negotiation. Whether the FOMC describes it as such or not, this is monetary financing of deficits on a scale not seen since World War II. It can be extremely beneficial to the economy in the short run. Whether it has long-term negative implications for fiscal restraint, inflation, and the dollar’s reserve currency status remains to be seen.

While policy effectiveness is constrained at the ZLB, it can still have a significant impact on markets. The FOMC has been telegraphing its upcoming regime change for quite some time now, and this has likely served to lower Treasury market volatility as well as the level of nominal and real interest rates. In addition, it is quite possible that real rates still have room to fall. For example, 5-year TIPS breakevens reflect inflation compensation that is well below two percent, and a far cry from where they theoretically should be if the FOMC is serious about engendering an inflation overshoot. Assuming nominal rates remain around current levels, real rates should continue to decline if and when inflation and inflation expectations rise. Some readers might be of the view that the FOMC has limited ability to impact inflation at the ZLB. But if that is the case, then the policy rate will be stuck at zero for even longer, an outcome that will put additional downward pressure on longer-dated Treasuries and continue to suppress interest rate volatility. This environment could be supportive of risk assets as well.


BEYOND FORWARD GUIDANCE AND "CONVENTIONAL" QE

An important lesson from of the past ten years is that while many investors believe the central bank is out of bullets, the FOMC does not. Or put differently, aware of the limitations of policy at the ZLB, the central bank doubles down and goes to even greater lengths to provide stimulus. These measures may ultimately have limited impact on the economy, but they certainly have apowerful impact on markets. Thus it is worth asking, if all of the above measures have limited effectiveness, and/or the healthand economic crisis worsens, what else could the Federal Reserve do? I think there are at least three options that would be on the table almost immediately for providing additional accommodation:

Yield Curve Control. Recent FOMC minutes reveal only limited support for the idea of yield curve control, under which the FOMC would specify caps on Treasury yields, and “enforce” these caps by buying Treasuries whenever upward pressure on yields intensifies. FOMC participants see a number of issues with this strategy, but perhaps the most challenging is the potential for monetary policy to become subordinated to fiscal policy and debt management (“risks to central bank independence”, in central bank parlance). Specifically, if the Federal Reserve were to enforce its yield caps, Congress might see it as an opportunity to engage in unrestrained monetary-financed fiscal expansion. Or at the very least, the Federal Reserve might face political pressure to maintain yield caps for much longer than necessary. This is precisely what happened in the aftermath of World War II, when it was not until 1951 that the Federal Reserve was able to extricate itself from yield caps. Still, the FOMC might actively consider yield curve control strategies more seriously if the crisis were to worsen, as it could be seen as a more efficient tool in the shortterm to keep rates low while Congress responds with even larger fiscal stimulus.

Credit Easing. The FOMC has already begun purchasing corporate debt in this crisis, but these purchases are aimed at improving market functioning rather than directly supporting the economy in the same manner as traditional QE. As a result, actual purchases of corporate debt have been quite modest, and the pace of purchases will decline further if market conditions continue to improve. However, this program could be transformed into corporate QE program with a targeted pace of monthly purchases that continues until economic objectives are met. The recent decision to base purchases on an index of securities leaves the Federal Reserve well-positioned to pivot towards higher-volume purchases should the FOMC view this as a tactic for providing additional stimulus.

Incentivizing Lending.The FOMC is hoping that its Main Street Lending Facility will incentivize banks to extend credit to small and medium-sized enterprises, but limited uptake of this program so far suggests that the terms may be overly restrictive. As a next step, the FOMC could work with Treasury─which provides risk capital to the facility─to further loosen program terms. But to move in this direction, Treasury would need to be comfortable with an increased likelihood of losses on the loan shares that the Fed purchases from banks. A second alternative would be a Funding for Lending (FLS) scheme similar to those used by the Bank of England and the Reserve Bank of Australia. In contrast to the Main Street facility, these schemes incentivize bank lending by providing central bank funding that is tied to bank lending─the more banks lend to targeted sectors, the more funding they can receive from the central bank at favorable rates. An FLS scheme could get around the main issue with Main Street, namely, that banks have little incentive to sell a large portion of attractive loans to the Fed in exchange for favorable funding and origination fees.