Economic Market Update: September 8, 2020
Tuesday, September 8, 2020
As summer comes to a close and autumn quickly approaches, we would like to provide an update on the current state of the economy, recent changes in the makeup of the Dow Jones Industrial Average, and recent changes in the Federal Reserve’s framework for monetary policy.
US Economic Outlook
After experiencing the sharpest economic contraction in recorded history during the second quarter of 2020, the US economy had initially appeared to rebound strongly. Activity resumed as lockdowns were lifted, spurring hopes of a V-shaped recovery. Despite a historic increase in unemployment, as tens of millions of Americans found themselves out of work, an equally historic fiscal policy response mitigated the worst of the damage and provided a bridge for businesses and consumers to traverse the economic gap created by efforts to contain the COVID-19 outbreak, setting the initial stage for a strong recovery.
During the depths of the crisis in April, net government transfers surged an unprecedented 182% as desperately needed fiscal stimulus kicked in. Support payments from government sources, mainly in the form of unemployment benefits, represented 24.9% of personal income at their peak in April, and have remained high, with July’s 17.4% figure higher than any month since 1959. To put this in proper perspective, in normal times, unemployment benefits account for around 1% of total income. During the Great Financial Crisis, they peaked at 2.5%. In total, the amount of fiscal stimulus that has been injected into the system is around 13.2% of 2019 GDP.
The Citi US Economic Surprise Index, which measures economic data surprises relative to market expectations, has soared in recent months to the highest levels recorded in its history, which dates back to 2003.
The positive news may be beginning to wane, however, putting the hopes of a strong V-shaped recovery at serious risk. The evidence indicates that the strong economic snapback experienced during the summer months was in most part due to the unprecedented amount of fiscal stimulus flowing through the system; but, on July 31, the fiscal stimulus spigot was largely turned off.
Once the calendar turned to August, the $600 enhancement to unemployment benefits terminated, followed shortly by the expiration of mortgage forbearance and eviction moratorium programs, the end of a freeze on federally subsidized student loan payments, and the conclusion of the PPP program. President Trump has issued executive orders in an attempt to cover these expiring programs, but these only offer a temporary fix, and legislation is still required. Initial optimism that policy makers would quickly reach an agreement on another round of fiscal stimulus has faded into skepticism, and more than a month later, the two sides continue to appear far apart. This failure on the part of policy makers to extend help to millions of Americans has significant ramifications for the course of the economic recovery in the months ahead.
High-frequency data indicates that the nascent bounce in spending seen over the past few months has stalled as these programs expired in August. A Wall Street Journal report from late August highlighted a significant decline in spending at grocery stores, based on weekly data from the month. Food prices and dining at restaurants were stable during the same period, indicating that they were not factors in the decline – evidence that the reduction in benefit payments was likely a key factor. This is further supported by the fact that states experiencing higher levels of unemployment also saw more significant declines. The authors concluded that, if the reduction in government fiscal support is already leading households to cut back spending on necessities, the outlook for discretionary spending in the months ahead could be grim if the current stalemate in Washington continues. Bloomberg Economics estimates that failure to pass an additional stimulus package will push their baseline estimate for 2020 GDP growth from -6.5% to -10%. The damage would bleed into 2021, as the expected path of the recovery would be shallower as well, reducing overall economic growth in 2021 by possibly 1% or more.
Three factors have combined during this crisis to increase the role unemployment benefits are currently playing in the US economy by a significant margin, relative to historical crises:
- First, an unprecedented number of Americans lost their jobs all at once in March and April.
- Second, Congress created the Pandemic Unemployment Assistance (PUA) program as part of the CARES Act passed in March, greatly expanding the number of unemployed workers eligible for assistance.
- Lastly, the Federal Pandemic Unemployment Compensation (FPUC) program, also part of the CARES Act, provided for an additional $600/week supplement to standard state unemployment benefit payments.
With unemployment insurance payments accounting for around 15% of total income nationally during the pandemic, a failure in Washington to reach an agreement to extend these programs – even at reduced amounts – has the potential to be disastrous.
Recent data releases highlight the importance of another round of stimulus for American families facing the potential of significant financial hardship, and for the economy as a whole. At the peak of the crisis this year, one in five US workers received unemployment insurance benefits – which is five times greater than the previous record. As of the most recent jobless claims report, slightly more than 13.1 million people were receiving payments under traditional unemployment insurance programs on a non-seasonally adjusted basis. To put this in historical perspective, the Great Financial Crisis period provides an apt comparison: at the depths of the crisis, during 2008-2009, continuing claims for unemployment insurance peaked at 6.6 million.
Additionally, due to the emergency programs created under the CARES Act, claims for traditional unemployment benefits only provide a partial picture of the turmoil that still embroils the labor market. Nearly 13.6 million people are currently receiving aid under the PUA program, which covers business owners and other self-employed individuals, independent contractors, and gig economy workers who are ineligible for traditional unemployment insurance schemes. Add in those filing initial claims during the week of the most recent report, and in total more than 29.2 million Americans are currently relying on some form of unemployment assistance as of the most recently available data. In historical context, that means 22.5 million more Americans are out of work now than were unemployed during the depths of the last recession. Another way to visualize this without distortions is through the employment-to-population ratio, which measures the percentage of the overall population that is currently employed. That ratio fell to the lowest level ever recorded earlier this year, and remains depressed, at levels not seen since the 1970s.
Uncertainty over future finances tends to affect how we spend today, and the double-digit increase in the savings rate indicates that consumers, even those who remain employed, are being cautious and have pulled back on spending as well. Academic research has consistently found that spending by the unemployed drastically declines when unemployment insurance benefits expire. To put the impact in perspective, the $600 weekly top-up to traditional unemployment insurance payments amounts to roughly $65 billion in monthly income, and nearly all of it is spent shortly after it is received, according to studies of consumer spending behavior during the pandemic. Harkening back to the axiom we’ve presented in previous updates─that one person’s spending is another person’s income─the removal of $65 billion a month in spending will have far-reaching effects; when annualized, the lost income from the expiration of the program approaches $1 trillion.
A recent working paper from economic and public policy researchers at the Becker Friedman Institute at the University of Chicago examined the impact of unemployment insurance benefits on consumer spending behavior during the pandemic. Historically, spending by those receiving unemployment benefits fell by 7% relative to their baseline while employed. During this crisis, however, the authors found “dramatically different spending patterns for the unemployed compared to normal times.” During the initial months of the crisis, spending fell across the board, with aggregate spending by even fully employed households falling by around 10%. Once unemployed households began receiving benefits, however, their spending rebounded quickly, ultimately increasing to rise above their pre-unemployment baseline. This finding correlates with other research estimating that around two-thirds of unemployed households saw their lost earnings replaced at a level greater than 100% by the CARES Act passed in March.
The authors’ findings on the impact of delays between the onset of unemployment and the initial receipt of unemployment benefits is perhaps even more instructive for the current moment. During normal times, unemployment recipients face little delay between a job loss and the start of benefits. During this crisis, however, the sheer volume of applicants overwhelmed the infrastructure states use to process and approve claims, resulting in some recipients experiencing a gap of two months or more between the onset of unemployment and the initiation of unemployment benefit payments. The authors found that, for these households, spending fell immediately and precipitously – down 20% from the baseline while employed. Given that more than 29 million Americans are currently relying on these benefits, these findings suggest that we likely face a significant decline in personal consumption, which accounts for nearly 70% of GDP, in the months ahead.
Another recent working paper, also from the Becker Friedman Institute at the University of Chicago, explored the dynamics of the labor market during the COVID crisis by examining the distribution of job losses and recoveries by wage tier. The authors found that aggregate employment fell by 21% between March and April, and has since begun to rebound, albeit slowly. In analyzing the drivers of the nascent recovery in the labor market, the researchers found that the most significant contributor to the increase in employment was worker recall as businesses that were closed temporarily or operating at limited capacity due to COVID restrictions, reopened. Half of continuing businesses shrank payroll at some point during the period between February and June. This meshes with the fact that the majority of those reporting job losses during the crisis indicated that the job loss was only temporary. More importantly, however, the authors found that the distribution of job losses varied significantly by wage tier, with the most significant impact falling on those earning the lowest wages─the cohort most impacted by even a temporary reduction in income. Using payroll data from ADP – the largest provider of payroll services in the United States – the researchers divided the workers (about 26 million in total) into quintiles by wage rate. By mid-April, 35% of workers in the lowest quintile lost their jobs, compared to just 9% of those in the highest quintile. Through June, the employment rate remained 20% below the level in February for the lowest wage cohort.
Taken in total, the research on employment dynamics, the economic impact of unemployment benefits, and consumer spending behavior during the pandemic, indicates that an unprecedented percentage of the American labor force remains reliant on unemployment benefits─and the households receiving those benefits are disproportionately those least able to weather a sustained period of financial hardship. Households experiencing a job loss spend unemployment benefit payments swiftly, and almost in their entirety. Finally, household spending declines precipitously for households not receiving benefits, and this decline continues to worsen as time increases. Policy makers have currently retreated into their partisan corners and remain far apart on an additional stimulus package. The longer the delay, the more significant the damage will be to the economic recovery, with the pain reverberating through 2021.
Stalemate Persists In Washington
Congressional lawmakers entered the summer recess without reaching a deal on another round of stimulus. In response, on August 8, President Trump issued a number of executive orders aimed at addressing the expiration of several relief programs authorized in prior legislation. These executive orders addressed:
- Supplemental unemployment benefits.
- Student loan payment relief.
- Payroll tax deferral.
- Moratoriums on evictions and foreclosures.
In practice, these executive orders may provide some needed relief, but they face several practical and legal challenges, and Congressional action is still needed to address these issues in the long-term. One of the most significant executive orders issued by President Trump on August 8 allows states to utilize disaster relief funds to provide an additional $300/week on top of the standard state unemployment benefit. While FEMA has approved 45 states for the program thus far, only six have actually begun to distribute the payments. The executive order is only a stopgap measure, however. Once the benefits begin flowing, which the Department of Labor estimates will take about three weeks, the funds will be exhausted within four to five weeks; Congress will have to act if this much-needed income support for the 29.2 million Americans relying on these crucial benefits is to be continued.
Unfortunately, the two sides appear to remain far apart. While recent reports indicate that an informal agreement has been reached on a continuing resolution to avert a government shutdown on October 1, stimulus talks between the two sides have remained at an impasse since negotiations stalled on August 7. Both House Speaker Nancy Pelosi and Senate Majority Leader Mitch McConnell have expressed doubt in recent days that lawmakers will be able to come to an agreement on additional aid in a timely fashion, once lawmakers return to Washington after a monthlong recess. While we remain optimistic that a deal will ultimately be reached, there is heightened risk that the stalemate will continue through Election Day. While Democrats have reduced their initial ask─the $3.5 trillion package proposed in the HEROES Act─by about $1 trillion, Democratic leaders have stated that they would only return to the negotiating table if Republicans were willing to spend significantly more than $1 trillion. In response, reports are indicating that Senate Republicans are contemplating bringing a slimmed-down $500 billion version of their initial $1 trillion proposal to the floor. The two sides clearly remain far apart. As we approach Election Day, and partisan rhetoric heats up, the probability of a deal may decline further.
A Bridge Too Short
The Congressional Budget Office projects that the cumulative gap between actual economic output and potential GDP caused by the crisis is around 12.5%. Using 2019 GDP figures, that suggests a gap of around $2.7 trillion. Moody’s Analytics has been closely tracking fiscal stimulus, and reports that around $2.3 trillion of the nearly $3.5 trillion in approved spending has been disbursed as of September 1. On the surface, it may appear that the approved stimulus spending already authorized should be more than sufficient to offset the economic output gap caused by the virus. If economic activity had returned to pre-pandemic levels, this might very well be true. Moody’s estimates that, after declining 40% during the worst of the crisis, economic activity experienced a strong rebound in the early summer months, as fiscal stimulus programs kicked in, and checks hit the bank accounts of the tens of millions of American families most impacted by the virus. In the weeks since many of these programs expired or were disrupted, however, economic activity has plateaued, and it remains about 21% below normal.
Until economic activity is back to normal, the removal of these stimulus programs from the system will stunt the recovery in 2020 – and will ultimately lower growth in future years as well. Fiscal stimulus acts as a bridge to help the economy make it across the output gap caused by efforts to contain the virus (in economic parlance, stimulus stabilizes aggregate demand when there is a shortfall). We have now arrived at the end of the bridge built in March and April, and it has become clear that the original bridge built by policy makers through legislation like the CARES Act is not long enough to span the economic chasm created by the virus. If policy makers fail to lengthen the bridge to span the entire chasm, the recovery is likely to stall, making the journey to reach the other side of the economic canyon we still face more costly and difficult.
Changes to the DJIA
A recent shake-up that has been in the news involves the changes to the composition of the Dow Jones Industrial Average that occurred on August 31. The most striking change is the removal of Exxon, the longest-tenured company in the index at the time of its removal. The company was first added to the blue-chip index in 1928, when it was then known as Standard Oil of New Jersey. Although the removal is largely symbolic, it caps off a yearslong decline in Exxon’s fortunes; just six years ago, in 2014, Exxon was the largest company in the United States by market cap, peaking at more than $415 billion. In the years since, the company’s value has declined to $180 billion, and is now dwarfed by tech giants such as Apple, Amazon, Microsoft, and Google.
The catalyst for the changes was the 4:1 stock split undergone by Apple on August 31. The Dow is unique among widely followed indices in that it is price-weighted rather than market-cap weighted (i.e., the weight of a company in the index is determined by the value of a single share of its stock, rather than the total value of the company). The split reduced Apple’s weight in the index from around 12% to about 3%, and lowered the weight of the information technology sector from 28% to 20%. The index is ostensibly designed to track the state of the US economy, and, given that the committee has made little secret of the fact that they have been trying to shed the index’s historically industrial-heavy background for years, the changes come as little surprise.
Change to Federal Reserve Framework
In 2019, the Federal Reserve announced the launch of a first-ever review of the monetary policy framework – the tools, communication practices, and overall strategy – employed in pursuing the goals of maximum employment and price stability mandated by Congress (also known as the “dual mandate”). Federal Reserve Chairman Jerome Powell recently announced the conclusions of this review at the annual Jackson Hole Economic Policy Symposium, which was held virtually, and, for the first time, was open to the public. The symposium kicked off with a speech from Chair Powell announcing the conclusions, and was accompanied by the release of a revised Statement on Longer-Run Goals and Monetary Policy Strategy. As was widely expected, the main outcome of the review was the adoption of average inflation targeting (AIT), a shift in how the Federal Reserve will approach the Congressional mandate of price stability. Going forward, the Federal Open Market Committee (FOMC) will aim for inflation moderately above 2% following periods when inflation has run persistently below 2%. Inflation has been persistently below the FOMC’s stated target of 2% in the decade since the Great Financial Crisis. Since the start of the expansion leading out of the GFC to now, core PCE, the Fed’s preferred inflation measure, has averaged 1.6%. Looking at monthly data from June 2009 to February of this year, reported core PCE measured 2% or greater in just 14 of the 128 months during the period.
During this time, members have consistently reiterated that the 2% target is meant to be symmetrical – meaning that inflation misses to the upside as often as it misses to the downside – and the adjustment to the framework represents a formalization of this consistently communicated goal.
The prior Federal Reserve framework was in large part dominated by the standard Taylor Rule reaction function. A detailed technical explanation of the Taylor Rule is beyond the scope of this update, but the basic concept is that the rule dictates that the FOMC should tighten monetary policy when realized inflation is higher than their target and/or the level of unemployment is lower than the Fed’s estimate of the natural rate of unemployment, and loosen monetary policy when the reverse is true. Implicit in this framework is a reliance on the Phillips curve, which is an economic concept derived empirically from historical data, stating that inflation and unemployment have a stable and inverse relationship (i.e., when unemployment is low, inflationary pressure increases). Historically, the Phillips curve was well-specified; as unemployment declined, inflation tended to increase, and as unemployment rose, inflation tended to decrease. This relationship, however, has not held during the past 20 years or so – the level of inflation has remained low, regardless of the level of unemployment. This phenomenon has been referred to as the “flattening” of the Phillips curve, and the recently announced adjustment to the FOMC’s monetary policy framework can largely be thought of as a response to this new reality. Put simply, the unemployment rate is no longer as important as it was previously for the Federal Reserve when determining the course of monetary policy.
In a research piece examining the policy framework adjustments, economists at Goldman Sachs offered two separate interpretations of the adjustment in the context of the standard Taylor Rule. The first is that, under AIT, the Taylor Rule framework would temporarily be modified, so that policy rule inflation target would be raised (say, from 2% to 2.5%) when inflation over some trailing period has averaged less than 2%. The second interpretation would add the cumulative inflation shortfall over some trailing window (say, the period since the beginning of the last recession) as an additional term in the Taylor Rule function. Each interpretation has the similar effect of allowing for higher levels of inflation following periods in which realized inflation falls short of the stated 2% target. Goldman’s team then ran simulations using the Fed’s macroeconomic model, FRB/US, to compare both moderate and aggressive interpretations of AIT policy to the old framework, in an attempt to determine what the framework adjustment will mean in practice. In short, the adjustments under AIT policies generate lower forecasted funds rate paths, slightly higher levels of inflation, and lower unemployment rates than the old framework.
Notably, however, each returns average inflation back to 2% very slowly over a number of years. In comparing the moderate AIT policy interpretation to the more aggressive approach, the more aggressive policy interpretation generates a much lower funds rate path forecast and results in somewhat larger economic effects. In announcing the adjustments to the framework, Chair Powell emphasized that the FOMC will remain “flexible,” and will not be tied to a particular mathematical formula in guiding monetary policy going forward. Another notable change is the announcement that the FOMC will now only respond to “shortfalls” of employment from its maximum level, rather than deviations both above and below this maximum level. In practice, this means that the FOMC will no longer preemptively tighten monetary policy in the face of a strong labor market.
The bottom line is that we expect monetary policy to be easy for the next several years, and the new framework suggests that policy may remain easier for a few quarters longer than was likely under the old framework. The realistic impact of this shift is likely to be relatively small during the initial onset of a recession. With the policy rate already at the lower bound, the result of the lower-for-longer policy implied by AIT will make only a limited incremental contribution to the economy during the depths of a recession. Furthermore, returning inflation to 2% will still require a quite lengthy expansion under the assumption that the Phillips curve remains flat, as has been the experience this century.
Despite this change in framework for the Federal Reserve, evidence indicates that the impact of the shift will be minimal during a recession and early recovery like the one we are experiencing today. Fiscal policy, therefore, has to do much of the heavy lifting, once the tools of monetary policy have been exhausted. The strong rebound the US economy has experienced in the early months of the recovery was largely fueled by the unprecedented amount of fiscal stimulus flowing through the system. Should Washington fail to put partisan differences aside and reach a deal to extend much-needed aid to American families and small businesses, the initial hopes for a strong V-shaped recovery will prove hollow, and economic growth will be lower going forward, as we recover from the worst drop in economic activity in history.
