Macro Roundup

Bart Hobijn and Ayşegül Şahin, "Maximum Employment and the Participation Cycle," National Bureau Of Economic Research, September 2021, https://www.nber.org/papers/w29222

Bart Hobijna and Aysegul Sahin argued that labor force participation is cyclical and lags the unemployment cycle; that LFP is driven by upticks in employment stability, for example lower job-loss risk and tight labor markets as opposed to labor force entry and exit. Unemployed workers are more likely to drop out of labor force than employed worker, ultimacy employment stability (or instability) is driving the participation cycle, "....The decomposition reveals thatthe source of the participation cycle is fluctuations in job-loss and job-finding rates, rather than cyclical movements in labor force entry and exit rates. The magnitude of the participation cycle is large. Cyclical downward pressures on employment from participation are two-thirds that of unemployment. Moreover,the participation cycle delays the recovery in employment because it lags the unemployment cycle. It also amplifies the unevenness of the impact of recessions. Groups that see large increases in their unemployment rates also experience more pronounced participation cycles. Despite differences in their magnitudes, the source of the participation cycle is the same for all groups..."

This implies that, "...the bulk of the drop in the participation rate since the onset of the pandemic is cyclical and that the cyclical recovery in participation likely will trail that of the unemployment rate.....They result in the estimate that 1.5 percentage points of the 1.7-percentage-point decline in the participation rate from February 2020 to June 2021 is due to the cyclical drag on the participation rate associated with the deterioration of job-loss and job-finding prospects since the onset of the pandemic. This result holds not just for the aggregate data. For all groups in our sample, the bulk of the decline in their participation rates is due to the participation cycle...."
Their dynamic, “..The entry and exit component captures the part of changes in the participation rate that can be traced back to changes in the probabilities of workers flowing across the participation margin, i.e., into and out of the labor force. This component puts upward pressure on participation at the beginning of recessions when the likelihood that the employed and unemployed drop out of the labor force goes down. Mostly, however, it captures the time-varying long-run trend behavior of the participation rate. The procyclical pressures on the participation rate come from the cycle component, which we call the participation cycle in the rest of this paper. These are the pressures exerted on labor force participation by job-loss (flows from employment to unemployment), and job-finding (flows from unemployment to employment). At first glance, it might sound puzzling that the flows that do not involve crossing the participation margin are the source of the procyclicality of the labor force participation rate. The intuition comes from recognizing that those who are unemployed are substantially more likely to drop out of the labor force than the employed.Specifically, the exit rate from the labor force for the unemployed averaged around 25 percent in 1978-2019. This is almost an order of magnitude larger than the labor force exit rate of employed workers which averaged 2.8 percent during the same period. The difference in these exit rates from the labor force creates a wedge that we refer to as the attachment wedge. The higher the fraction of the labor force that is unemployed, i.e., the higher the unemployment rate, the more likely workers are to drop out of the labor force in the future, which lowers the participation rate going forward. Since movements in job-loss and job-finding are the main drivers of fluctuations in the unemployment rate (Elsby et al., 2015), the unemployment cycle and the participation cycle are closely tied together. In terms of magnitude, we find that the procyclical forces that affect the labor supply during business cycles are large. Our analysis reveals that the downward pressure that the participation cycle puts on the employment-to-population (EPOP) ratio is about two-thirds that of the unemployment cycle. Moreover, the participation cycle lags the unemployment cycle. On average, the participation cycle bottoms out nine months after the unemployment rate peaks. This lag is even longer for deeper recessions. The participation and unemployment cycles tend to converge during labor market recoveries, moving in tandem especially in later phases of expansions. When the unemployment rate gets close to or is below its natural rate, declines in unemployment and cyclical upward pressures on participation have about the same effect on the EPOP ratio. This observation is consistent with the Perry-Okun rule that during strong labor markets a 1-percentage-point decline in the unemployment rate results in cyclical upward pressure of 0.65 percentage point on the participation rate. Using a very different methodology and more recent data we find a rule of thumb very similar to Perry (1971) and Okun (1973)... Our results establish that increases in employment stability for those in the labor force during strong labor markets put upward pressure on participation. That is, the reductions in job-loss and increases in job-finding when labor market conditions improve result in fewer and shorter periods of unemployment (see Marston, 1976), which reduces the likelihood that participants drop out of the labor force...."

Implications, "...The source of the upward pressures on participation during expansions has important policy implications especially in light of the maximum employment mandate being a broad based and inclusive goal. If the prevailing view were true, progress towards maximum employment when the labor market is hot and the unemployment rate is low mainly would be made through the re-entry into the labor force of workers from marginalized groups. Therefore, it would be important for policymakers to focus on movements in the participation rates of the groups of workers that bear the brunt of labor market downturns when the unemployment rate is low. Instead, our results suggest there is little need to shift the attention from the unemployment to the participation rate, especially of marginalized groups, at the tail end of expansions because the same forces are at play for all groups in the labor market. For all groups, the Perry-Okun rule is a reasonable rule of thumb. This means that, during the latter stages of the business cycle when the unemployment rate is low, cyclical upward pressures on participation move almost in lockstep with changes in unemployment rates. This is because for all groups the main procyclical forces on participation are driven by fluctuations in job-loss and job-finding rates that also account for the bulk of the movements in the unemployment rate. However, the magnitude of the participation cycle is highly uneven across groups as a consequence of the differences in the cyclicality and levels of unemployment rates. We find that the participation cycle amplifies the well-documented unevenness of recessions, as captured by different increases in group-specific unemployment rates. Groups with a higher increase in the incidence of unemployment also have larger procyclical pressures on their participation rate. Therefore, the groups hardest hit during recessions have the largest cyclical upward pressures on participation during recoveries and expansions. This includes low-skilled workers and workers who identify as Black, or African American, and Hispanic. Our new finding complements Wolfers’ discussion of Aaronson et al. (2019) where he shows that disadvantaged groups with higher unemployment rates also tend to have more cyclical unemployment rates….”

Evidence"... Figure 7 plots the cumulative contribution of the entry/exit and cycle to the change in the LFPR since the start of 1978 for the total population for 1978-2019 in panel (a).25 The entry/exit component is the main driver of the long-run trend in the LFPR. Interestingly, it also exhibits a countercyclical pattern putting upward pressure on the LFPR during recessions and a downward pressure later in expansions. This finding challenges the popular view that the procyclicality of the participation rate has its origins in discouraged workers leaving the labor force during recessions and re-entering the labor force as labor market conditions improve. We find that the net contribution of labor force entry and exit is not procyclical at all. On the contrary, the entry/exit component is decisively countercyclical, pushing against the procyclical forces we identify. It is the exit component that is responsible for the upward trend in participation from 1978 through 2000, largely driven by women (panel (c)). The increase in labor force participation among women was not because women was not because those who were not part of the labor force became more likely to join. Instead, it was driven by the increased attachment to the labor force of those women who already were part of it. While the entry component was relatively muted for women until the 2000s, it accounts for most of the decline since early 2000 in the participation rate of both men and women...."

"...The results are reported in the bottom half of Table 3. The “cycle” column of the row labeled “Total” shows that the estimated cyclical downward pressures on participation in 2021 are 0.4 percentage points lower than those in 2014. This finding has three important implications. First of all, in spite of a lower participation rate, the cyclical downward pressures on participation in 2021 are close to those in 2014. Thus, the similarities between early fall 2014 and the summer of 2021 extend beyond the unemployment and labor market flow rates to the participation cycle. The second is that it provides information about the average annual trend decline in LFPR. If the participation cycle now is comparable with that in September 2014, as the table suggests, then this implies that most of the 1.2-percentage-point decline in the participation rate from September 2014 through June 2021 is due to the secular downward trend in the participation rate. This observation suggests that the annual average decline in trend participation was about 0.17 percentage points a year during that period.41 This is in line with the estimate of the CBO of around 0.2 percentage points a year and slightly lower than the cohort-based estimates presented in Figure 3. The third implication is that the cyclical gains in participation that accumulated from 2014 through 2019 have not completely been erased by the COVID-19 Recession. On the contrary, the results by group in the “cycle” column of the second half of Table 3 show that cyclical downward pressures on the participation rates of young workers, those with less than a high school education, as well as Black or African Americans are all estimated to be a percentage point or more lower now than in the wake of the Great Recession. Put differently, the expansionary labor market in the 2014-2019 period helped create employment stability for these groups, which increased their attachment to the labor force. Even though the COVID-19 shock brought the participation cycle to a similar point as in September 2014, these groups still carried over part of the gains of the previous expansion, as evident in the differences in the cycle component in the second half of Table 3...."

Greg Mankiw footnotes his recent NYT op-ed, "...economists disagree about the how far the tax-based explanation goes. A reasonable reading of the literature is that lower labor effort and incomes in Europe are likely due to a combination of higher tax rates, stronger unions, and greater regulations......we find that the divorce and tax mechanisms together can explain 45% of the variation in labor supply between the United States and the European countries."......Europeans today work much less than Americans because of the policies of the unions in the 1970s, 1980s, and part of the 1990s and because of labor market regulations. Marginal tax rates may have also played a role......Based on our reading of the micro evidence, we recommend calibrating macro models to match Hicksian elasticities of 0.3 on the intensive and 0.25 on the extensive margin and Frisch elasticities of 0.5 on the intensive and 0.25 on the extensive margin. Hence, it would be reasonable to calibrate representative agent macro models to match a Frisch elasticity of aggregate hours of 0.75. These elasticities are consistent with the observed differences in aggregate hours across countries with different tax systems..."

Follow-up references

In my most recent Times column, I did not have the space to fully explain the body of work that follows up on the Prescott hypothesis that higher tax rates explain lower work effort and national incomes in Western Europe. For interested readers (that is, the more nerdy ones), here are a some relevant references together with brief excerpts (emphasis added):

1. Steven Davis and Magnus Henrekson

"Lastly, let us return to the recent studies by Prescott (2002, 2003), which consider the output, employment and welfare consequences of personal taxes in an equilibrium model with one production sector and a simple labor-leisure choice for the representative household. Our evidence supports the view that tax rate differences among rich countries are a major reason for large international differences in market work time. At the same time, however, our evidence strongly suggests that labor and consumption taxes operate with powerful effect on several margins: substitution between legal and underground activity, substitution between home and market production, the mix of market production activity, and the composition of market expenditures."

2. Indraneel Chakraborty et al.

"Americans work more than Europeans. Using micro-data from the United States and 17 European countries, we document that women are typically the largest contributors to the cross-country differences in work hours. We also show that there is a negative relation between taxes and annual hours worked, driven by men, and a positive relation between divorce rates and annual hours worked, driven by women. In a calibrated life-cycle model with heterogeneous agents, marriage and divorce, we find that the divorce and tax mechanisms together can explain 45% of the variation in labor supply between the United States and the European countries."

3. Alberto Alesina et al.

"Our punch line is that Europeans today work much less than Americans because of the policies of the unions in the 1970s, 1980s, and part of the 1990s and because of labor market regulations. Marginal tax rates may have also played a role, especially for women's labor force participation, but our view is that in a hypothetical competitive labor market without unions and with limited regulation, these tax increases would not have affected hours worked as much. Certainly micro evidence on the elasticity of labor supply is inconsistent with a mainly tax-based explanation of this phenomenon, even though social multiplier effects may help in this respect."

4. Raj Chetty et. al.

"Based on our reading of the micro evidence, we recommend calibrating macro models to match Hicksian elasticities of 0.3 on the intensive and 0.25 on the extensive margin and Frisch elasticities of 0.5 on the intensive and 0.25 on the extensive margin. Hence, it would be reasonable to calibrate representative agent macro models to match a Frisch elasticity of aggregate hours of 0.75. These elasticities are consistent with the observed differences in aggregate hours across countries with different tax systems."

As I noted in my column, economists disagree about the how far the tax-based explanation goes. A reasonable reading of the literature is that lower labor effort and incomes in Europe are likely due to a combination of higher tax rates, stronger unions, and greater regulations.

Greg Mankiw, “Follow Up References,” Greg Mankiw’s Blog, September 22, 2021, http://gregmankiw.blogspot.com/2021/09/follow-up-references.html

Justin Fox takes a snapshot of autumn 2021 in New York, "...Payroll employment in the city fell 19.5% in the first three months of the pandemic and isstill down almost 11% from February 2020 on a seasonally adjusted basis. Nationwide the decline is now just 3.5%, and a not-seasonally-adjusted long view (seasonally adjusted numbers are only available back to 1990) shows how massive the recent hit to New York has been from a historical perspective.... Weekday subway ridership in New York City, meanwhile, is now at a little over 50% of the pre-pandemic norm....At present, visits to retail and recreational establishments in Manhattan (a city-wide estimate for New York is not available) are still down more than 60% from the pre-pandemic norm, a decline similar to that in European cities but much bigger than in Hong Kong or Mexico City....OpenTable’s statistics on seated diners at the restaurants using its reservation service, meanwhile, show New York about even with San Francisco near the bottom of the pack....he picture is even worse for the city’s lodging industry, with a recent report from the American Hotel & Lodging Association and Kalibri Labs projecting an 88.4%, decline in business travel revenue for New York in 2021 relative to 2019.... In the weekly office occupancy statistics maintained by Kastle Systems, New York was ahead of only San Francisco and San Jose with office security-card swipes at 28.1% of the pre-pandemic norm in the week of September 15.... Zumper’s asking-price data, meanwhile, shows the median one-bedroom rent in New York hitting $2,810 in August, pushing it past San Francisco as the country’s most expensive market. These rising rents are a sign that confidence in the city’s future is returning, but they’re also somewhat tragic, given that most New Yorkers can’t afford $2,810 one-bedroom apartments...."

Justin Fox, "Is New York Back? Crime and Jobs Data Point to a Long Recovery," Bloomberg, September 23, 2021, https://www.bloomberg.com/opinion/articles/2021-09-23/new-york-city-is-still-a-shell-of-its-former-self

Is New York Back? Crime and Jobs Data Point to a Long Recovery

One could get the impression from reading headlines these days that New York City is back. Landlords are jacking up apartment rents by as much as 70% from pandemic lows. Google just said it would buy a downtown Manhattan building for $2.1 billion, the biggest office transaction in the U.S. since Covid-19 hit. Subway ridership is reaching new pandemic-era highs. And so on.

Indeed, the city has come far from the awful days of spring 2020. By a few metrics — apartment rents, mainly — it really is back to where it was before the pandemic. By most others, though, it still falls miles short of a full recovery.

I write this, I might as well admit, partly in response to my San-Francisco-based Bloomberg Opinion colleague Noah Smith’s claim last week that New York “is emerging as one of the world’s most resilient big cities in the wake of the pandemic.” Sure, it looks pretty lively in comparison with San Francisco, where hardly anybody has gone back to the office and tourist zones like the area around Union Square remain deserted (or at least were when I was there a few weeks ago). It also isn’t spiraling ever downward in a repeat of the terrible 1970s, as some feared last summer. There are even tantalizing hints that it may come back as something better than it was — at least that’s what I told myself while biking home at dusk on Sunday after dropping off a U-Haul truck in East Harlem (long story) and encountering scene after lovely scene of New Yorkers dining outdoors, playing and socializing on closed-off streets, and getting around on two-wheeled vehicles of many varieties.

New York City is going to have trouble making much more headway, though, without a far bigger jobs recovery than it has seen so far. Payroll employment in the city fell 19.5% in the first three months of the pandemic and is still down almost 11% from February 2020 on a seasonally adjusted basis. Nationwide the decline is now just 3.5%, and a not-seasonally-adjusted long view (seasonally adjusted numbers are only available back to 1990) shows how massive the recent hit to New York has been from a historical perspective.

Weekday subway ridership in New York City, meanwhile, is now at a little over 50% of the pre-pandemic norm. That’s much higher than Bay Area Rapid Transit (in San Francisco and environs) and the Washington Metro have seen so far, an indication of how much more reliant New Yorkers are on public transportation. But it’s about the same as what the London Underground is experiencing and well short of where the Paris Metro and commuter-rail systems were early in the summer.

Google’s community mobility statistics offer another way to compare how much people are getting out and about around in the world. At present, visits to retail and recreational establishments in Manhattan (a city-wide estimate for New York is not available) are still down more than 36% from the pre-pandemic norm, a significantly bigger decline than in Hong Kong, Berlin, Mexico City or Amsterdam.

Apartment rents are of course another matter, with Apartment List’s estimate of the median rents paid by recent movers showing New York’s surpassing pre-pandemic levels for the first time in August. That’s in contrast to San Francisco and Washington, where rents still haven’t fully recovered. On the other hand, rents in the Florida coastal city of Boca Raton have gone from 11% lower than New York’s to 7% higher over the course of the pandemic.

These rising rents are a sign that confidence in the city’s future is returning, but they’re also somewhat tragic, given that most New Yorkers can’t afford $2,810 one-bedroom apartments. Unlike San Francisco and its suburbs (and even more, the San Jose-Sunnyvale-Santa Clara metro area just to its south), New York is not especially affluent. The city’s median household income has been barely higher than the nation’s in recent years, and probably dropped below it in 2020. There were indications that high real estate prices were driving people away from the city before the pandemic hit, and while New York’s surprisingly high population count in the 2020 Census was taken by some as evidence that earlier Census estimates showing a population decline starting in 2016 were wrong, I think that’s a misreading: the population decline may well have been real, just starting from a higher base than thought.

The spasm of violent crime that began last summer is another problem. It’s worse in other places than in New York, where murders are now trending downward and shootings have stopped rising relative to last year, but the city remains more dangerous than it was before the pandemic. It’s also gotten a lot dirtier because of Sanitation Department cutbacks and worker shortages — which again are not unique to New York but are an especially big challenge here. Finally, while tax collections have come in a lot stronger than forecast last year and early this year, they’re still running billions short of what the city will need to pay its bills when federal pandemic aid runs out.

The 20th anniversary of the Sept. 11, 2001 terrorist attacks provided an opportunity to marvel at New York’s past resilience. In the immediate aftermath of 9/11 the city seemed permanently damaged. Some speculated that big cities and tall office buildings would go out of fashion. Instead, the country’s biggest city came roaring back.

Will it do so again? Yeah, I think so. But Covid-19 has killed many more New Yorkers and caused far more economic damage than 9/11 did, and the remote-work revolution it spawned has already been more wide-reaching and durable than post-9/11 tall-building-phobia ever was. It’s going to be a long, hard road to recovery for New York, and acting like we’re already there seems like a big mistake.