Macro Roundup

Douglas Belkin, "What If Colleges Used Only Test Scores to Fill Campuses?,"Wall Street Journal, June 23, 2019, https://www.wsj.com/articles/what-if-colleges-used-only-test-scores-to-fill-campuses-11561282201

What If Colleges Used Only Test Scores to Fill Campuses?

As Americans debate the roles of wealth, race and access in higher education, researchers at Georgetown University wondered: What would the nation’s most selective colleges and universities look like if they admitted students solely on the basis of SAT scores?

Their answer: The most prestigious U.S. campuses would be wealthier, whiter and more male—leading researchers to question the role standardized testing plays in a fair, comprehensive admissions process.

More than half the students now enrolled at the top 200 colleges and universities would lose their seats to students who performed better on the test—and the median SAT score would rise by 70 points to 1320, the study found.

The biggest losers in this reshuffling were black and Latino students, whose numbers would be cut nearly in half, to 11% of all students from 19%. The share of Asian students would slip to 10% from 11%. The principal winners were wealthy white male students, whose ranks would increase. But a large number of white students would lose their seats and be replaced with other white students.

Many elite colleges consider a range of factors such as grades, extracurricular activities and teacher recommendations in what they call a holistic admissions approach. They also try to build a diverse student body and might look for students from underrepresented ethnic or racial backgrounds or geographical regions.

The holistic admissions process has come under fire following the disclosure this spring of the biggest cheating ring federal prosecutors say they have ever seen. In addition, pending lawsuits filed against Harvard University and the University of North Carolina, Chapel Hill, among others, charge that the schools racially discriminate against Asian applicants.

“The SAT does not and should not measure excellence on its own. Data are overwhelming that grades and test scores together better predict college success than either does alone,” a spokesman for The College Board said in a statement. “Comprehensive research demonstrates that sustained commitment to an activity in high school outside of class further predicts success in college and beyond. Resourcefulness in response to challenges has long been honored in college admissions as a dimension of merit and success in life. A focus on a single score would leave so much talent unseen.”

The College Board, the New York-based nonprofit that oversees the SAT, is introducing an adversity score to accompany a student’s SAT results. The score is an amalgamation of socioeconomic data that measures factors like the crime rate, housing cost, family structure and school resources where an individual student lives and studies.

Researchers at Georgetown sorted admissions at the nation’s top schools as determined by Barron’s Profiles of American Colleges. They took the 300,000 seats from those colleges and universities and filled them with top SAT and ACT scores earned by the 2.7 million high-school seniors who took the test in 2012. Then they compared who got the highest test scores to who got in by the actual admissions system, using Education Department data that didn’t include the students’ names.

The researchers used a composite score that standardizes both the SAT and ACT scores in terms of the SAT’s grading scale. Using that scale, the median SAT score of accepted students climbed to 1320 from 1250. The new bar for admissions meant that 159,000 students lost their seats.

The percentage of male students at the schools climbed to 54% from 50%. The most dramatic increase was among white students, who grew to 75% from 66% of the aggregated student body. The proportion of students coming from families in the top quartile of income distribution—where the median income is $122,000—increased to 63% from 60%.

“When used as part of a holistic strategy, standardized testing offers a check on opaque admissions, but it’s still no guarantee of fairness,” Mr. Carnevale said. “It’s time to find a better approach.”

"...We will pursue aggressive new government policies to support American workers. And we will start with two major changes: Aggressively using all of our tools to defend and create American jobs…. Consolidating existing government programs that affect job creation into a new agency with the sole responsibility to create and defend quality, sustainable American jobs…”

Elizabeth Warren, "A Plan For Economic Patriotism,"Medium, June 4, 2019, https://medium.com/@teamwarren/a-plan-for-economic-patriotism-13b879f4cfc7

A Plan For Economic Patriotism

I come from a patriotic family. All three of my brothers joined the military. And I’m deeply grateful for the opportunities America has given me. But the giant “American” corporations who control our economy don’t seem to feel the same way. They certainly don’t act like it.

Sure, these companies wave the flag — but they have no loyalty or allegiance to America. Levi’s is an iconic American brand, but the company operates only 2% of its factories here. Dixon Ticonderoga — maker of the famous №2 pencil — has “moved almost all of its pencil production to Mexico and China.” And General Electric recently shut down an industrial engine factory in Wisconsin and shipped the jobs to Canada. The list goes on and on.

These “American” companies show only one real loyalty: to the short-term interests of their shareholders, a third of whom are foreign investors. If they can close up an American factory and ship jobs overseas to save a nickel, that’s exactly what they will do — abandoning loyal American workers and hollowing out American cities along the way.

Politicians love to say they care about American jobs. But for decades, those same politicians have cited “free market principles” and refused to intervene in markets on behalf of American workers. And of course, they ignore those same supposed principles and intervene regularly to protect the interests of multinational corporations and international capital.

The result? Millions of good jobs lost overseas and a generation of stagnant wages, growing inequality, and sluggish economic growth.

If Washington wants to put a stop to this, it can. If we want faster growth, stronger American industry, and more good American jobs, then our government should do what other leading nations do and act aggressively to achieve those goals instead of catering to the financial interests of companies with no particular allegiance to America.

It’s not a question of more government or less government. It’s about who government works for.

That’s why today I’m announcing that, as President, I would pursue an agenda of economic patriotism, using new and existing tools to defend and create quality American jobs and promote American industry.

My Administration will pursue fundamental, structural changes in our government’s approach to the economy, finally putting American workers and middle-class prosperity ahead of multinational profits and Wall Street bonuses.

In the weeks ahead, I’ll be releasing longer individual plans on how economic patriotism should shape our approach to specific parts of the American economy, from trade policy to Wall Street. All of these proposals will share this common vision for economic policymaking in America. Today I’m also releasing the first specific example — a plan for American manufacturing.

But first, let me explain how economic patriotism works.

An End to the Excuses

It’s time to reject the excuses we’ve heard for decades about why we can’t do more to help American workers.

Some people blame “globalization” for flat wages and American jobs shipped overseas. But globalization isn’t some mysterious force whose effects are inevitable and beyond our control. No — America chose to pursue a trade policy that prioritized the interests of capital over the interests of American workers. Germany, for example, chose a different path and participated in international trade while at the same time robustly — and successfully — supporting its domestic industries and its workers.

Others blame “automation” for American job losses, especially in manufacturing. It’s a good story — robots and other new technologies made American manufacturing workers more productive, so companies needed to hire far fewer actual human beings. A good story, except it’s not really true. Recent research finds this story is based on a widely-held misunderstanding of the data on American manufacturing output, and a statistical quirk about how productivity is measured in our computer industry. There is actually no “evidence that productivity caused manufacturing’s relative and absolute employment decline” in America since the 1980s. Meanwhile, Germany has nearly five times as many robots per worker as we do and has not lost jobs overall as a result.

And a lot of people blamed a supposed “skills gap” for job losses — that American workers lacked the skills or credentials they needed to fill the jobs available. Except that wasn’t true either. It was just a symptom of high unemployment rates. Companies felt comfortable demanding more skills from workers as an excuse to be more selective about which workers to hire.

The truth is that Washington policies — not unstoppable market forces — are a key driver of the problems American workers face. From our trade agreements to our tax code, we have encouraged companies to invest abroad, ship jobs overseas, and keep wages low. All in the interest of serving multinational companies and international capital with no particular loyalty to the United States.

In my administration, we will stop making excuses. We will pursue aggressive new government policies to support American workers. And we will start with two major changes:

Aggressively using all of our tools to defend and create American jobs. The prevailing view in Washington — from both political parties — has been that our government should not aggressively intervene in the markets to boost American workers. (This “rule” goes out the window when it comes to subsidizing Wall Street and multinational corporations.) We have tried that approach, and it has failed spectacularly. From our own experience and the experience of other countries, we know what types of government actions actually work to promote sustainable job growth and industrial development. It’s time to have the courage to pick up the tools we have and use them.

Consolidating existing government programs that affect job creation into a new agency with the sole responsibility to create and defend quality, sustainable American jobs. The new Department — the Department of Economic Development — will replace the Commerce Department, subsume other agencies like the Small Business Administration and the Patent and Trademark Office, and include research and development programs, worker training programs, and export and trade authorities like the Office of the U.S. Trade Representative. The new Department will have a single goal: creating and defending good American jobs.

Aggressive Intervention on Behalf of American Workers

If we can aggressively intervene in markets to protect the interests of the wealthy and well-connected — as we have for decades with bailouts and subsidies — then we can damn well use all the tools at our disposal to protect the interests of American workers. That’s why we should use a variety of more aggressive tactics, including:

More actively managing our currency value to promote exports and domestic manufacturing. One of the most important factors in our trade deficit and our weak export levels is the value of our currency. Other countries have actively managed the value of their currency to boost exports and develop their domestic industries. And foreign investors and central banks have driven up the value of our currency for their own benefit. We should consider a number of tools and work with other countries harmed by currency misalignment to produce a currency value that’s better for our workers and our industries.

Leveraging federal R&D to create domestic jobs and sustainable investments in the future. We spend only half as much as we did in the 1980s on federal research and development. Meanwhile, when American taxpayers do invest in R&D, we often see American companies take that research and use it to manufacture products overseas, like Apple did with the iPhone. The companies get rich, and American taxpayers have subsidized the creation of low-wage foreign jobs. Other countries have adopted different approaches to public R&D funding that have produced strong outcomes for the economy and for taxpayers. Learning from these approaches, my administration will substantially increase our investments in R&D, but with three critical new conditions:

Production stemming from federally funded research should take place in the United States. If taxpayer investments in R&D lead to new ideas and products, those products should be made here. The federal government already includes this requirement in some of its programs, but it should be a standard requirement, absent truly extraordinary circumstances.

Taxpayers should be able to capture the upside of their research investments if they result in profitable enterprises. Like any investor, taxpayers should get a return on the risky investments they are making in R&D. That can take various forms. Taxpayers can: get an equity stake in any company that relies on intellectual property these investments create; retain royalties on publicly funded innovation or a golden-share of the patent revenue; or require the companies benefitting from publicly funded R&D to reinvest profits back into domestic production, R&D, and worker training programs, rather than into stock buybacks.

R&D investments must be spread across every region of the country, not focused on only a few coastal cities. There are talented Americans in every part of the country, but too often cities and towns experience brain drain and shrink because corporations move jobs and opportunities overseas or to a small handful of American cities. We must allocate R&D funding across the country, to ensure that there are economic opportunities in every region and that funding is targeted at solving regional problems.

Increasing export promotion to match the efforts of our competitors. In 2017, our main export promotion agency, the Export-Import Bank, provided $200 million in total medium- and long-term financing to support American exports. China’s equivalent agency provided more than a hundred times as much support, while Germany’s agency provided more than thirty times as much support. We must spend more to boost American exports so we can level the playing field for American workers. And while historically a large chunk of the Export-Import Bank’s support has gone to a handful of big companies, our export promotion should focus more on smaller and medium-sized businesses.

Deploying the massive purchasing power of the federal government to create markets for American-made products. The federal government spends hundreds of billions of dollars each year to purchase goods and services. We should require whenever possible for the government to purchase American-made products, and use large federal procurement commitments as a tool to create demand for new American-made goods and to develop particular domestic industries.

Restructuring worker training programs to deliver real results for American workers and American companies. Nearly half of the German workforce has graduated from a post-secondary apprenticeship program, which gives people access to good jobs without a four-year college education and provides German companies with a steady stream of capable workers. We should take aggressive steps to overhaul our worker training programs so they produce better results for American workers and companies.

Dramatically scale up apprenticeship programs. We currently invest $200 million annually in apprenticeship programs. We should increase that tenfold and make a $20 billion commitment to apprenticeship programs for the next ten years. These efforts should bring together community colleges, technical schools, unions, and companies.

Institute new sectoral training programs. We should also create sectoral training programs — a model that has been successful in Wisconsin. These local or regional sector training partnerships would help align training with the local job market, leverage the community college system, and, by designing training based on an entire sector, ensure that workers gain skills that are transferable across employers.

Economic patriotism is about using all the tools we have to boost American workers and American industries so they have the best opportunity to compete internationally. While those tools can include certain things like tariffs, our principal goal should be investing in American workers rather than diminishing our competitors. If our workers are on a level playing field, I know they can take on any challenger and win.

The Department of Economic Development

Our international competitors like China, Germany, and Japan develop concrete plans for promoting domestic industry and then make serious investments to achieve their goals. China’s Made in China 2025 plan aims to dominate advanced manufacturing in the coming decades using various incentives and “hundreds of billions of euros” in subsidies. Germany and Japan have also developed plans that identify long-term goals for domestic production and put real money behind achieving them.

This is a pretty straightforward idea. But outside of the defense context, the United States has nothing remotely like it.

Government programs that affect job creation are an afterthought, uncoordinated and scattered across the government, and submerged in larger agencies with different primary missions:

There are 58 programs in 11 federal agencies that provide support to American manufacturing — all of them tacked on to the primary missions at those agencies.
There are at least nine offices in five different agencies primarily responsible for trade policy and export promotion.
And there are 47 different employment and training programs spread across nine different federal agencies.

Even worse, there are some government agencies that undermine sustainable American jobs. For example, the Office of the United States Trade Representative — whose mission is to negotiate trade deals on behalf of America — is captured by the interests of corporate executives and lobbyists. Its actions across Administrations demonstrate a deep ideological opposition to anything that might put the interests of American workers above the interests of multinational corporations or Wall Street.

We should put all of these offices and programs in the same place, get rid of the ones that are redundant or don’t work, and bring the rogue ones to heel — to make it clear that the unified mission of the federal government is to promote sustainable, middle-class American jobs.

That’s why I’m proposing the creation of a new agency — the Department of Economic Development — that will have the single goal of defending good-paying American jobs and creating new ones.

The Department will be responsible for creating a National Jobs Strategy (NJS) every four years, just as countries like Germany and China produce regular strategic plans. The NJS will be a long-term plan that examines the worldwide economic environment and identifies new risks and opportunities. It will focus not just on the overall American economy, but on regional economies. It will examine trends that have disproportionate effects on rural communities and smaller cities. And it will establish clear goals for American jobs and American industry that will guide how the Department of Economic Development prioritizes its investments and direct its programs.

Critically, all of our trade-related programs will fall within the new Department. By placing our trade programs within this new Department, we will make clear that trade policy must defend and create American jobs.

________________________________

It’s becoming easier and easier to shift capital and jobs from one country to another. That’s why our government has to care more about defending and creating American jobs than ever before — not less. We can navigate the changes ahead if we embrace economic patriotism and make American workers our highest priority, rather than continuing to cater to the interests of companies and people with no allegiance to America.

Greg Mankiw, "The National Debt Is Still a Problem,"New York Times, June 20, 2019, https://www.nytimes.com/2019/06/20/business/national-debt-trump.html

The National Debt Is Still a Problem

“I’m the king of debt.” So declared Donald Trump while running for president in 2016. Alas, he was not kidding.

When President Trump moved into the White House, the government debt held by the public stood at $14.4 trillion. The most recent figure is $16.1 trillion. And under current law, that number will keep rising with no end in sight.

The debt rises when the government borrows to finance a budget deficit, and in some circumstances, running a budget deficit is reasonable.

Historically, the most important cause of large deficits has been spending on wars. During World War II, for example, deficits ballooned, and the government debt increased to 106 percent of gross domestic product in 1946 from 37 percent in 1941.

The alternative to running a large budget deficit during a major military conflict is substantially raising taxes. Doing so would be economically debilitating. It would also arguably be unfair to current taxpayers, who are already offering their blood and toil to preserve freedom for themselves and future generations. By running up the debt, the government requires future generations to foot part of the bill.

Running a large budget deficit is also reasonable during an economic downturn. Whenever the economy goes into recession, revenue from income and payroll taxes automatically falls, and spending on programs such as unemployment insurance automatically rises. Balancing the budget would require tax increases or spending cuts precisely when the economy can least handle them.

In fact, presidents of both parties often embrace deficit-enlarging policies to combat downturns. President Ronald Reagan cut taxes in 1981, and President George W. Bush cut them in 2001 and 2003 as the economy struggled with recessions. Similarly, President Barack Obama signed a sizable stimulus bill in 2009 to deal with the effects of the financial crisis. While economists debate the relative merits of tax cuts and spending increases to prop up a moribund economy, few criticize budget deficits during a time of high and rising unemployment.

But the current budget deficit is not as easily justified. President Trump says his goal is to scale back military entanglements around the world, so temporary wartime spending is no excuse.

Nor is the state of the economy. It is currently strong, with employment having grown robustly over the past two years. The unemployment rate is 3.6 percent, its lowest level since 1969.

In these circumstances the norm should be a small budget deficit, or even a surplus. But that is not the case now. In 2018, the deficit was 3.9 percent of G.D.P., compared with an average deficit of 2.1 percent of G.D.P. over the previous 70 years.

Even so, the current deficit is less worrisome than its trajectory. The Congressional Budget Office estimates that, under current law, the deficit will increase to 4.6 percent of G.D.P. in 2023. That projection is based on the plausible but optimistic assumption of continued economic growth and no major military conflict.

Because of these deficits, the government debt held by the public will rise to 85 percent of G.D.P. in 2023, reaching the highest level since 1947, the Congressional Budget Office says. And if current law remains unchanged, the debt will keep rising as a percentage of G.D.P., eventually reaching levels never seen in the United States. That is simply not sustainable.

What is to be done? Perhaps the wisest words on this topic come from an old New Yorker cartoon. In it, the president’s advisers are huddled around his desk. They summarize the situation this way: “Our deficit-reduction plan is simple, but it will require a great deal of money.”

If we are not going to saddle future generations with ever-increasing government debt, we need to find a great deal of money. That means either spending less or taxing more.

I would prefer to curb spending. For example, to prevent Social Security’s funding shortfall from enlarging the government debt, we could slowly increase the age of eligibility. The government would still provide a safety net for the very old, but others would have to keep working or use their savings to pay for an earlier retirement.

But I recognize that my preferences might not reflect those of the body politic. Many people want a more expansive government to address such problems as the rising cost of health care and the explosion of student debt. Doing so would require higher taxes.

The conceit of some on the left is that all this can be done just by taxing the rich more. Yes, some revenue can be found there. Congress should reform the tax treatment of carried interest, which the finance industry uses to secure the lower capital gains rate for some forms of compensation. It should repeal the overly complicated and inequitable treatment of pass-through businesses in the 2017 tax bill. It could increase the corporate tax rate from 21 percent to, say, 25 percent, which would still be well below the 35 percent rate that prevailed when President Trump took office. Finally, it should strengthen tax enforcement so families like the Trumps pay what they owe.

But the problem of rising government debt is too large to solve simply by taxing a small sliver of the population. Congress will need to consider broader-based taxes, such as the value-added taxes that efficiently raise substantial revenue in many European nations.

If higher taxes are the eventual solution to rising government debt, as seems likely, then we all are going to have to pay more. If we don’t, our children will.

Chang-Tai Hsieh and Esteban Rossi-Hansberg, "The Industrial Revolution in Services," National Bureau of Economic Research, June 2019, https://www.nber.org/papers/w25968

new NBER offers nice datapoints supporting your tack that much of the rise in industry concentration in the US was driven by the rise of ~ "chain stores" in a broad sense (driven by their ability to produce at lower marginal costs)

"...We argue that the rise in industry concentration is due to companies similar to the Cheesecake Factory and Steward Healthcare that have adopted technologies that enable them to standardize and scale up the delivery of non-traded services. In this sense, what has happened in non-traded services is akin to the industrial revolution unleashed by Henry Ford more than a hundred years ago when Ford introduced mass production to a car industry dominated by independent artisans. We use micro-data from the Longitudinal Business Database from 1977 to 2013 to document the following facts.First, we show that the phenomena of rising concentration documented by Autor et al. (2017) is only seen in three broad sectors - services, wholesale, and retail.As Autor et al. (2017) suggest, top firms have become more efficient over time, but our evidence indicates that this is only true for top firms in these three sectors. In manufacturing, for example, concentration has fallen.Second, rising concentration in these sectors is entirely driven by an increase the number of local markets served by the top firms. Within a typical market served by a top firm in sectors with increasing concentration, we find that employment of top firms is either constant or falling. Specifically, we find that average employment per establishment of top firms falls in sectors with rising concentration.The same is true for employment of top firms in each county they serve.3Third, we find that total employment rises substantially in industries with rising concentration.This is true even when we look at total employment of the smaller firms in these industries. This evidence is consistent with our view that increasing concentration is driven by new ICT-enabled technologies that ultimately raise aggregate industry TFP. It is not consistent with the view that concentration is due to declining competition or entry barriers, as suggested by Gutierrez and Philippon (2017) and Furman and Orszag (2018), as these forces will result in a decline in industry employment.Fourth, we show that the top firms in the economy as a whole have become increasingly specialized in narrow set of sectors, and these are precisely the non-traded sectors that have undergone an industrial revolution. At the same time, top firms have exited many sectors. The net effect is that there is essentially no change in concentration by the top firms in the economy as a whole. The “super-star” firms of today’s economy are larger in their chosen sectors and have unleashed productivity growth in these sectors, but they are not any larger as a share of the aggregate economy.....The rise in national industry concentration in the US between 1977 and 2013 is driven by a new industrial revolution in three broad non-traded sectors: services, retail, and wholesale. Sectors where national concentration is rising have increased their share of employment, and the expansion is entirely driven by the number of local markets served by firms. Firm employment per market has either increased slightly at the MSA level, or decreased substantially at the county or establishment levels. In industries with increasing concentration, the expansion into more markets is more pronounced for the top 10% firms, but is present for the bottom 90% as well. These trends have not been accompanied by economy-wide concentration.Top U.S. firms are increasingly specialized in sectors with rising industry concentration, but their aggregate employment share has remained roughly stable. We argue that these facts are consistent with the availability of a new set of fixed-cost technologies that enable adopters to produce at lower marginal costs in all markets.We present a simple model of firm size and market entry to describe the menu of new technologies and trace its implications...."

Note attached is a WSJ op-ed on the paper by Hsieh and an excellent blog post reviewing the paper by Geoffrey Manne

Tedeschi responds to the FRBSF paper (that found the LFP has been driven up by a lack of exits we looked at a few weeks ago) which he finds unconvincing, think that that prime age participation still has run room".... conclude that labor force entry likely explains between ⅓ to ½ of the rise in US prime-age participation from its recent trough, and that the level and slope of labor force flows suggest prime-age participation may have even further to rise in the absence of exogenous shocks.....Analyzing the contribution of exit and entry on labor supply changes depends crucially on three dimensions — the frequency of the flows, the adjustments made for consistency and survey error, and the very definition of “labor supply” itself.....The unadjusted month-to-month CPS flows used in Barnichon (2019) suggest that all of the prime-age participation rise — actually, more than all since unadjusted month-to-month labor force entry falls somewhat between October 2015 and May 2019 — was due to falls in labor force exit. A sensitivity analysis that compares the unadjusted month-to-month flows with alternative labor force status flows that correct for possible errors or inconsistencies shows that labor force entry consistently played a role too. These corrections yield results that are qualitatively in line with unadjusted year-on-year flows, as well as flows into and out of a probability-weighted synthetic labor supply measure.Quantitatively, labor force entry contributes between 11% and 73% of the rise in prime-age labor force participation since October 2015 under the ten alternatives I analyzed, with a median across all the metrics of 36%. Moreover, because these flow measures are different from one another, their steady-state troughs each occur at different dates other than October 2015. If I measure entry and exit contributions beginning at each metric’s steady-state trough rather than the same October 2015 start point, the range of alternative entry contributions rises to 23% — 93%, with the median rising to 42%. In other words, when looking at a robust array of measures, prime-age labor force entry rose during the recovery. Based on the range of results a reasonable conclusion is that labor force entry has contributed between ⅓ and ½ of the rise in prime-age labor force participation during the recovery.Flows alone provide no conclusive evidence on potential participation, but for most of these alternative measures, entry is rising but still below pre-crisis peaks, and exit is falling but still above pre-crisis troughs. Both of these behaviors suggest prime-age participation has further to rise in the absence of exogenous shocks...."
Ernie Tedeschi, "Participation and the Hot Labor Market," Employ America, June 21, 2019, https://medium.com/@employamerica/participation-and-the-hot-labor-market-a84ef77a3bb1

Participation and the Hot Labor Market

• The labor force participation rate of prime-age (25-54 year old) Americans has risen almost 2 percentage points since its trough in 2015.
• Labor force growth must be due to some combination of higher labor force entry rates and/or lower labor force exit rates; it is an open question how much each has contributed to the recent rise.
• A recent study published by the Federal Reserve Bank of San Francisco (Barnichon (2019)) attributes virtually all of this rise to lower labor force exit and concludes that the tightening labor market of the last five years has not pulled in new workers.
• By contrast, I show that this finding is not robust when looking at flows that are: 1.) adjusted for internal errors and inconsistencies; 2.) measured year-on-year rather than month-on-month; and, 3.) calculated from broader or narrower measures of labor supply.
• I conclude that labor force entry likely explains between ⅓ to ½ of the rise in US prime-age participation from its recent trough, and that the level and slope of labor force flows suggest prime-age participation may have even further to rise in the absence of exogenous shocks. Introduction

The Great Recession hit the American labor force hard. By 2015 Q3, the share of Americans age 25-54 who were classified as in the labor force — those with or actively looking for a job — had dropped to 80.7%, the lowest prime-age participation rate since 1984.

Some of this decline was due to long-term structural factors: a persistent decline in prime-age male participation, and a plateauing in prime-age female participation after decades of gains.

But in retrospect, we now have strong evidence that much of the decline was due to the soft economy. One tell is how participation has behaved in the tightening labor market of recent years: by 2019 Q1, the prime-age participation rate had risen to 82.6%, almost two percentage points up from its 2015 trough.

A recent Economic Letter from the Federal Reserve Bank of San Franciscomakes the case that the rise since 2015 was almost exclusively because workers began leaving the labor force at smaller and smaller rates (“labor force exit”), not because of new entrants into the workforce (“labor force entry”). This would mean that the tight American labor market hasn’t really pulled workers back into the nation’s labor supply.

The San Francisco Fed analysis based its conclusion solely on month-to-month flows in the Current Population Survey (CPS), but there are two weaknesses to its approach.

First, a growing literature shows evidence of possible internal inconsistencies and errors in CPS labor force status. I find that correcting for these errors meaningfully changes, and under certain assumptions fully reverses, the month-to-month flow story.

Second, the San Francisco Fed analysis did not test the sensitivity of its conclusions against other approaches, either by using year-on-year rather than month-to-month flow calculations, or by looking at different labor force definitions.

Addressing both of these issues, I come to a wholly different conclusion. Bothlabor force exit and entry contributed to the recent participation rise, the latter responsible for between ⅓ and ½ of the rise. Moreover, the level and slope of these metrics’ reads since 2015 are consistent with further rise in prime-age participation in the absence of exogenous shocks.

On bathtubs and participation

Labor supply can expand through two different mechanisms: by workers coming back into the labor force (a rise in labor force entry), or by a stem in the flow of workers leaving the labor force (a fall in labor force exit).

The more formal depiction of this logic is often called a bathtub model, and it’s used in economics, demography, the natural sciences, and engineering.

The upshot of the bathtub model is that the “steady state” level of a metric is the level that would prevail in equilibrium if both entry/inflow and exit/outflow rates remained unchanged over time.

But the math is far less important here than the underlying mechanics. Labor force participation can rise either because the rate of people entering the labor force has risen (in keeping with the bathtub analogy, this is like more water coming out of the faucet) or if the rate of people exiting the labor force falls (like if the drain was clogged), or some combination of the two.

________________________________

A recent FRBSF analysis suggests all of the participation rise stemmed from lower labor force exit.

In a May 2019 Economics Letter, San Francisco Federal Reserve labor economist Regis Barnichon looks at the behavior of entry and exit rates in the CPS and concludes that virtually all of the rise in prime-age participation after 2015 was due to a fall in labor force exit. Based on these estimates, the hot economy had little to no impact on labor force entry rates.

I’ve reproduced the Barnichon (2019) data series below, which is month-to-month and takes CPS labor force status as-is with no adjustments. Both unadjusted labor force entry and exit rates spiked in the wake of the Great Recession, and then fell subsequently. Entry rates flatlined beginning around 2015, while exit rates have generally continued to fall. Barnichon (2019) concludes then that the rise in participation must have been driven entirely by declines in the level of labor force exit.

Barnichon (2019) goes on further to make a slope argument, pointing out that there’s no indication from this month-to-month flow data that the labor force entry rate is poised to rise. He therefore concludes that the US is unlikely to see sustained higher labor force entry in the short- to medium-term.

Note that unadjusted month-to-month labor force exit is falling and may still have further to fall.

However, even on Barnichon’s own terms, a flattening of one mechanism (entry) doesn’t necessarily mean that the level of prime-age labor force participation has peaked. Exit rates, the other mechanism, are falling but remain elevated from their 2007 and 2000 lows. If one concluded that labor force exit had further to fall, then the implication is that prime-age participation by extension still has further to rise.

Under the Barnichon analytical framework, I calculate that if unadjusted month-to-month labor force entry were unchanged but labor force exit continued falling, then the prime-age labor force participation rate would rise another 1 percentage point under the 2007 exit rate, and 2 percentage points under the 2000 exit rate.

This is not dispositive of course — there may be factors preventing entry and exit from reaching pre-crisis levels. But the point is that even using the Barnichon methodology, we may not be at peak prime-age labor force participation yet.

________________________________

The distinction between unemployment and nonparticipation is often inconsistent within the CPS.

Economists should be cautious about drawing strong conclusions solely from unadjusted month-to-month CPS flows, because the distinction between participation and nonparticipation can be inconsistent for some CPS respondents over time.

Determining whether a person in the CPS is entering or exiting the labor force often hinges on the distinction between being “unemployed” — out of work but actively looking for a job — and being a “nonparticipant”: out of work and not actively looking.

Both involve periods of nonwork, but an unemployed person is still in the official labor force, whereas a nonparticipant is not. The CPS includes a series of questions meant to guide and prompt household respondents with these distinctions.

Most respondents to the CPS appear to have consistently-coded labor force status over time. But some labor force status records appear to be susceptible to classification error.

For example, Ahn & Hamilton (2019) notes duration inconsistencies: some people who are nonparticipants in the CPS one month and then switch to being unemployed the next consecutive month also report that the duration of their unemployment spells have been greater than a month. But this is internally inconsistent within the CPS: a person switching statuses into unemployment should report being unemployed no more than one month.

Meanwhile, Abowd and Zellner (1985), Poterba and Summers (1986), and Elsby, Hobijn, and Şahin (2015) also identify cases of spurious transition, where households will move from, say, unemployment to nonparticipation and back again over three consecutive months. In principle this is entirely possible, but in practice these temporary flows may effectively be noise rather than signal.

Once we adjust for these issues, labor force entry becomes a more important factor in month-to-month transitions.

To test the robustness of the Barnichon (2019) approach, then, I make two adjustments to account for possible errors mentioned above.

The first is an adjustment suggested in Elsby et al (2015) which the authors colorfully term “deNUNifying”. For CPS records in the first four or second four monthly rotations, a record that switches from N to U and then back to N again in consecutive months (NUN) is treated as N for all three (NNN); likewise if an N record was U in both the prior and subsequent month (UNU), it’s changed to U for all three months (UUU). The two exceptions are if the record switched between U and N each of the four rotation months (UNUN or NUNU); in those cases, I leave the records as-is since either state is an equally-likely counterfactual.

The second adjustment is one employed in Ahn & Hamilton (2019) and which I uncreatively dub “deNUing” here. If a record transitions from N to U in consecutive months (NU), but then reports an unemployment duration of greater than 5 weeks, I treat the prior month’s N as a U instead (UU).

DeNUNed month-to-month entry contributed 43% of the participation rise since October 2015 and 55% since the deNUNed steady-state trough of October 2013. When the data is also deNUed, the entry contribution rises to 73% and 93% respectively. This is almost the mirror image of the decomposition under the Barnichon (2019) approach.

This is confirmed when we look at year-to-year transitions.

But just as economists should not draw strong conclusions solely from unadjusted month-to-month CPS flows, so too should they be reticent to rely only on adjusted flows, no matter how internally-consistent or transparent those adjustments are. After all, the adjustments here literally involve changing responses to the CPS.

A useful cross-check for the month-to-month flow analysis is to look at year-on-year flows. If a concern with the CPS is that short-term labor status flows are subject to classification errors, then looking at the same flows 12 months apart may provide a more reliable signal of persistent status shifts. Year-on-year flows also have the advantage of being seasonally-aligned, since they compare the same month across two different years.

And in fact, the year-on-year flow data — even the raw unadjusted flows — are consistent with the story told by our deNUNed and deNUed month-to-month flows: the rise in prime-age participation has been driven by bothhigher labor force entry and lower labor force exit. The 12-month flow rates themselves differ in level terms between the unadjusted and adjusted series, but their relative behaviors are all consistent with one another and tell similar stories. The entry contributions for the year-on-year flows range from 33-36% since October 2015 and 38-45% since their July 2014 steady-state troughs.

Economists focus on labor force participation often because it is a proxy for how we think of labor supply — the population of workers who have or are pursuing a job in any given month.

But the participant-versus-nonparticipant distinction isn’t always meaningful when it comes to job-finding. The unemployed are more likely to transition into employment over 12 months than nonparticipants are, but there is a great deal of heterogeneity among different types of nonparticipants. Some nonparticipants, for example, still say they want a job; these nonparticipants often have job-finding rates not dramatically different from the unemployed.

Moreover, more than two-thirds of the prime-age workers hired out of nonemployment were nonparticipants rather than unemployed, and in fact most of these hired nonparticipants had previously said they didn’t want a job.

Economists generally think of employment as a labor demand rather than supply measure, since there are always nonemployed Americans transitioning between jobs or ready and willing to take a job should one arise, margins that in principle a labor supply metric would account for.

But the point here is not to hold up employment as a superior labor supply measure. Rather it’s simply to find a measure that can bookend a sensitivity analysis on the narrow side. For this purpose, using employment is straightforward and comprehensible.

The results further support the conclusion from the adjusted month-to-month flows and the year-on-year flows: the rise in prime-age employment is a function of both job finding and a fall in job exit, though with employment the trough in entry was pre-2015. Month-to-month employment entry contributed 60% of the prime-age employment rise since October 2015 and 36% since the steady-state employment trough in December 2009. Year-on-year results are 21% and 23% respectively.

Another cross-check, which bounds the other side of the analysis, uses a broader measure of labor supply than the official labor force. To try to keep the analysis here similarly straightforward to the one using employment, I define an “augmented prime-age labor force” equal to the official labor force plus prime-age Americans who are nonparticipants but say they want a job, as well as prime-age Americans who say they are out of the labor force due to school enrollment (based on the logic that Americans investing in more education probably intend to reenter the labor force at some point). This alternative definition is about 5 percentage points higher in level terms than the official prime-age labor force participation rate in 2019, a gap that has grown over time.

As a final robustness test, I construct a synthetic labor supply metric. All workers with a job are included 1:1, but I weight those without a job, whether unemployed or nonparticipants, based on the real-time probability that they will find jobs the next month. I use a rolling centered three-year probit model. The model predicts the probability of one-month job finding based on sex, age, parenthood, marriage status, race, nativity, education, size of metro area, and reason for nonemployment (short- or long-term unemployment, discouragement, retirement, enrollment, disability/illness, home/family care, or other). To calculate flows, I randomly assign the nonemployed CPS records to synthetic labor force status based on these binomial probabilities.

In one sense, my synthetic measure is broader than the official labor force in that every margin of nonemployment is given at least some weight, including nonparticipants. But in practice because my synthetic measure no longer weights the unemployed at 100% but at a (usually much) smaller weight, its level is lower than the official prime-age labor force, and so ought not be thought of as a broader labor force measure. It is however more cyclical than the official labor force: the rise since 2015 has been twice that of headline labor force participation.

Conclusions: Putting it All Together

Analyzing the contribution of exit and entry on labor supply changes depends crucially on three dimensions — the frequency of the flows, the adjustments made for consistency and survey error, and the very definition of “labor supply” itself.

The chart below decomposes the change in prime-age participation implied by different metrics. In this particular figure, the starting point is always October 2015 (when the 12M moving average of the prime-age participation rate troughed) and runs through May 2019.

The unadjusted month-to-month CPS flows used in Barnichon (2019) suggest that all of the prime-age participation rise — actually, more than all since unadjusted month-to-month labor force entry falls somewhat between October 2015 and May 2019 — was due to falls in labor force exit.

A sensitivity analysis that compares the unadjusted month-to-month flows with alternative labor force status flows that correct for possible errors or inconsistencies shows that labor force entry consistently played a role too. These corrections yield results that are qualitatively in line with unadjusted year-on-year flows, as well as flows into and out of a probability-weighted synthetic labor supply measure.

Quantitatively, labor force entry contributes between 11% and 73% of the rise in prime-age labor force participation since October 2015 under the ten alternatives I analyzed, with a median across all the metrics of 36%.

Flows alone provide no conclusive evidence on potential participation, but for most of these alternative measures, entry is rising but still below pre-crisis peaks, and exit is falling but still above pre-crisis troughs. Both of these behaviors suggest prime-age participation has further to rise in the absence of exogenous shocks.

________________________________

End Notes

My data here are 12-month moving averages whereas Barnichon (2019) shows calendar year averages; the two series nevertheless appear to be consistent with one another. I weight my flow metrics using a constructed weight that constrains the sum of longitudinally-linked monthly records to be equal to the total population along sex, 5-year age group, and labor force status interactions. Later on in this analysis when I look at year-on-year flows, I use a similarly-constructed weight but limited to records longitudinally-linked across 12 months.

The CPS surveys the same household a total of eight months. The first four months are consecutive. Then the household is off the survey for eight months. And finally the household returns to the CPS for another four consecutive months. All told, sixteen months elapse between a households first and last months in the CPS.

Ahn and Hamilton set the cutoff at >4 weeks of reported duration rather than >5; I choose 5 instead of 4 to conservatively avoid false error-identification in case of seasonal timing variations, since a typical month is greater than 4 weeks long.

"...In Figure 1A, we look specifically at China's payments for the use of U.S. intellectual property. The payments grew significantly, from $755 million in 1999 to $8.3 billion in 2017—more than 11-fold. Even more interesting, as shown in Figure 1B, is that China's royalty payments to the United States (blue line) grew faster than China's GDP (red line)....."

Ana Maria Santacreu and Makenzie Peake, "A Closer Look at China’s Supposed Misappropriation of U.S. Intellectual Property," Federal Reserve Bank of St. Louis, February 8, 2019, https://research.stlouisfed.org/publications/economic-synopses/2019/02/08/a-closer-look-at-chinas-supposed-misappropriation-of-u-s-intellectual-property

Amy Glasmeier, “Living Wage Calculator,” Massachusetts Institute Of Technology, Accessed May 2020, http://livingwage.mit.edu/

About the Living Wage Calculator

INTRODUCTION

Analysts and policy makers often compare income to the federal poverty threshold in order to determine an individual’s ability to live within a certain standard of living. However, poverty thresholds do not account for living costs beyond a very basic food budget. The federal poverty measure does not take into consideration costs like childcare and health care that not only draw from one’s income, but also are determining factors in one’s ability to work and to endure the potential hardships associated with balancing employment and other aspects of everyday life. Further, poverty thresholds do not account for geographic variation in the cost of essential household expenses.

The living wage model is an alternative measure of basic needs. It is a market-based approach that draws upon geographically specific expenditure data related to a family’s likely minimum food, childcare, health insurance, housing, transportation, and other basic necessities (e.g. clothing, personal care items, etc.) costs. The living wage draws on these cost elements and the rough effects of income and payroll taxes to determine the minimum employment earnings necessary to meet a family’s basic needs while also maintaining self-sufficiency.

The living wage model generates a cost of living estimate that exceeds the federal poverty thresholds. As calculated, the living wage estimate accounts for the basic needs of a family. The living wage model does not include funds that cover what many may consider as necessities enjoyed by many Americans. The tool does not include funds for pre-prepared meals or those eaten in restaurants. We do not add funds for entertainment, nor do we incorporate leisure time for unpaid vacations or holidays. Lastly, the calculated living wage does not provide a financial means to enable savings and investment or for the purchase of capital assets (e.g., provisions for retirement or home purchases). The living wage is the minimum income standard that, if met, draws a very fine line between the financial independence of the working poor and the need to seek out public assistance or suffer consistent and severe housing and food insecurity. In light of this fact, the living wage is perhaps better defined as a minimum subsistence wage for persons living in the United States.

FAMILY COMPOSITIONS

The living wage calculator estimates the living wage needed to support families of twelve different compositions: one adult families with 0, 1, 2, or 3 dependent children, two adult families where both adults are in the work force with 0, 1, 2, or 3 dependent children, and two adult families where one adult is not in the work force with 0, 1, 2, or 3 dependent children.

For single adult families, the adult is assumed to be employed full-time. For two adult families where both adults are in the labor force, both adults are assumed to be employed full-time. For two adult families where one adult is not in the labor force, one of the adults is assumed to be employed full-time while the other non-wage-earning adult provides full-time childcare for the family’s children. Full-time work is assumed to be year-round, 40 hours per week for 52 weeks, per adult.

Families with one child are assumed to have a ‘young child’ (4 years old). Families with two children are assumed to have a ‘young child’ and a ‘child’ (9 years old). Families with three children are assumed to have a ‘young child’, a ‘child’, and a ‘teenager’ (15 years old).

TECHNICAL DOCUMENTATION

For a more detailed description of the methodology used, please see the technical documentation here.

“…To see what might happen to restaurant jobs in states like Alabama if the minimum wage increases to $15 an hour and if the tip credit is eliminated, the chart above shows shows some of the fallout in New York City for jobs at full-service restaurants. The significant loss of food jobs in the Big Apple over the last few years are partly at least, if not mostly, a direct result of the negative impact of the $15 an hour minimum wage law that went into effect on December 31, 2018. The increase to $15 an hour at the end of last year followed three annual $2 an hour increases starting at the end of 2016 that increased the city’s minimum wage by 67% from $9 an hour in 2016 to $15 an hour at the end of last year….”

Mark Perry, "Minimum wage increases could devastate restaurants and food jobs. NYC may be the ‘canary in the coal mine’," American Enterprise Institute, June 22, 2019, https://www.aei.org/carpe-diem/minimum-wage-increases-could-devastate-restaurants-and-food-jobs-nyc-may-be-the-canary-in-the-coal-mine/

Minimum wage increases could devastate restaurants and food jobs. NYC may be the ‘canary in the coal mine

Want to earn a living wage in Alabama? Try working in a full-service restaurant. I’m a restaurant franchisee with nine locations throughout Alabama and Florida, and my tipped employees report an hourly wage between $18 and $28, inclusive of base pay and gratuities. That’s more than the living wage for a single adult in Alabama, as calculated by the Massachusetts Institute of Technology.

Unfortunately, some well-meaning members of Congress are trying to pass a minimum-wage bill that would destroy my business model. Tipped employees are currently paid a lower base wage and are legally guaranteed to earn at least the minimum wage with tips included. My employees average roughly three times the relevant minimum wage when tips are accounted for.

But Rep. Terri Sewell (D., Ala.) has introduced one of several bills that would upend this system. Under her bill, employees would have to earn at least the minimum wage before tips. A handful of states—Alaska, California, Minnesota, Montana, Nevada, Oregon and Washington—have adopted this approach. Were it to take effect in Alabama, it would represent as much as a 600% increase in the hourly cost of paying a tipped employee to work for me. In pursuit of a “living wage,” Ms. Sewell could cause my employees to lose their incomes entirely by forcing me to shut down my business.

Despite the widespread popularity of restaurants the public has surprisingly little understanding of how the business works. At my restaurants we sell a six-piece plate of chicken wings for $10. Customers typically assume we keep $6 or $7 in profit from that sale. In fact, our profit on that six-piece is closer to 50 cents, after accounting for food and labor costs, rent, overhead and other expenses.

My employees actually make more money from a sale than I do. I earn 5% on each customer check, while they earn 20% or more of the check in tip income. If that doesn’t seem shocking to you, consider this: labor costs are about a third of a restaurant’s budget. Now, imagine if a household expense that takes up a third of your income—maybe a mortgage payment or rent—increased by 600%. You’d probably be panicking. So am I.

MP: To see what might happen to restaurant jobs in states like Alabama if the minimum wage increases to $15 an hour and if the tip credit is eliminated, the chart above shows shows some of the fallout in New York City for jobs at full-service restaurants. The significant loss of food jobs in the Big Apple over the last few years are partly at least, if not mostly, a direct result of the negative impact of the $15 an hour minimum wage law that went into effect on December 31, 2018. The increase to $15 an hour at the end of last year followed three annual $2 an hour increases starting at the end of 2016 that increased the city’s minimum wage by 67% from $9 an hour in 2016 to $15 an hour at the end of last year.

Based on employment data released yesterday by the Bureau of Labor Statistics, jobs at New York City full-service restaurants fell by 4,000 from last May, or by 11 jobs lost every day over the last year. The job losses in New York City over the last few years have been much worse than during the Great Recession, and even slightly worse than the loss of food service jobs during the 2001 recession and the devastating effects of the 9/11 terrorist attacks that year. New York City could be the “canary in the coal mine” that predicts what happens when politicians force significantly higher labor costs on restaurant owners that put restaurant jobs at great risk as well as jeopardizing the financial profitability and long term viability of restaurants.

We hear a lot about the living wages of unskilled, low skilled and limit-experienced workers, but what about the “living profitability” of small business owners and restaurateurs like Mr. Fox who may not survive a 600% increase in their labor costs?

new John Fernald (FRBSF) on the impact of demographics and productivity on GDP growth

"....The solid growth in GDP over the past year naturally raises the question of whether we have been understating the potential trend. Nevertheless, it is far too early to declare that trend growth is higher than current estimates of 1.6-1.7%. The most natural interpretation of the strength in hours worked is that any structural element represents an effect on the levels of employment, hours, and potential output—perhaps in response to the TCJA—not a persistent growth effect.Policy changes such as family-friendly policies common in other OECD economies might lead to trend increases in labor force participation (Daly et al. 2018). But such policies are unlikely to be implemented soon.....slow pace relative to history reflects two factors. Most importantly, demographic trends leave little doubt that the labor force will grow slowly at best over the next decade. There is more uncertainty about the other factor, productivity growth. Our forecast assumes that productivity will grow at a pace similar to recent decades. Averting this slow-growth future would require a sustained increase in productivity growth..... Demographics are the key reason future trend growth is expected to be low relative to history. Figure 1 shows that labor force growth has slowed markedly since the 1970s, when baby boomers reached working age and female participation in the labor force rose rapidly. However, stagnating growth in female participation since the 1990s combined with projections for birth rates, immigration, and mortality suggest growth of the traditional working-age population ages 16-64 is slowing to a historically low pace.....Figure 2 shows growth of real GDP per hour, a broad measure of labor productivity, broken into periods since 1973 that reflect variation in productivity growth. Figure 2 depicts how productivity growth has shifted between normal and exceptional periods (Fernald 2016). Unusually influential innovations—such as the steam engine, electric dynamo, internal combustion engine, and microprocessor—typically led to a host of complementary innovations that boosted productivity growth broadly for a time.....This measure of productivity growth has slowed since 2004. The green dashed line in Figure 2 is productivity growth averaged over 2004-2018, which is very close to the 1973-1995 average pace. During the fast-growth period from 1995 to 2004, productivity growth averaged 2½%. During the slower periods of 1973-1995 and 2004-2018, growth averaged only 1% to 1¼%, and that pace dropped dramatically lower most recently in 2010-2018....The black dot in the figure shows 2018 growth in real GDP per hour of 1.1%. This is higher than the average growth rate over 2010-2018 and may suggest a pickup in productivity growth. However, in our view, it is best to view the years since 2004 as a whole; in this case, the 2018 pace is close to the period average. We focus on the average since 2004 because labor productivity has some persistent cyclical dynamics. The large shock of the Great Recession caused a temporary boost in productivity growth from 2007 to 2010 that has since unwound. For one thing, less-skilled workers were more likely to lose jobs in the Great Recession than more-skilled workers, who typically have higher productivity. For another, businesses came out of the Great Recession with ample productive capacity but limited demand. Over time, increases in capacity per worker are an important source of productivity growth, but for a time after the recession, businesses could meet the recovery in demand without investing. Hence, that capacity channel was temporarily dialed down. Once we average over the transitory ups and downs, it appears that we have been in a slow growth regime for the past 15 years. Regimes generally appear persistent—lasting a decade or longer. For this reason, our most likely or “modal” estimate is that the slow-growth regime will continue for at least the next few years.That said, productivity is highly uncertain in both a statistical and an economic sense. Under the regime view of productivity growth, much of the statistical uncertainty is about which regime we will be in. Neither economists nor statisticians have a good track record of forecasting changes in trend productivity growth. In addition, even within a regime, productivity is inherently volatile from year to year. Economically, major technological gains could remain narrowly focused, leaving modest and incremental productivity growth as the norm for the broader economy, as we assume in our benchmark. Or, we might see another broad-based wave of the IT revolution. Alternatively, future growth might look substantially different from the past, reflecting the innovative contributions of robots and machine learning. Despite this inherent uncertainty, our best guess is that productivity growth over the next five to six years will be in line with previous slow regimes. During those regimes, GDP per hour rose around 1-1¼% per year...."

John Fernald and Huiyu Li, "Is Slow Still the New Normal for GDP Growth?," Federal Reserve Bank of San Francisco, June 24, 2019, https://www.frbsf.org/economic-research/publications/economic-letter/2019/june/is-slow-still-new-normal-for-gdp-growth/

"....In the years since the Great Recession, social scientists have anticipated that economic recovery in the United States, characterized by gains in employment and median household income, would augur a reversal of declining fertility trends. However, the expected post-recession rebound in fertility rates has yet to materialize....If fertility responses to economic recessions are in fact cyclical, then economic recovery and expansion should predict increased fertility rates in the same way that economic recessions predict decreased fertility rates. However, fertility rates have yet to return to pre-recession levels. Cherlin and colleagues’ (2013) analysis indicated an 11 percent decrease in TFR between 2007 and 2011 - a drop from a 2.1 to 1.9. By 2013, six years after the onset of the Great Recession, the national TFR remained below the level of population replacement (Mather 2014). Only in 2014 did the TFR briefly reverse course and increase before decreasing yet more in 2015 and 2016 (Hamilton et al. 2017; Hamilton and Kirmeyer 2017). This ongoing fertility decline in the U.S. is at odds with slow, yet steady improvements in economic conditions, including unemployment and household income. Nationally, household income returned to pre-recession levels in 2015 while the unemployment rate returned to pre-recession levels in 2016....In this study, I propose an economic explanation for why fertility rates have continued to decline regardless of improvements in conventional economic indicators. I argue that ongoing structural changes in U.S. labor markets have prolonged the financial uncertainty that leads women and couples to delay or forgo childbearing. Combining statistical and survey data with restricted-use vital registration records, I examine how cyclical and structural changes in metropolitan-area labor markets were associated with changes in total fertility rates (TFRs) across racial/ethnic groups from the early 1990s to the present day, with a particular focus on the 2006-2014 period. The findings suggest that changes in industry composition—specifically, the loss of manufacturing and other goods-producing businesses—have a larger effect on TFRs than changes in the unemployment rate for all racial/ethnic groups. Because structural changes in labor markets are more likely to be sustained over time—in contrast to unemployment rates, which fluctuate with economic cycles—further reductions in unemployment are unlikely to reverse declining fertility trends....."

Nathan Seltzer, "Beyond the Great Recession: Labor Market Polarization and Ongoing Fertility Decline in the United States,"Demography, June 18, 2019, https://link.springer.com/article/10.1007/s13524-019-00790-6