Category: Labor Markets

Across races, women bolster family economic security

Posted on by David Hudson
Between 1979 and 2013, women of all races saw an increase in their work hours. On average, however, women from middle-class African American and Latino families worked more hours annually than women from white families.

Overview

Over the past half-century, the rise in women’s employment and earnings in the United States have boosted family incomes up and down the income ladder. Women’s increased time spent in paid employment means that families have lost time inside the home for caregiving, leaving many struggling to cope with the competing demands of work and life. Despite the time squeeze at home, however, women’s additional earnings have not meant that family incomes have increased faster than in earlier eras.

Changes in how women spend their time and its effect on family earnings look different based on where a family sits in the income distribution. Over the period from 1979 to 2013, women’s additional earnings made up for men’s declining earnings within middle-class families. Among low-income families, while women’s earnings helped, they were not enough to completely offset lower men’s earnings. The view is different at the top, where women’s earnings helped pull professional families to an even higher standard of living.

The United States is a nation where income trends also differ markedly across racial ethnic groups; families of color have lower incomes, on average, than white families. This issue brief explores the role that women’s added hours and earnings play in families across income and race and ethnicity.

Download File
Across races, women bolster family economic security (pdf)

Read the full pdf in your browser

Using data from the Current Population Survey, this brief documents how family incomes changed between 1979 and 2013 for low-income, middle-class, and professional white, African American, and Latino families, decomposing the differences in male earnings, female earnings both from greater pay and more hours worked, and other sources of income over this time period. This analysis is an extension of the research presented in the forthcoming book, “Finding Time: The Economics of Work-Life Conflict,” authored by one of the co-authors of this issue brief, Heather Boushey.

Looking over the period from 1979 to 2013, our key findings are:

  • Women from African American and Latino families work more hours, on average, than women from white families.
  • While income grew at about the same pace for white, African American, and Latino families, actual levels of income are much smaller for families of color.
  • Among low-income and middle-class families, the fall in men’s earnings meant that women’s added contributions were vital to ensuring positive—or less negative—changes in income across all racial and ethnic groups.
  • Among professional families, women’s additional work hours improved family income more within white families than within African American and Latino families.

Introduction

Between 1979 and 2013, were it not for women’s rising work hours and earnings, low-income and middle-class families in the United States would have seen sharp drops in household income. Across families up and down the income ladder, women’s increased pay and work hours made the positive difference that American families needed in the wake of increasing instability and stagnant wage growth.

Though women have been vital for the economic success of their families, our analysis is also a reminder that soaring levels of income inequality in the United States—which may be further exacerbated by macroeconomic trends—have particularly harsh consequences for low-income and middle-income families. On average, between 1979 and 2013, low-income families lost income, middle-class families saw their income stall, and professional families experienced a nearly 50 percent income gain.

While we know that families of color—specifically African Americans and Latinos—face ongoing discrimination in the labor market and have lower earnings than their white counterparts, this issue brief explores whether these trends in the importance of women’s added work hours and earnings look different across families of different races or ethnicities.

This issue brief uses data from the Current Population Survey to calculate how incomes changed between 1979 and 2013 for low-income, middle-class, and professional families of different races in the United States. We decompose these changes into differences in male earnings, female earnings from more pay, female earnings from more hours worked, and other sources of income to find how much of a difference the transformations in female labor force participation have made to family incomes across income groups and races. As noted above, this analysis is an extension and update of the analysis presented in “Finding Time,” which covered the period from 1979 to 2012.

We find that while white, African American, and Latino families have all seen income growth at about the same pace, the changes in the actual levels of income are much smaller for families of color. Women’s earnings have made all the difference for low-income and middle-class families of all races, as men’s earnings fell. For professional families, however, women’s increased earnings from additional work hours has a much smaller positive role in the change in income for both African American and Latino families than for white families. Much of this disparity is driven by the fact that women’s earnings have traditionally played a larger role in total family income for African American families as compared to white families.

Defining income groups and family race

This analysis follows the same methodology presented in “Finding Time.” For ease of composition, we use the term “family” throughout the brief, even though the analysis is done at the household level. We split households in our sample into three income groups. Low-income households are those in the bottom third of the income distribution, earning less than $25,440 per year in 2015 dollars. Professional households are those in the top fifth of the income distribution who have at least one household member with a college degree or higher; these households have an income of $71,158 or higher in 2015 dollars. Everyone else falls in the middle-class category.

It’s more complex to define what race or ethnicity a family should be categorized as, especially considering the growing share of multiracial and multiethnic households in the United States. In this brief, we categorize a family as white, African American, or Latino based on the race and ethnicity of the person identified by the Census as the “household head.” We recognize this is an approximation and ideally the categories would be more inclusive. Breaking the category down smaller, however, does not give us a large-enough sample size for our analysis.

Table 1 breaks down the distribution of racial and ethnic groups within each income group. Low-income families have a higher share of African American and Latino families, compared to middle-class or professional families.

Table 1

Setting some context

Before decomposing the changes in income, we will first set the broad context for how family incomes and women’s hours have changed across low-income, middle-class, and professional white, African American, and Latino families between 1979 and 2013.

How did family income change for different races across income groups?

Between 1979 and 2013, across racial and ethnic groups, incomes for low-income families in the United States were either stuck or falling. Low-income white and African American families saw a fall in their income—4.5 percent and 3.4 percent, respectively. Latinos saw a modest increase in family income from $25,499 in 1979 to $25,958 in 2013, but these gains are fairly small—only a 1.8 percent increase over 34 years. (See Figure 1.)

Figure 1

 

Middle-class families have seen parallel trends across all racial and ethnic groups. Between 1979 and 2013, incomes for white, African American, and Latino families grew by 12.9 percent, 12.0 percent, and 14.1 percent, respectively. (See Figure 2.) While all three groups saw modest gains, however, this growth did not close the income gap across racial and ethnic groups. In both 1979 and 2013, the average African American family had an income of nearly $10,000 less than the average white family. While Latino families also have a smaller family income than white families across the time period, the margin of difference (roughly $2,000) is much smaller.

Figure 2

 

The biggest gains in income have been for professional families. Between 1979 and 2013, professional families saw substantial increases in family income across all three racial and ethnic groups. (See Figure 3.) White and African American families experienced similar rates of growth in their incomes (around 50.7 percent), while Latino families saw much slower growth (34.8 percent). As of 2013, Latino professional families had incomes closer to those of African American families compared to white families—a significant shift from 1979.

Figure 3

 

How did women’s working hours change for different races across income groups?

Between 1979 and 2013, low-income women from white, African American, Latino families put in more hours in paid employment. (See Figure 4.) Over this time period, women from white and African American families had similar increases in work hours. This meant that the employment gap between the two groups remained fairly steady: As of 2013, women in low-income African American families continued to work more hours than white women from low-income families.

Also noteworthy is the sharp increase in work hours for women from low-income Latino families over this time period. In 1979, these women worked an average of 584 hours per year—the lowest number among the three racial and ethnic groups—but by 2013, that number had risen by 50 percent, up to 874 hours. They now put in longer hours than either white or African American women in low-income families. This could be due to a variety of factors and was likely affected by the changing patterns in immigration and shifts in gender roles as these women acclimated to the U.S. workforce.

Figure 4

 

Between 1979 and 2013, women from middle-class households across all three racial and ethnic groups also increased their hours. (See Figure 5.) In both 1979 and 2013, within middle-class families, women from white families put in fewer hours than those from African American or Latino families. While the gap closed over time and women from white homes saw the largest percent increase in their work hours, the longest work hours among women in middle-class families in 2013 still came from African American families, followed by Latinos, then whites.

Figure 5

 

The trends in women’s hours of employment differ more across race and ethnicity for professional families than middle-class and low-income families. (See Figure 6.) Among professional families, between 1979 and 2013, women from white families saw the greatest rise in their work hours (32 percent). In 1979, they were putting in an average of 1,315 hours per year (25 hours per week), compared to 1,738 hours per year (33 hours per week) in 2013.

Yet women in professional white families still put in fewer work hours than women in professional African American or Latino families. Women from African American homes saw much slower rates of growth but still increased their work hours by close to 10 percent over the time period. Women from Latino homes experienced the smallest change (4.0 percent), raising their annual hours worked only by 69 hours.

Figure 6

 

Decomposing the changes in family income across income and race groups

We’ve seen that between 1979 and 2013, across all income groups and racial and ethnic groups, women increased their working hours. Women from African American and Latino families are working more hours on average than those from white families. Incomes have risen modestly for middle-class families across race and ethnic groups. In low-income families, incomes have fallen for African American and white families, while increasing marginally for Latinos. Professional families have seen their incomes rise across racial and ethnic groups, although white families continue to have the highest incomes.

Based on these broad trends, our question is: How did the changes in women’s work hours contribute to changes in family incomes across white, African American, and Latino families up and down the income ladder?

To unpack this, we decompose the changes in average family income into male earnings, female earnings, and income from Social Security, pensions, or any other non-employment-related source for each income group and racial and ethnic group. We specifically divide female earnings into actual wage differences between 1979 and 2013. We then use a calculation that finds the difference between 2013 female earnings and a counterfactual that holds 2013 hourly wages constant and multiplies them by the annual hours women worked in 1979 to arrive at the female earnings from additional hours.

As shown in the following breakdowns by income groups, the trends between white, African American, and Latino families are similar: On the whole, women’s increased earnings from both higher pay and additional hours worked has been a large positive factor in keeping family incomes steady. For professional African American or Latino families, women working additional hours has made a much smaller positive difference in boosting family income, undoubtedly because they started with higher average hours.

Low-income families

For white, African American, and Latino low-income households, female earnings from additional work hours shielded families from large decreases in income. (See Figure 7.)

Figure 7

 

Women’s earnings from working more hours and earning more per hour boosted family income and were the only source of positive gains in African American and Latino families. For white families, women’s earnings from more hours contributed an average of $1,060 per year, and women’s earnings from more pay secured an extra $318. African American families saw much smaller changes in income over the time period, but here, too, women’s increased hours added $795 and women’s higher pay added $849.

Similarly, for low-income Latino families, women’s earnings were the only reason they experienced any income growth between 1979 and 2013. In these families, women’s added hours were especially important: Within Latino families, women’s increased work hours added an average of $2,925 and their higher pay added $670 to annual income.

It is notable that within low-income families across racial and ethnic groups, men’s earnings dragged down family income. Between 1979 and 2013, men in low-income families pulled down family income the most in white families, reducing overall family income by $3,446. This was larger than in Latino and African American families where declines in men’s earnings pulled down income by $1,624 and $1,825, respectively.

It’s also notable that between 1979 and 2013, African American and Latino families saw a decline in “other income.” This category includes any unearned income, such as interest payments or benefits, like Social Security or other income supports. While our data does not allow us to dig into the details, it is possible that this decline stems from the loss of income supports due to change in federal welfare policy in the mid-1990s.

Middle-class families

As in low-income families, across all three race groups for middle-class families, women’s earnings—both from more hours and more pay—and other sources of income have been the only components to increase family income. (See Figure 8.)

Between 1979 and 2013, women’s earnings and added hours accounted for the largest share of the gains for all three racial and ethnic groups. Women in white and Latino families added the most to income through their increased working hours ($5,750 and $4,884, respectively, in comparison to $4,880 and $4,425 from increased pay), while African American women’s earnings from increased work hours contributed much less to changes in family income than earnings from increased pay ($2,801 versus $5,925).

Figure 8

 

Like in low-income families, between 1979 and 2013, middle-class men’s average annual earnings fell across racial and ethnic groups. Without the added earnings of women, middle-class families of all races would have seen their incomes fall, all else being equal. The rising hours of paid employment among middle-class women documented in Figure 5 was a key factor in sustaining economic security for all kinds of families.

Professional families

While women’s earnings from working more hours have played a relatively large, positive part in supporting family incomes for both low-income and middle-class families across all three racial and ethnic groups, we document two different trends for professional families: Men’s earnings have boosted family incomes across all three racial and ethnic groups, and non-white professional family income hasn’t benefited as much from an increase in female labor force participation. (See Figure 9.)

Figure 9

 

Between 1979 and 2013, among professional families across racial and ethnic groups, the combination of women’s added hours of work and more earnings per hour comprised the largest share of family income change. Within all these families, men’s earnings have also added to family income over time, unlike low-income or middle-class families.

When we decompose women’s earnings over this time period, women’s earnings from additional hours made up a larger portion of the change ($15,128, or 22 percent) in white families. African American and Latino families saw smaller gains from women’s increased work hours. African American women’s earnings from more work hours added $6,178 (or 11 percent of the total change) to family income, which was less than the changes brought in by other sources of income. And Latino women’s earnings from more work hours only added $2,281 (or 5 percent of the total change) to family income. This is consistent with Figure 6 where we showed a sharp rise in employment for women in white professional families and a fairly small increase for women in Latino families. It is notable, however, that that the gains from higher pay (as opposed to higher hours) are about the same across racial and ethnic groups.

Conclusion

Between 1979 and 2013, across income groups and across race and ethnicity, women’s increased working hours helped stabilize the family incomes of American families, though at varying degrees. Among low-income and middle-class families, women’s contributions were vital to ensuring positive—or less-negative—changes in incomes for white, African American, and Latino families. For professionals, women’s additional hours have made a larger difference for white families than for African Americans and Latino families, but women’s higher pay has improved incomes in all kinds of professional families.

What is clear is that all families have lost time. As women increasingly spend their days employed outside the home, we need to consider the variety of ways in which our labor market and social policies do—and often do not—support families. As one moves down the income ladder, workers are increasingly less likely to have access to employer-provided policies that help them address conflicts between work and life. Workers of color have additional barriers in accessing these policies, such as the ability to have a flexible schedule, controlling for a variety of factors.

As we consider solutions to help workers address conflict between work and family life, policymakers need to bear in mind the increase in average annual hours of work for women in families of color and the important role women’s earnings play within these families. We need policies to address the times when a worker needs to be at home to provide care for a few days or weeks; policies that ensure workers have access to predictable schedules and, ideally, some flexibility that works for employees and employers alike; policies that address the need for high-quality, affordable care for children and, increasingly, elders. And we also need to ensure that these policies are fairly implemented.

But it is not enough to just create more inclusive caregiving policies for workers of color. While generally, across race groups, women have made the difference for their families’ income, there’s also another important paradox at play here: Women from families of color have worked more hours on average than women from white families, yet their families still don’t enjoy the same incomes as white families, especially in the middle or at the top. To truly help women of color establish economic security for their families, we also have to endorse policies that help raise their wages.

When thinking about how to address the time squeeze for all families, caregiving and wage policies that are crafted to be equitable—especially across income and race groups—will go a long way.

Heather Boushey is the Executive Director and Chief Economist at the Washington Center for Equitable Growth and the author of the forthcoming book from Harvard University Press, “Finding Time: The Economics of Work-Life Conflict.” Kavya Vaghul is a Research Analyst at Equitable Growth.

Acknowledgements

The authors would like to thank John Schmitt, Ben Zipperer, Dave Evans, David Hudson, and Bridget Ansel. All errors are, of course, ours alone.

Methodology

The methodology used for this issue brief is identical to that detailed in the Appendix to Heather Boushey’s “Finding Time: The Economics of Work-Life Conflict.”

In this issue brief, we use the Center for Economic and Policy Research extracts of the Current Population Survey Annual Social and Economic Supplement for survey years 1980 and 2014 (calendar years 1979 and 2013). The CPS provides data on income, earnings from employment, hours, and educational attainment. All dollar values are reported in 2015 dollars, adjusted for inflation using the Consumer Price Index Research Series available from the U.S. Bureau of Labor Statistics. Because the Consumer Price Index Research Series only includes indices through 2014, we used the rate of increase between 2014 and 2015 in the Consumer Price Index for all urban consumers from the Bureau of Labor Statistics to scale up the Research Series’ 2014 index value to a reasonable 2015 index estimate. We then used this 2015 index value to adjust all results presented.

For ease of composition, throughout this brief we use the term “family,” even though the analysis is done at the household level. According to the U.S. Census Bureau, in 2014, two-thirds of households were made up of families, defined as at least one person related to the head of household by birth, marriage, or adoption.

We divide our sample into three income groups—low-income, middle-class, and professional households—using the the definitions outlined in “Finding Time.” For calendar year 2013, the last year for which we have data at the time of this analysis, we categorized the income groups as follows:

  • Low-income households are those in the bottom third of the size-adjusted household income distribution. These households had an income of below $25,440 (as compared to $25,242 and below for 2012). In 1979, 28.3 percent of all households were low-income, increasing to 29.7 percent in 2013. These percentages are slightly lower than one third because the cut-off for low-income households is based on household income data that includes persons of all ages, while our analysis is limited to households with at least one person between the ages of 16 and 64. The working-age population (16 to 64) typically has higher incomes than older workers, and as a result, the working-age population has somewhat fewer households that fall into this low-income category.
  • Professionals are those households that are in the top quintile of the and size-adjusted household income distribution and have at least one member who holds a college degree or higher. In 2013, professional households had an income of $71,158 or higher (as compared to $70,643 or higher in 2012). In 1979, 10.2 percent of households were considered professional, and by 2013, this share had grown to 16.8 percent.
  • Everyone else falls in the middle-class category. For this group, the household income ranges from $25,440 to $71,158 in 2013 (as compared to $25,242 to $70,643 in 2012); the upper threshold, however, may be higher for those households without a college graduate but with a member who has an extremely high-paying job. This explains why within the middle-income group, the share of households exceeds 50 percent: The share of middle-income households declined from 62 percent in 1979 to 53.4 percent in 2013.

Note that all cut-offs above are displayed in 2015 dollars, using the inflation adjustment method presented earlier.

In our analysis, we limit the universe to persons with non-missing, positive income of any type. This means that even if a person does not have earnings from some form of employment but does receive income from Social Security, pensions, or any other source recorded by the CPS, they are included in our analysis.

These data are decomposed into income changes between 1979 and 2013 for low-income, middle-class, and professional families. The actual household income decomposition uses a simple shift-share analysis to find the differences in earnings between 1979 and 2013 and calculate the extra earnings due to increased hours worked by women.

To do this, we first calculate the male, female, and other earnings by the three income categories. To calculate the sex-specific earnings per household, we sum the income from wages and income from self-employment for men and women, respectively. The amount for other earnings is derived by subtracting the male and female earnings from total household earnings. We average the household, male, female, and other earnings by each income group for 1979 and 2013, and take the differences between the two years to show the raw changes in earnings by each income group.

To find the change in hours, for each year by household, we sum the total hours worked by men and women. We average these per-household male and female hours, by year, for each of the three income groups.

Finally, we calculate the counterfactual earnings of women. We use the 2013 earnings per hour for women and multiply it by the 1979 hours worked by women. Finally, we subtract this counterfactual earnings from the female earnings in 2013, arriving at the female earnings due to additional hours.

We repeated this analysis for families of different races, specifically white, African American, and Latino families. Defining what race or ethnicity a family should be categorized as more complex, especially as the share of multiracial and multiethnic households is growing. In this brief, we categorized a family as white, African American, or Latino based on the race and ethnicity of the person identified by the Census as the “household head.” In the CPS, this is coded as the first person observation within each household grouping. We recognize this is an approximation and, ideally, the categories would be more inclusive. However, breaking the category down smaller does not give us enough of a sample size for our analysis.

One important point to note is that because of the nature of this shift-share analysis, the averages don’t exactly tally up to the raw data. Therefore, when presenting average income, we use the sum of the decomposed parts of income. While economists typically show median income, for ease of composition and the constraints of the decomposition analysis, we show the averages so that the data are consistent across figures. Another important note is that we make no adjustments for changes over time in topcoding of income, which likely has the effect of exaggerating the increase in professional families’ income relative to the other two income groups.

Equitable Growth in conversation: An interview with Byron Auguste

Posted on by Nick Bunker

“Equitable Growth in Conversation” is a recurring series where we talk with economists and other social scientists to help us better understand whether and how economic inequality affects economic growth and stability.

In this latest installment, Heather Boushey, Executive Director and Chief Economist here at Equitable Growth, talks with Byron Auguste, Managing Director of Opportunity@Work. The two discuss the current problems with the labor market, how these problems may be mostly on the demand side, and how we might “rewire” the labor market.

Read their conversation below.

Heather Boushey: Byron, thank you so much for talking with us. The big topic that I want to focus on with you is about the demand-side problems when it comes to opportunity in the labor market. When we’re thinking about policy, we think a lot about the supply-side problems. And I know that you’ve spent some time thinking about the demand side and I’m eager to learn more from you.

Since the end of the Great Recession, the number of job openings in the United States has increased much quicker than the number of hires. And many economists have been interpreting that as a sign of supply-side problems—that workers don’t have the skills that employers are looking for. But you actually see this as a demand-side, or employer-side, problem. Can you tell us why?

Byron Auguste: The first step to really understanding what’s going on in the labor market is to think about it as a market, take it seriously as a market, understand its market characteristics, the information that’s available to the different actors, the incentives that they face, and then to look at their actual behavior and how it’s changed over time.

Over the past 30 years, there’s been an increasing sense that somehow there are these mismatches in the labor market. We have a situation where in 2015, open jobs that employers were trying to fill in the United States were at a record high, while at the same time, labor force participation among working-age adults was at a 40-year low.

A number of commentators and particularly businesspeople talk about this as a skills gap, which gives the impression that if only people were getting the right skills or more skills there would be no problem. But if you look at the labor market, both the changes in the demand side and the supply side, it’s really striking how much the supply side—that is to say, education and training—has been relatively stable, whereas the behavior of the demand side—how employers hire and fire, who and how they train, and just everything about their HR behavior—has changed dramatically in the last 30 years.

If you’re looking for the main reason to change the labor market, all the data and the stylized facts should lead you to look at the demand side first. And if you look at the demand side first, you see some really striking things.

First of all, you see that employers have significantly changed their model with respect to hiring and training. Thirty years ago, maybe half of hiring was in some sense entry-level hiring—hiring people out of high school, out of college, out of Ph.D. programs, whatever it was. The expectation was that the companies would train the new hires and they would learn generalized work skills, as well as the specific skills that companies need on the job.

But if you look at 2014, although we don’t have great data on this, Wharton professor Peter Cappelli has noted that entry-level hiring accounted for just 6 percent of all new hires in 2012—much lower—and that much more of hiring now is for experienced workers with very specific education and experience profiles.

When you’re looking at entry level hiring, it’s much easier to find poaching from other companies. Then on top of that, when employers look for these sorts of specific profiles, they tend to characterize them, like in a job description, in terms of their specific education and employment history.

The requirements for a four-year college degree, in particular, are rising dramatically. Burning Glass, a data analytics company in the labor market, showed that only about 20 percent of administrative assistants in this country have college degrees, but that two-thirds of the new job postings for administrative assistants require a bachelor’s degree. In other words, about 80 percent of existing administrative assistants can’t apply for two-thirds of the new jobs in their own field.

If you think about it, the way this “credential creep”  moved through many parts of the labor market, it might connect the dots to the fact that we have low voluntary quits. Voluntary job mobility is down by 23 percent since 2001. And of course that’s partly cyclical, but it’s much lower now than at a similar point in past cycles. And it’s because a lot of people are stuck. They can’t move, based on the way that employers have increasingly depended on these kind of backward-looking heuristics. That’s an example of the demand-side behavior that really stops people from making progress.

In addition, when you look at the demand behavior over the business cycle by employers, it’s a really big difference. In the first five recessions after World War II, U.S. employers only laid off about one-third of the workers they would have needed to lay off to fully offset the drop in demand for their products or services. In other words, two-thirds of their workforce, in the aggregate, was absorbed by what economists would call labor hoarding, but what a CEO would now call “missing quarterly earnings targets,” right? This happens so consistently in the post-World War II period that economists called it Okun’s Law [after macro economist Arthur Okun].

But employers don’t do [much of] that anymore; their behavior started to change in the early ’80s. It went to 50/50 in that recession in the early ’80s, then to 25/75—the other way—in the early ’90s. And in the last two recessions, U.S. employers laid off 100 percent of the workforce that was needed to offset the demand drop. In other words, there was no more labor hoarding, profits were maintained, and layoffs for workers absorbed the entire demand drop.

Okun’s “Law” has been repealed, but only in the United States. In the United Kingdom, employers are still laying off one-third, two-thirds; Okun’s Law is alive and well there. In Germany, employers have moved in the other direction and they had almost no layoffs in the last recession.

So, ten million people can get laid off all at once and employers still maintain the hiring heuristic that, well, if you’ve been laid off for a while, then you’re a riskier hire. We have lots of evidence that if you are unemployed—well, if you’re unemployed, period, but particularly if you’re unemployed for over six months—you’re somewhere between half as likely and a quarter as likely to get an interview, even when you have identical education and work experience as someone who currently has a job.

When you add up all of those factors—how employers hire, how much they fire, how they train, who they train—the data on internal employer training is not so great, so we don’t know exactly what’s going on. But we do know that training matters a lot. We know that employers, for example, spend probably something like 20 times as much as the federal government does on training.

We can estimate that on a per-worker basis, training expenditures have dropped by about 30 percent in the last 20 years or so. Although it’s harder to quantify, there’s a lot of evidence that the pattern of expenditure on training has shifted towards the middle and the top of the occupation and wage stack in companies, and most of the cutbacks in training expenditures are at the lower end, or frontline workers.

Typically, if you’re a frontline worker (e.g. working in a retail store, a factory, a warehouse, or a call center, for example) and no one’s reporting to you, then you’re probably just being trained for safety, compliance, and efficiency. You’re not being trained like higher-income workers who are being trained for job progression and cross-training, developing their human capital in ways that would allow them to contribute more and to earn more over time.

There’s tremendous bifurcation now in the labor market. It’s driven more from the demand side and it really flows back into the supply side of the market.

We need to look at the demand side harder, however it’s not at all to say that there isn’t a lot of improvement that’s needed in higher education and job training and K-12 education and the like; there really is. But when you look at job training and at the parts of higher education where the implicit or explicit promise to the student is that they’ll be able to earn a higher wage, get a better job, the fact that the demand side is so misaligned makes it very difficult to change the supply side.

To put it another way, we say there’s a skills gap and that we want to train a bunch of people for skills. But the first five steps of the six-step hiring process are not about skills, rather they are actually about pedigree and history. Even if you train someone to have those skills, you don’t have a time machine, so you can’t go back in time and change their history or pedigree. All you can change is their skills. Until we have a labor market where people can get hired based exclusively on their skills, abilities, readiness, independent of how they got there, then the labor market is going to continue to keep a lot of people stuck, shut a lot of people out, and ultimately a lot of people will drift off.

As a result, you get lower labor force participation, lower voluntary mobility, and less wage growth. When people are able to go find a better job (to work at something, and then to earn more as a result) that’s where half of wage growth comes from, not just sticking in your role and getting an occasional cost-of-living increase.

HB: So there are two things that I want to hit on specifically. First, I just want you to connect the dots for me about what you said—how employers look for resumes, which is about what workers have done, and not competencies, which is what workers can do. I believe that’s exactly what you were just speaking to, so I just want to make sure I’m getting the lingo and how you’re thinking about it correct.

But second, one thing that I think is very interesting is how this looks different up and down the skills ladder. You talked about how there are fewer openings for entry-level positions and that there’s less training at the bottom and still more training at the top. Could you say more about how the demand side is driving that across the ladder? Is it that employers don’t think that folks at the bottom are trainable? Or is it that they don’t feel that those jobs need much training? Could some of this be that jobs are shifting so much that the ones at the bottom are being so deskilled that that’s a shift employers are making? Or is it tied up with other kinds of demand-side issues? 

So a two-part question. Take them in whichever order you would like.

BA: I’ll take the first question first—this observation that employers look for resumes instead of competencies.

Any competent employer will tell you, “No, we’re looking for competencies. We want people who can really do the job.” But if you break down their hiring process, you’ll find that competencies and sort of demonstrating what you can do often make up the last step or perhaps the last two steps in the hiring process.

The smart businesses will look for competencies. But again, they’ll do that with a handful of people that are finalists for the position, right? That might have started with several hundred or several thousand people sending in their resumes for that position and others coming through references and the like.

Companies need a way to screen, in the sense that it would not be cost effective for them to evaluate the competencies of 3,000 different people, on their own, through their existing processes. They narrow it down by keyword algorithms on resumes, so most of those resumes are never seen or evaluated by a human being.

They’re screened on job-applicant tracking software that’s looking for certain keywords associated with that job. And they are screened based on educational qualifications. Jobs get defined as, “this job requires a four-year degree,” or “this job requires specific years of work experience in specific roles,” and the like.

Unfortunately, this process often leaves out somebody who has not walked this sort of straight path where they had a good high school education, then went to college, graduated from college, or perhaps from a community college in a very in-demand, well-structured program with good employer ties.

There are a few ways to getting a job that work, but the ways that work more straightforwardly represent about half of our labor market, give or take. And for the other half, they’re going a more circuitous route.

There are 35-40 million Americans who went to college but never graduated. They don’t have that degree. On average, they make a little bit more than high school graduates, but their earnings are much closer to high school graduates than to bachelor’s degree graduates.

Then there are workers who have developed skills on the job but don’t have any credentials associated with it. They managed to stay at their company and their company is doing well. The people around them know that they can do that job. But if they ever want to apply for even the same job at another company, they’re very likely to be screened out by the educational requirements, which is another reason you see people not being able to move jobs and this huge decline in job mobility that’s been largely unexplained. If we start looking at some of those institutional factors and demand-side behaviors, we’re going to get a lot more of an explanation for why that is happening.

So consider the alternative. What if an employer, defined it in terms of “This is what we need you to be able to do to meet this standard in this context, and here’s how you can demonstrate it” instead of making a job description and the associated processes of hiring based on someone’s history and pedigree?

And what if there are a variety of ways you can demonstrate that you can do this now. In other words, if you can do the job, you can get the job. What if that were the norm and we built systems around that?

I’ve been working with this organization Opportunity@Work on the information technology occupations and trying to apply it there. Take coding or computer programming, for example. Today, if you want to know whether someone can code, you can look at their resume or you can look at their code. And there are standard repositories like GitHub that you can use.

So instead of using a software algorithm on someone’s resume, why not use a software algorithm to look at the quality of someone’s code?   That changes the nature of the demand signal.

And this is a very important point, because there are two big implications of changing that structure of the demand side of employment. The first is that there are millions of people today who can do more than they are allowed to do. In other words, you might be a bus driver and you’re running your church website, but you can’t get a job running a website for a company because you’re a bus driver. You might get a good reference from someone, but you wouldn’t get through the typical hiring process. And there are millions of people who could do more, earn more. They could fill those jobs already for which employers say they have trouble hiring.

So that’s one phenomenon. But that’s not enough. There are not enough of those people [who can do the job] to fill all the jobs that employers would characterize as a skills gap. But if you change that demand signal to be truly based on competency and ability and skill, then you change the business model for training which benefits those who don’t yet have the skills, but are capable of learning. Because today, if you train someone for a job and that person lacks the pedigree for the job, lacks the past history, the professional history, the educational history, that person is unlikely to get the job, even if you trained them.

Even if they have the ability, the underlying potential, and you train them well, they probably still can’t get the job. And as a result, it doesn’t make sense to train them. It’ll be a failed government program if the government pays for that training. It’ll be an unsustainable nonprofit if a nonprofit does that training. And it’ll be a failed business if a business does that training. And the individual will have wasted their time because they won’t actually get the job under today’s standard hiring heuristic.

But if, instead, a business had a robust hiring channel where you could get the job if you acquired the skills, then to train someone with the aptitude and the motivation for that job would be an excellent business investment, an excellent public investment, and an excellent investment of that person’s time.

So if you create that, then, that’s the second-order effect and it’s much larger, because there are many more people with the underlying potential to learn and to master a set of skills than there are people who already have those skills and can’t get the job.

That’s the second big wave. But then the third thing is, if that’s true, then everyone from entrepreneurs to social entrepreneurs to a creative kind of government can enable entirely new sorts of business models or policy approaches, but underpin the scaling of human capital acquisition, which I think is sort of the breakthrough.

But again, once you start understanding that this is a market, you understand that you actually have to change that initial demand signal first. For example, until there’s a sufficiently large market for a certain manufactured product, there’s no demand for enough factories to build it. And if there’s not enough demand for factories, then there’s not enough demand for machine tools for those factories. You see, it sort of goes back in the chain.

The labor market is an $8.5 trillion market in the United States and essentially the market in which investments in human capital realize their return, then you see that it flows back to all of our human capital kind of systems, which is a large part of our economy.

HB: So you say that the first step is to figure out the policy solutions. What do you think the first key policy steps are on the demand side to make that happen? What would you say the most urgent ones are—or are you guys there yet at Opportunity@Work?

BA: I think it’s a matter of both policy and practices, in the sense that not all of this can be solved by public policy. It has to be solved by employer practices, which public policy can enable, can incentivize, but can’t mandate at the level of specificity that would be necessary to still keep up with changes in the labor market.

The approach we’ve taken at Opportunity@Work is to really think about where the capabilities and mandates of government, business, the nonprofit sector, and these sort of technology platforms can be rewired, combined in different ways.

If you think about the problem we’re trying to solve, it is a market-based collective action problem. Hiring people, it’s not consumption for a company. It is an input to a complex and rapidly changing set of production processes meant to try to target a moving market, or set of moving markets. So it moves very quickly and government really can’t do market-based activities at that level.

On the other hand, the phenomenon we’re talking about on the demand side is really a collective action problem, because companies transactionally, individually, have a very strong incentive to poach an already experienced and trained worker instead of hiring a promising worker that they would have to train and give experience. But if they’re all poaching out the front door, then everybody’s being poached out the backdoor, and they’re not creating a new supply of people.

HB: Did you just make an argument for why the non-compete clauses that many workers are being forced to sign might actually be a good thing?

BA: No, I don’t believe these non-competes are a good thing. I don’t believe excessive requirements for licensure and so forth are a good thing. I think they’re all bad things.

I think they’re all part and parcel of a reality that much of our public policy—a lot of it at the state and local level—and many of our business practices really make it more difficult for people to find their path to the work that would give them the most satisfaction, that would produce the most value, and in which they could be most highly compensated in whatever mix of passion, freedom, and joy that reflects what they want.

So, no, I think they’re pretty much all bad, because there’s been a reduction in voluntary job mobility as a result of a lot of these things, including the licensure, the non-competes, and the like.

HB: But if we have a labor market where people are changing jobs a lot, isn’t there an upside to policies that might make it harder for people to switch, because it would create more incentives for businesses to invest in the talent that they have?

BA: It’s a logical argument on the face of it, but when you look at the data, you see it’s a myth that there’s some Millennial wanderlust and people just want to switch jobs more. They’re job-hopping more.

The reality is that businesses are treating workers more as a kind of sort of contingent workforce, whereas before, a much larger portion of businesses had a model where they thought of their employee base a little bit more like a long-term asset.

It’s not that people wouldn’t change jobs, but on average, businesses were investing as if they were going to keep people for a long time, to the point of accepting significantly lower profits during recessions so that they could hold on to two-thirds of the people they would otherwise have to lay off.

It’s businesses that stop doing that, right? Employers stop doing that.  The volatility in the labor market is driven much more by employers hiring up when demand goes up and then firing and laying off people quickly when demand drops.

That’s where the volatility is coming from, much more from the employer side than on the employee side. On the employee side, actually, the voluntary quits are lower at any given point in the business cycle than it used to be. And geographic mobility is much less; it’s half of what it was 30 years ago.

So in theory, what you’re asking is plausible. But when you look at the data, that’s not what’s going on. The volatility in people’s working lives is being driven much more by changes in employer behavior than by changes in individual worker preferences. And that has a lot of implications.

HB: Well, I think that’s all the time we have, but this has been very insightful and enormously helpful. Thank you so much, Byron.

BA: Thank you. I enjoyed it. Take care.

Women have made the difference for family economic security

Posted on by David Hudson

Overview

The steady movement of women into the U.S. workforce over the past half-century has dramatically changed the composition of family incomes across the country and up and down the income ladder. All these additional earnings, however, have not meant that family income has increased faster than in earlier eras. A breadth of research demonstrates that overall family economic security in the United States has been declining since the 1970s, especially among low- and middle-income families. As a result, even as more women have joined the labor force and families have lost their time for caregiving, too many families’ continue to face economic insecurity.

This issue brief explores how over the past four decades, women’s increased earnings and increased annual hours of work have been essential as families across the United States seek to find and maintain economic security. Using data from the Current Population Survey, we document how family income has changed between 1979 and 2013 for low-income, middle-class, and professional families, decomposing the differences in male earnings and female earnings from greater pay, female earnings from more hours worked, and other sources of income over this time period.

Download File
Women have made the difference for family economic security (pdf)

Read the full PDF in your browser

This analysis is an extension and update of the analysis presented in the forthcoming book “Finding Time: The Economics of Work-Life Conflict” authored by one of the co-authors of this issue brief, Equitable Growth’s Executive Director and Chief Economist Heather Boushey. Here are our key findings:

  • Between 1979 and 2013, on average, low-income families in the United States saw their incomes fall by 2.0 percent. Middle-income families, however, saw their incomes grow by 12.4 percent, and professional families saw their incomes rise by 48.8 percent.
  • Over the same time period, the average woman in the United States saw her annual working hours increase by 26.4 percent. This trend was similar across low-income, middle-class, and professional families.
  • Across all three income groups, women significantly helped family incomes both because they earned more per hour and worked more per year. Women’s contributions saved low-income and middle-class families from steep drops in their income.

These findings establish that working women, especially those within low-income and middle-income families, have made the key difference in securing earnings for their families. Without women’s added earnings, families would be much worse off.

Women’s changing role in the U.S. labor force

An increasing number of families across the United States, especially low- and middle-income families, are becoming less economically secure. This trend began more than four decades ago, long before the Great Recession of 2007–2009. In light of this increasing instability and stagnant growth in family incomes, families have had to find ways to cope—including an growing reliance on the earnings of women. The role of women in the United States has transformed from predominantly being a wife or mother to being all of these things and a breadwinner.

Half a century ago, women—especially middle-class women—began entering the U.S. labor force and staying there, although they were still more likely to be their family’s caregiver for children, the aging, and the ill. And starting in the 1970s, more women started gaining professional degrees, which along with other factors contributed to a sharp rise in women’s labor force participation as well, especially for prime-age working women (ages 25 to 54). By 2000, about 60 percent of all U.S. women were in the labor force, which remained the case until the financial crash in 2007 and the ensuing Great Recession. (See Figure 1.) 

Figure 1

 

As women engaged more in the labor force, they also started working more hours on average. In 1979, 28.6 percent of all U.S. women were working full time. By 2007, right before the Great Recession, this number had increased to 43.6 percent. This change has made a substantial difference at a macroeconomic level: Our gross domestic product in 2012 would have been $1.7 trillion less if women had not increased their working hours over the past four decades.

Updating and building on the findings that Heather Boushey presents in her forthcoming book, “Finding Time: The Economics of Work-Life Conflict,” this brief explores the difference that women’s increased work hours have made at the family level, and how this effect varies for different types of families. Using data from the Current Population Survey, we calculate how family income changed between 1979 and 2013 for low-income, middle-class, and professional families in the United States, decomposing it by differences in male earnings, female earnings, and other sources of income over that time period. We further dissect the change in female earnings to find what portion of that change is due to increases in pay and what portion is due to working more hours in 2013 in comparison to 1979.

Defining income groups

This analysis follows the same methodology presented in “Finding Time.” For ease of composition, we use the term “family” throughout the brief, even though the analysis is done at the household level. We split households in our sample into three income groups. Low-income households are those in the bottom third of the income distribution, which means those earning less than $25,440 per year in 2015 dollars. Professional households are those in the top fifth of the income distribution who have at least one household member with a college degree or higher; in 2015 dollars, these households have an income of $71,158 or higher. Everyone else falls in the middle-class category.

Setting some context

Before we delve into the main analysis, let’s first set some context for how family incomes and women’s hours have changed across low-income, middle-class, and professional families.

How did family incomes change between 1979 and 2013?

The changes in average U.S. family incomes between 1979 and 2013 show widening inequality, consistent with other research. In 1979, low-income families had an average annual family income of $23,697 in 2015 dollars—by 2013, that number dropped by 2.0 percent to $23,224. Middle-class families only fared marginally better: In 1979, they had an average income of $72,168, which by 2013 rose by 12.4 percent to $81,152. Over the same time period, however, professional families saw a 48.8 percent increase in their average income, going from $132,492 in 1979 to $197,141 in 2013. (See Figure 2.)

Figure 2

 

How did women’s working hours change between 1979 and 2013?

Women across all three income groups saw their working hours rise. In 1979, on average, women from low-income families worked 629 hours per year (or 12 hours per week), and by 2013 their annual work hours had grown by 24.1 percent to 780 hours (or 15 hours per week). Similarly, middle-class and professional women’s hours grew by 25.7 percent (from 23 per week in 1979 to 29 hours per week in 2013) and 29.4 percent (from 26 hours per week in 1979 to 34 hours per week in 2013) over that same time period, respectively. (See Figure 3.)

Figure 3

 

Decomposing the changes in family income between 1979 and 2013

Figures 2 and 3 establish that between 1979 and 2013, even as women’s working hours increased at parallel rates for all income groups, family income was either stuck or stalled for low-income and middle-class families. This paradox is resolved when we break down the changes in average family income into male earnings, female earnings, and income from Social Security, pensions, or any other non-employment-related source over that same time period.

Specifically, we divide female earnings into the portion due to women earning more per hour and that due to women working more per year. To calculate female earnings stemming directly from the additional hours worked, we take the difference between 2013 female earnings and the hypothetical earnings of women if they earned 2013 hourly wages, but worked the same hours as women did in 1979. (For more on how we did this calculation, please see our Methodology.) It turns out that this is an important distinction because, as shown in Figure 4, the difference in female earnings due to additional hours has made a significant positive difference for every income group, especially for low-income and middle-class families. (See Figure 4.)

Figure 4

 

Between 1979 and 2013, low-income families saw their income fall, as shown in Figure 2; in Figure 4, we see that this is because men’s earnings fell. Women not only increased their working hours but also their pay per hour—so much so that their overall contribution grew the average annual family income by $1,929 making up for much of the losses from declining male earnings. Women’s earnings from working additional hours alone added $1,473.

In middle-class families, the average annual income grew by $8,984. But, like low-income families, this change was also entirely due to women’s added hours and earnings. Between 1979 and 2013, women’s earnings from working more hours accounted for the largest share of the gains, adding $5,703. The second-largest factor was once again women’s pay, which contributed about $4,925. Also similar to low-income families, men’s earnings pulled down overall income within middle-income families, falling by $4,278.

Professional families experienced significant income growth between 1979 and 2013 across all earnings and income categories. As with low-income and middle-class families, women’s earnings from higher pay and increased hours were the most important factor. Women’s earnings from pay increased by an average of $20,274 per year, while women’s earnings from more hours accounted for $14,188. However, as opposed to families further down the income spectrum, men in professional families earned on average $24,936 more per year over this period, helping boost family income.

Conclusion

Across every income group, women’s increased working hours have helped bolster family economic security. But for low-income and middle-class families, women’s contributions, particularly from working more per year, have been essential in abating the effects of stuck or stalled family income growth. Simply, women’s participation in the workforce has made the key difference for middle-class families and more vulnerable families on the brink.

Families have benefited greatly from the changing roles of women in the home and the work force. Yet across the U.S. job market and across all levels of the income distribution, men and women face daily conflicts between work and family. As researchers strengthen the finding that what happens within a family is just as important to the economy as what happens within a business, it is imperative that we design policies that support families where they live, work, and play. These policies, among them work flexibility, paid family and medical leave, and child care, must be designed so that their distributional effects help all families equitably in order to truly strengthen our economy. 

Heather Boushey is the Executive Director and Chief Economist at the Washington Center for Equitable Growth and the author of the forthcoming book from Harvard University Press, “Finding Time: The Economics of Work-Life Conflict.” Kavya Vaghul is a Research Analyst at Equitable Growth.

Acknowledgements

The authors would like to thank John Schmitt, Ben Zipperer, Dave Evans, David Hudson, and Bridget Ansel. All errors are, of course, ours alone.

Methodology

The methodology used for this issue brief is identical to that detailed in the Appendix to Heather Boushey’s “Finding Time: The Economics of Work-Life Conflict.”

In this issue brief, we use the Center for Economic and Policy Research extracts of the Current Population Survey Annual Social and Economic Supplement for survey years 1980 and 2014 (calendar years 1979 and 2013). The CPS provides data on income, earnings from employment, hours, and educational attainment. All dollar values are reported in 2015 dollars, adjusted for inflation using the Consumer Price Index Research Series available from the U.S. Bureau of Labor Statistics. Because the Consumer Price Index Research Series only includes indices through 2014, we used the rate of increase between 2014 and 2015 in the Consumer Price Index for all urban consumers from the Bureau of Labor Statistics to scale up the Research Series’ 2014 index value to a reasonable 2015 index estimate. We then used this 2015 index value to adjust all results presented.

For ease of composition, throughout this brief we use the term “family,” even though the analysis is done at the household level. According to the U.S. Census Bureau, in 2014, two-thirds of households were made up of families, defined as at least one person related to the head of household by birth, marriage, or adoption.

We divide our sample into three income groups—low-income, middle-class, and professional households—using the the definitions outlined in “Finding Time.” For calendar year 2013, the last year for which we have data at the time of this analysis, we categorized the income groups as follows:

  • Low-income households are those in the bottom third of the inflation- and size-adjusted household income distribution. These households had an income of below $25,440 (as compared to $25,242 and below for 2012). In 1979, 28.3 percent of all households were low-income, increasing to 29.7 percent in 2013. These percentages are slightly lower than one third because the cut-off for low-income households is based on household income data that includes persons of all ages, while our analysis is limited to households with at least one person between the ages of 16 and 64. The working-age population (16 to 64) typically has higher incomes than older workers, and as a result, the working-age population has somewhat fewer households that fall into this low-income category.
  • Professionals are those households that are in the top quintile of the inflation- and size-adjusted household income distribution and have at least one member who holds a college degree or higher. In 2013, professional households had an income of $71,158 or higher (as compared to $70,643 or higher in 2012). In 1979, 10.2 percent of households were considered professional, and by 2013, this share had grown to 16.8 percent.
  • Everyone else falls in the middle-class category. For this group, the household income ranges from $25,440 to $71,158 in 2013 (as compared to $25,242 to $70,643 in 2012); the upper threshold, however, may be higher for those households without a college graduate but with a member who has an extremely high-paying job. This explains why within the middle-income group, the share of households exceeds 50 percent: The share of middle-income households declined from 62 percent in 1979 to 53.4 percent in 2013.

Note that all cut-offs above are displayed in 2015 dollars, using the inflation adjustment method presented earlier.

In our analysis, we limit the universe to persons with non-missing, positive income of any type. This means that even if a person does not have earnings from some form of employment but does receive income from Social Security, pensions, or any other source recorded by the CPS, they are included in our analysis.

These data are decomposed into income changes between 1979 and 2013 for low-income, middle-class, and professional families. The actual household income decomposition uses a simple shift-share analysis to find the differences in earnings between 1979 and 2013 and calculate the extra earnings due to increased hours worked by women.

To do this, we first calculate the male, female, and other earnings by the three income categories. To calculate the sex-specific earnings per household, we sum the income from wages and income from self-employment for men and women, respectively. The amount for other earnings is derived by subtracting the male and female earnings from total household earnings. We average the household, male, female, and other earnings by each income group for 1979 and 2013, and take the differences between the two years to show the raw changes in earnings by each income group.

To find the change in hours, for each year by household, we sum the total hours worked by men and women. We average these per-household male and female hours, by year, for each of the three income groups.

Finally, we calculate the counterfactual earnings of women. We use the 2013 earnings per hour for women and multiply it by the 1979 hours worked by women. Finally, we subtract this counterfactual earnings from the female earnings in 2013, arriving at the female earnings due to additional hours.

One important point to note is that because of the nature of this shift-share analysis, the averages don’t exactly tally up to the raw data. Therefore, when presenting average income, we use the sum of the decomposed parts of income. While economists typically show median income, for ease of composition and the constraints of the decomposition analysis, we show the averages so that the data are consistent across figures. Another important note is that we make no adjustments for changes over time in topcoding of income, which likely has the effect of exaggerating the increase in professional families’ income relative to the other two income groups.

(Featured photo credit: Flickr/yooperann)

Home economics

Posted on by David Hudson

The following excerpt is from an essay adapted from Heather Boushey’s forthcoming book, “Finding Time: The Economics of Work-Life Conflict.” Read the full essay here, originally published in the latest issue of Democracy.

American businesses used to have a silent partner. This partner never showed up at a board meeting or made a demand, but was integral to profitability. That partner was the American Wife.

She made sure the American Worker showed up for work well rested (he didn’t have to wake up at 3 a.m. to feed the baby or comfort a child after a nightmare), in clean clothes (that he neither laundered nor stacked neatly in the closet), with a lunch box packed to the brim with cold-cut sandwiches, coffee, and a home-baked cookie. She took care of all the big and small daily emergencies that might distract the American Worker from focusing 100 percent on his job while he was at work. Little Johnny got in a fight on the playground? The American Wife will be right there to talk to the school. Aunt Bea fell and broke her hip? The American Wife can spend the afternoon bringing her groceries and making her dinner. The boss is coming over for dinner? The American Wife already has the pot roast in the oven. Even if she had a job—and a certain percentage of American wives, like my mom, worked and had to work despite suburban cultural expectations—she was still the primary caregiver when work-life conflicts arose. The presumption was that she would be the one at home.

This meant that for decades, the American Wife gave American businesses a big, fat bonus. Her time at home made possible the American Worker’s time at work.

This unspoken yet well-understood business contract is now broken. Moreover, it doesn’t look like we’re going back to it anytime soon. Nor should we. American families look different today. Wives—and women more generally—work outside the home because they need to and because they want to.

“Finding Time: The Economics of Work-Life Conflict” will be available on April 19, 2016.

Give working women their due for caregiving in economic policy debates

Posted on by David Hudson

Economists love their economic models and their footnote citations just as much as politicians love it when economists ply their tools of the trade in support of these politicians’ economic policies. In this regard, Sen. Bernie Sanders is no different than Sen. Marco Rubio, and former Secretary of State Hillary Clinton no different than former Gov. Jeb Bush. Yet fact-based, data-driven economic modeling based on the best available evidence often misses the mark due to seemingly obvious missing factors.

Take the still-simmering internecine battle among left-leaning economists over Senator Sanders’ economic policy proposals—which University of Massachusetts Amherst economist Gerald Friedman models, showing that U.S. economic growth would top 5 percent if he becomes president and if he is able to implement his entire platform immediately. Setting aside all of the politically unknowable “ifs” in that last sentence, Friedman’s model tells us that the Sanders growth machine will be powered by a swiftly rising employment rate—the share of the U.S. population with a job—up to 65 percent by 2026, 8 percentage points higher than current forecasts by the Congressional Budget Office. These are big employment gains.

In support of this projection, Friedman argues in footnote 22 of his paper that women’s labor force participation will rise because women’s wages will rise after President Sanders signs into law, implements, and enforces the Paycheck Fairness Act, which would reduce discrimination against women workers. Pay is certainly an important reason that women work, but it’s not the only reason that women stay out of the labor market. Research shows that policies such as paid family and medical leave, stable workplace schedules and scheduling flexibility (that works for workers, not just their employers), and, especially, safe, affordable, and enriching child care and elder care boost women’s employment and that of caregivers more generally.

Generations of economists have failed to put the need for care and the effects of that care provided predominantly by women into their labor supply economic models. I cannot make every economist fix their models, but it’s entirely unrealistic—and unjustifiable—to make assumptions about work and time spent in the workforce that ignore the everyday economic realities facing families. Politicians, however, are starting to get what so many economists have missed. Senator Rubio, Secretary Clinton, and Senator Sanders all have policy platforms that include paid family leave, and Secretary Clinton talks a lot about her plans for expanding access to child care.

These policies have demonstrable effects on employment and earnings. Researchers have shown that California’s paid family leave law has increased leave-taking for mothers and fathers and improved mothers’ employment outcomes. Mothers who have access to this new benefit are more likely to return to work after the birth of a child, and when they return to work, they put in more hours at work compared to mothers who did not use paid leave. These outcomes are supported by the fact that leaves are equally available to men and women; because fathers are using their leave, too, this gives mothers the support they need to address work-life conflict. More striking are the labor supply effects of universal child care policies. After Quebec implemented a universal child care program, for example, researchers found that mothers’ labor force participation rose by 13 percent.

This academic and policymaking attention to the issues of work-life conflict in our economy and our society on the presidential campaign trail follows many successes in implementing new policies at the state and local level. Very recently, Vermont’s legislature passed a bill that Gov. Peter Shumlin has expressed support for, giving workers employed more than 18 hours a week the right to earn up to five paid sick days each year (although only three days in the first two years of the law’s implementation). This is the 28th place in the United States to put such a policy in place, following four other states, 21 cities, one county, and the District of Columbia.

Yet, too often, politicians tend to see work-life policy as just another sop to just another interest group: women. (Never mind that we’re the majority of the population.) This is a big mistake. Economists and policymakers alike need to put women’s participation in the economy and in family life at the center of how our economy grows and works now and over time.

To understand what makes the economy grow, we need to know what hinders people from fully engaging in the labor force. Our nation’s inattention to the causes and consequences of work-life conflict is a serious hurdle for many families. Fixing it will require serious research followed by evidence-based policymaking to change how work is done in our society.

The consequences of higher labor standards in full service restaurants: A comparative case study of San Francisco and the Research Triangle in North Carolina

Posted on by David Evans
Workers prepare for lunch in a San Francisco restaurant kitchen.

Overview

Across the United States, policymakers in states and cities are grappling with a groundswell of public support for higher local minimum wages as well as other improvements in labor standards, among them better health care and paid sick leave for employees. In many of these communities the so-called “Fight for $15” movement, led predominantly by fast-food restaurant workers seeking to raise local minimum wages to that level, is spurring policymakers to consider the economic merits and possible adverse effects of such a policy move.

Download File
inside-monopsony-ib Download this Issue Brief

Overall, most economists agree that moderate increases in the minimum wage do not result in job losses. In fact, boosting the minimum wage may reduce employee turnover—a net positive result for employers who could spend less on hiring and training new workers and enjoy sustained productivity from their employees. Yet economists are less sure how locally-enacted minimum-wage raises and higher labor standards reshape employment practices within individual companies. One way to find out is to compare one community where higher labor standards are in place with one where there are no enhanced standards, focusing on one industry in particular.

Such a comparison is particularly apt in the full service restaurant industry in San Francisco and the Research Triangle communities in and around the cities of Durham, Raleigh, and Chapel Hill, North Carolina. Local labor standards in San Francisco include the nation’s highest minimum wage, a mandate for employee health care, and paid sick leave. In contrast, full service restaurants in Research Triangle communities follow lower federal minimum wage guidelines, including much lower hourly wages for tipped employees ($2.13 per hour), and are not required to provide employee health care or paid sick leave.

This issue brief details the findings of a comprehensive research paper that I presented at the Labor and Employment Relations Association Conference in January of this year. I examine employer responses to higher labor standards through a qualitative case comparison of the full service restaurant industry across these two fundamentally different institutional settings. The results are striking.

In San Francisco, higher labor standards led to greater “wage compression” in specific occupations within the restaurant industry, meaning that employers had less “wiggle room” to offer slightly higher wages to cooks or dishwashers or food servers or bartenders. Concurrent with this wage compression was a rise in professional standards as employers sought to hire and keep already well-trained workers at higher wages and with expanded benefits. Both developments reduced turnover and attracted more professional employees who maintain a high level of customer service.

In the Research Triangle region, the lack of higher labor standards led to a wider distribution of wages across the industry and within individual establishments. Thus employers could differentiate wage levels within job categories to a greater degree. This allows a labor practice that offers low-wages for entry-level employees with little experience but accepts a high rate of turnover as a result. This practice also translates into higher training expenditures for firms. In San Francisco, employers required more experience and professionalism from their new hires.

In both cities, however, a large wage gap remains between front-of-house and back-of-house occupations—a gap that correlates strongly with existing racial and ethnic divisions within restaurants (Latinos in the back; whites in the front). Some employers in San Francisco are addressing this gap by radically restructuring their compensation practices by adding service charges and, in some cases, eliminating tipping. In these restaurants, wages are more balanced across workers, whether they are making food in the kitchen or taking the orders and serving food and drinks to customers.

These findings are important for state and local policymakers to consider. Full service restaurants in the United States added 811,700 jobs nationally between the end of the Great Recession and October 2015, outpacing overall private-sector job growth by nearly 7 percent. What’s more, this trend is expected to continue as jobs in food service occupations are projected to grow faster than the overall labor market through 2030. Thus the restaurant sector is a useful harbinger for the predominant labor market conditions that policymakers can expect going forward—namely the proliferation of low-wage jobs in service industries that can’t be offshored.

Understanding how labor standards affect the pace of job creation and more general aspects of the employment is critical. In the full service restaurant industry in San Francisco, higher labor standards suggest the following results may occur in other cities enacting similar policies:

  • Higher professional standards may result in lower employee turnover and more productive workers.
  • Lower employee turnover and more productive workers may increase sales for owners and ultimately create better dining experiences for customers through better service.
  • Higher professional standards may limit entry-level opportunities within the industry, while lower standards may result in more employer-provided training for new workers.
  • Currently large wage gaps based on race and ethnicity between restaurant workers in the kitchen and servers and bartenders interacting directly with customers are not fully resolved by higher labor standards.
  • Steps to end tipping in favor of salaried employees in the front and back of restaurants may result in more uniform wages within restaurants, and may result in less ethnic and racial inequality within individual restaurants.

There are, of course, limitations in how much local labor standards can improve the quality of jobs within the restaurant industry. More research is needed to fully assess the impact on overall wage inequality and opportunity structures more generally. But broadly, my research suggests that “high-road” labor standards may well lift wages overall while reducing wage inequality and improving professionalism.

The full service restaurant industry today

The restaurant industry employs more than 10.2 million workers today, of which about 5.3 million are employed in full service restaurants, according to the U.S. Bureau of Labor Statistics. The restaurant industry overall epitomizes two trends evident in service industries overall in the United States—relatively high growth alongside low job quality—which is arguably why the sector is facing new demands for minimum wage increases at state and local levels.

The U.S. economy is slowly recovering from the depths of the Great Recession, yet the labor market continues to show weakness even as the unemployment rate has fallen to a below 5 percent. Despite optimistic accounts of the “re-shoring” of manufacturing and a recovering housing market, economic inequality is on the rise. That’s because the majority of jobs created since the end of the recession are relatively low-wage. Indeed, a unique feature of the current recovery is the so called “missing middle” in the pattern of job growth, with relatively few new middle-income positions or career pathways for workers who lack advanced skills. As a result, an increasing number of workers remain in low-wage positions for longer periods of time. Jobs that were once viewed as “stepping-stone” positions, among them restaurant work, are increasingly becoming relatively permanent careers that have little opportunity for long-term wage growth.

Recent research conducted by Restaurant Opportunity Centers United and affiliated scholars provides new evidence on wages, benefits, working conditions, and the extent of racial and ethnic discrimination. In 2011, the organization released a study conducted in eight large metropolitan regions that consisted of surveys and interviews with both employees and employers that documented the prevalence of low wages, lack of access to health benefits and sick leave, and persistent occupational segmentation by race.

More recently, scholars Rosmary Batt and Jae Eun Lee of Cornell University and Tashlin Lakhani of Ohio State University presented results based on a national employer survey across 33 large metropolitan areas focused on variation in human resource practices across restaurant market segments. They find a clear link between higher quality human resource practices and lower turnover. In a related study Batt highlights case studies of restaurants that pursued what she calls “high-road” practices, which include higher relative wages, more full-time work, and more investment in training—all of which resulted in lower employee turnover and improved productivity.

Conditions in the low-wage service sector overall are at a historically low level, with stagnant wages, uncertain working conditions and hours, and a hostile regulatory environment toward organized labor. Given these trends, a central concern for policymakers is what can be done to reduce wage inequality in these growth industries of the future. The impact of publicly mandated labor standards on employment and employee benefits is well studied and continues to be debated among economists and policy makers. But there are important missing pieces in our understanding of how locally enacted labor laws may have deeper consequences. In many ways, the behavior of actual firms remains a “black box” for researchers in the field. Raising labor standard has an affect beyond just the level of, among them changes in the rate of turnover, productivity, training, tenure, and professional expectations and norms.

A tale of full service restaurants in two cities

Using a methods-comparative case design that analyzes employment practices across the restaurant industries in two institutionally divergent urban labor markets—San Francisco, which has the nation’s strongest local labor standards, and North Carolina’s Research Triangle region, which does not—one can discern how higher labor standards affect wages and professionalism in the full service restaurant industry. (See the brief methodology at the end of this issue brief or the LERA website for a link to the full research paper.)

San Francisco employers must pay the nation’s highest minimum wage ($10.74 per hour rising to $15 by 2018), a pay-or-play health care mandate (up to $2.33 per hour) in which either the employer or the city provide employee health insurance, and paid sick leave requirements. In addition, tipped workers must be paid the full minimum wage. In the Research Triangle region of North Carolina there are no locally enacted labor standards. Thus, the effective wage in San Francisco is more than $13.00 per hour. North Carolina, in comparison, follows the federal standards of a $7.25 minimum wage and $2.13 tipped minimum wage, and has no paid sick leave or health care spending mandate.

While the regional labor markets of these two cases differ on a number of dimensions beyond the strength of local labor standards, there are a number of similarities that make this comparison plausible for detecting causal effects of labor standards. First, both regions are home to many high-tech employers and both have comparably tight overall labor markets. Lastly, both cases have a similar number of full service restaurant establishments, resulting in a similarly sized sampling universe.

The first finding in my research is perhaps the most telling—San Francisco’s higher minimum wage compared to North Carolina means that the San Francisco restaurants experience less variation in existing wages in different occupations within their establishments and among the establishments themselves. In other words, San Francisco experiences a general convergence of “high road” employer practices compared to the Research Triangle region’s existing “low-road” employer practices. (See Figure 1.)

Figure 1

The same pattern is apparent in other jobs within these restaurants in the two cities. Overall, the variation in wages is greater and the wages lower in the Research Triangle region North Carolina compared to San Francisco. (See Figure 2.)

Figure 2
Pushing the High Road Higher in San Francisco

My research also finds that other non-pay related labor standards also had an effect on restaurant labor practices. This is particularly evident in how some employers reacted to the enactment of San Francisco’s pay-or-play health care mandate. Rather than requiring employers to provide insurance directly to workers, the San Francisco Healthy Families Act of 2007 requires employers to pay up to $2.33 per hour worked for each employee. These payments can go either directly to the county health system—where resident workers can receive low-cost care—or into a separate health care spending account set up for each worker. This mandate involves a significant but uniform cost increase for all employers in the industry.

After the passage of this law, some employers decided to spend more than the mandated minimum for their employees’ health insurance in order to provide actual employer-subsidized health insurance to all workers—a benefit that is extremely rare in the industry. As one employer said: “This year for example, we did employee health insurance for everyone…now everyone has real insurance, not just the city thing. We think and hope it will help retain employees.”

Another San Francisco employer echoed the logic of providing full employer-sponsored health insurance rather than simply paying the lower cost option of a per-hour fee to the City. The manager of one neighborhood-based fine dining restaurant explained: “Part of our decision to offer health care goes beyond a simple cost-benefit. What’s another thousand dollars if you already have to spend a certain amount of money. There is a kind of revolutionary like revolt thing happening in that I’m not going to just sign a check over to the city. I’m going to actually give it to my employees. And then the other part is it becomes part of your hiring and your attraction is that you say hey, we offer full benefits.”

This manager’s initial sentiment reflects animosity toward the city government for enacting the Healthy Families law in the first place. Yet the employer’s actual behavior in paying more for full insurance indicates how the labor standard induced the employer to go above the minimum and embrace the potential retention and morale benefits for their workforce.

Beyond direct wage and benefit offers, employers in San Francisco reshaped other aspects of their employment relationship in an effort to differentiate themselves from other employers in the market, and to ultimately retain valued employees. Several interview subjects discussed how they attempted to create a unique work “culture” that is “exciting,” “fun,” or offers indirect benefits to workers, even in cases where employers cannot raise wages beyond the mandated level. One case in point: An owner-manager of a casual neighborhood restaurant allows line cooks to use the resources of the restaurant to further their career development and pursue income-generating work as part-time caterers. Another employer tries to retain key workers in lower-paid occupations through the use of in-kind compensation that is matched to the specific needs of the individual worker—in this case an employee-of-the-month award for back-of-the-house employees in the form of calling cards to reach families back in Mexico.

While these may seem like relatively minor gestures on the part of some employers, these forms of non-wage compensation represent additional ways in which employers try to differentiate themselves in order to retain workers. In the face of strong, binding labor standards that effectively limit the degree to which they can vary wage levels (“taking away the low-road”), employers try to structure their relationship with their workers in other ways. In these two examples we observe restauranteurs who—perhaps implicitly—are adopting some of the same progressive human resource practices typically associated only within high-skill industries or occupations. Specifically, they are recognizing and seeking to accommodate the individual needs of each worker, whether that relates to the worker’s need for outside income through catering or in-kind support of family obligations.

Worker training, retention, and productivity in the two cities

One reason employers need to do worker training in the San Francisco restaurant industry compared to the Research Triangle region is to retain workers. More stable jobs are definitely more in evidence in the city. The rate of turnover for the overall full service restaurant industry in San Francisco was 15.9 percent in 2012, according to official statistics from the Quarterly Workforce Indicators program. This compares to 31.1 percent in the Research Triangle region.

Importantly, however, this stark contrast in turnover is largely due to the relatively high rate of short-term workers who enter and exit employment at a given firm within the same quarter in the two cities. The difference in the turnover rate for “stable” jobs—meaning jobs that last more than one quarter—is much lower (12.9 percent versus 15.8 percent. This means that the full service restaurant sector in the Research Triangle region features a significantly higher number of unsuccessful, or weaker job matches than San Francisco’s restaurant sector. (See Figure 3.)

Figure 3


Those very short term jobs—lasting less than one quarter—are often described by labor economists and other observers as evidence of bad matches between employees and employers. Such high turnover is because workers quit to take a better job, stop working altogether, or were fired. But such high turnover is also indicative of employers operating in labor markets with lower standards hiring workers with weaker expectations of worker quality, which leads to a lower bar for entry level jobs and ultimately more firing of low-quality workers.

Such differing expectations are evident in the age and education of restaurant workers in in these two cities. The restaurant industry in the Research Triangle region tends to hire younger workers with a lower level of formal education. Specifically, 49.5 percent of workers in there are under age 24 or have less than a high school education, compared to 38.9 percent in San Francisco. Conversely, 40.6 percent of workers in San Francisco have some college or a bachelor’s degree or higher, compared to 29.7 percent in the Research Triangle Region. (See Figure 4.)

Figure 4

In addition to hiring an older and more educated workforce, San Francisco employers generally engage in more careful searches, which lead to overall better matches. First, employers in San Francisco report higher experience requirements for new hires across the occupational spectrum. As seen in Figure 4, only 8 percent of survey respondents in San Francisco reported that new servers could be hired without any previous experience in the restaurant industry, compared to 46 percent in the Research Triangle region. Also, a larger proportion of the San Francisco employers reported experience requirements of over one year—33 percent in San Francisco, compared to 25 percent in North Carolina.

The lower bar for entry into employment is also confirmed in employer interviews. One manager of a neighborhood bistro in Raleigh explained what he looks for in a new front-of-house worker: “Basically, we require [that a server] can work a four-shift minimum per week and go an entire shift, an entire eight-hour shift without smoking a cigarette and [without] any facial piercings or anything. Beyond that, just come in with a smile on your face.”

Even at restaurants in the region that do prefer experienced workers, managers and owners did not articulate how experience matters or which specific skills and industry-specific knowledge they require. As one upscale bar-and-grill manager explained: “We look for at least one year’s experience, but the biggest thing we look for is we look for the person. We don’t look for the skill. I could teach anybody how [to] wait tables [and] pour drinks. I can teach anybody how to cook steaks. What I can’t teach is how to be a good person.”

Employers in San Francisco discussed the minimum level of experience needed to work in front-of-house positions in a distinctly different tone. Rather than viewing servers as essentially interchangeable laborers who can be trained quickly and easily if they possess a modicum of personal hygiene and a friendly personality, employers in San Francisco exhibited a clear description of what a “professional server” was. One mid-scale restaurant employer said of her front-of-house staff: “We have a lot of people who have made it a career and they’re investing in the knowledge of the product and learning their trade or already know their trade because they’ve done it for years.”

Another San Francisco neighborhood bistro owner described the level and nature of experience needed to fill a server position at his restaurant. “Realistically, to work here, I would say [a server needs] five years of experience, because there’s a wine knowledge level that I expect that you really just couldn’t get any other way,” he said. “If you have ten years of experience at Applebee’s, that doesn’t do anything for me.”

Ultimately these responses indicate that employers in San Francisco are looking carefully at each candidates’ resume and approaching the hiring process with a set of expectations about the nature of work, the skills (how to manage a customer’s dining experience rather than just take orders), and industry-specific knowledge needed to perform at a high level. San Francisco employers tend to view their employees—front-of-house more so than back-of-house—as professionals rather than basic labor inputs.

This rise of professional norms—or the exhibited expectations of employers for certain worker traits that are typically associated with highly trained professionals—can also be seen in the unexpected finding on employer-provided training. Some labor market economists argue that “high-road” employers will spend more resources on training while “low-road” employers expect their low-wage workers to quit and because their low-wage workers seem easily replaceable. But at least in the full service restaurant industry just the opposite is true. San Francisco employers reported spending less time offering formal training periods for both front-of-house and back-of-house staff. Instead, they seek out and expect to find workers who already possess a high level of skills in the industry.

In contrast, more employers in the Research Triangle region discussed a recruitment and training model that was more likely to involve formal screening mechanisms for a high volume of applications and a longer, more formal training period for new hires (particularly for front-of-house workers). These training strategies are maintained to deal with the high level of labor turnover and the reliance on relatively less-skilled workers. The manager of a large sports bar-and-grill described the recruitment and training process at his establishment as highly scripted. “We do all of our applications online,” he said. “When people come in, we don’t physically hand them a piece of paper. We hand them a card. It tells them what website to go on. They go ahead and take an assessment. The assessment is scored, and then we get all those almost instantly. This web-based system pulls all the information up on a Manpower Plan, it tells us what they’ve applied for, where they’ve worked. Gives us a resume, and then it gives us a score on the assessment.”

The manager continued to explain that once an employee is hired, they enter into a formal training period that is standardized for each occupation. “Training is a huge investment for us and it is constant,” he said. “Training days depend on the position. Bartending training is ten days and servers require eight days. In the kitchen it’s probably about ten days. Every day they write note cards on all their recipes. But they’ll take a final. When they take their final, their test in the kitchen, they have to know every ingredient, every ounce, and every item, for the entire station. That’s why we require them to write note cards.”

Even at higher-end restaurants, employers in the region have built a human resource system that accepts a high rate of turnover. “We try to stay ahead of the game so that we’re always hiring, we’re always interviewing, but hopefully it’s not desperation hires,” says another manager. “And we try to have a mix of needs like people who need fulltime, who can work lunches and brunches and all of that, to servers who really want very part time so that you can kind of over staff on busy shifts and then there’s always someone that wants to go home. There’s always a student that would like a Saturday night off.”

Training in San Francisco is decidedly different. Employers there stress “professional norms,” which translates into efforts to support continuous skill upgrading and quasi-professional development activities that are integrated into the jobs themselves. One employer described that in addition to limited initial training for servers, the restaurant has designed a system to support ongoing knowledge development. “Sometimes we’ll assign different topics like rum to one person and then they come back and they’re responsible for training everyone else, doing kind of an in service just to keep it interesting, keep them motivated to learn,” he explained. “If they’re having to present it to someone else, they’re going to want to know the product. It’s sort of a team approach, you would use the whole team to train the rest of the team. Next week somebody gets vodka, next week somebody gets some small winery up in Napa. And we don’t just do products, sometimes we’ll do a certain vegetable, they have to find out the history of it.”

Another San Francisco employer explained that the opportunity to learn on the job actually becomes a recruiting and retention tool for his staff. “The attraction of working here is that they get to taste a lot of wines,” he says. “It’s a big wine list. They can kind of flex their wine muscles a little bit and be like kind of like mini-sommeliers on the floor. They don’t hand over all the wine sales decisions to me or someone else. They handle it themselves. We’ve had no turnover for two years.”

The upshot? San Francisco employers seem to be seeking out better trained, more experienced workers and expecting more from them, which in turn leads to greater professionalism in San Francisco establishments. Specifically, employers in San Francisco readily describe their ideal employees in language typically used to describe professionals—meaning workers who have recognizable industry-specific skills, typically work full time, and invest in their own training.

The persistence of ethnic and racial divides in the full service restaurant industry

One persistent pattern in full service restaurants that hasn’t changed because of differing labor standards in the two cities is this—employers still view back-of-house workers (line cooks, prep cooks, dishwashers) in a less formal, more racialized frame. Listen to the manager of a corporate chain restaurant in Chapel Hill who also previously managed several independent restaurants in the region:

“The Latino workforce, these guys know how to work. They’ve been typically cooking in their own kitchens for large extended families. This is how they typically grew up. So it’s not like me cooking for a family of four at my house, or a family of five, or even doing a Thanksgiving dinner for maybe nine people. They’re cooking three meals a day or whatever it is, for their extended family or for many people in the household. I think that’s where a lot of those skills come into it just based on how they grew up. Compared to those workers with formal culinary education, I’ve probably kicked more people out of my kitchen who had a formal education, because they think they know everything now. It’s one of those things where if somebody taught you how to cook eggs right, if somebody taught you how to do certain things right then that’s wonderful, but can you actually get in that kitchen and perform and do multi tasks.”

The stated preference for Latino workers as prep cooks and line cooks undermines the utility of formal credentialing programs and codified skills that can be marketed across firms. The connection between ethnic background and perceived work ethic can lead to an assumption that Latino workers are monolithic and interchangeable. This ultimately limits the opportunities for individual workers to move up the pay scale.

In San Francisco, employers also offered a view of back-of-house workers that emphasized ethnic stereotypes rather than formal skills or credentials. Explained one ethnic restaurant manager in the city:

“You know, a line cook position, I hate to say it, most of them are my people, most of them are Mexican. And you know, you try to stay away from anyone who went to serious cooking school, went to a culinary academy, or has an AA in culinary kitchen skills. Mexicans are just a better quality cook, they really are. I hate to say it. They might not know what sous-vide is, but if you teach them once how to braise something, how to do it correctly, they’ll do it better than the guy who went to school. It’s just innate.”

Equating ethnic status with work ethic or “innate” ability may lower barriers to entry for new workers seeking a back-of-house job, yet the way employers frame skill through an ethnic lens reinforces the barrier between front-of-house and back-of-house workers. This barrier is important not only because it limits access to better paid server positions, but also because as labor standards rise wage differential grows. The barriers between back-of-house and front-of-house occupations is an observation that nearly all respondents in San Francisco brought up in response to direct questions about how they reacted to rising minimum wage and other labor standards. In particular, employers claim that higher labor standards exacerbate the difficulty they have in finding and retaining high quality line cooks and prep workers. In their view, since the mandates require them to give raises across the board, including tipped workers whose total hourly income already exceeds the new mandate, they have less financial flexibility to offer higher wages to non-tipped workers.

One of the more interesting way in which employers in the full service restaurant industry in San Francisco are responding to higher labor standards–and the persistent ethnic and racial divide between the back and front of the house–is through radically restructuring compensation practices. Specifically, some employers are eliminating tipping and applying an across the board service charge of 18 or 20 percent in order to redistribute income between front-of-house and back-of-house positions.

The elimination of tips is a relatively rare business model in the U.S. restaurant sector, but there have been a number of recent, high-profile examples that have accelerated the pace of change. The nationally recognized restauranteur Danny Meyer, who owns several upscale restaurants in New York City (Gramercy Tavern, Union Square Café), announced that all of his New York-based restaurants would go “hospitality included” within a year. Meyer told the New York Times that he specifically cited the need to rebalance the pay scale for kitchen staff after the recent increase in minimum wage for restaurant workers in New York.

Some interview respondents in San Francisco gave unprompted support for this compensation model. Said the manager of multiple fine-dining restaurants in the city:

“If I opened a new restaurant of my own tomorrow, I would 100 percent put everybody on salary. I would charge a flat percentage surcharge, and I would, I’d put everybody on salary. Direct-to-customer employees probably start at $65,000 dollars a year and they cap out at $110,000 and non-direct-to-customer employees probably start at $45,000 and also would likely cap out at $110,000. And you know, they would be eligible for raises annually based on performance, and then two bonus structures a year.”

While the ability to raise prices or add significant surcharges in order to eliminate tipping may be limited to higher priced restaurants–or very profitable establishments–it is clear that rising labor standards in cities like San Francisco and New York are accelerating this trend. But one barrier to a more widespread adoption of this approach is the way payroll taxes are assessed. If a service charge is collected by the employer—rather than the employee in the case of tips—and paid to workers in salary or higher hourly wages, then the employer must pay additional payroll taxes into the unemployment system. Two additional interview subjects cited this added cost as a minor barrier to moving to a tip-less model.

What is most interesting about this recent restructuring of compensation practices is not that it will be immediately adopted throughout the industry, but that it illustrates how alternative business models can be possible, including ones that focus on evening the playing field between front-of-house and back-of-house workers. Such employment practices would reduce wage inequality as well as racial and economic inequality.

—T. William Lester is an Assistant Professor in the Department of City and Regional Planning
at the University of North Carolina-Chapel Hill

Methodology

The methodology employed in the larger research paper (that this issue brief is based on) consists of a set of semi-structured interviews with approximately 15 employers in each case. The interview subjects were restaurant owners, general managers, or other key staff who have direct control or influence over the firm’s human resource strategy. Subjects were solicited from and represent all major restaurant market segments (e.g. family-style, casual fine dining, and fine dining), offering a range of observations according to price point and revenue. Interview subjects were initially solicited via a web-based survey inquiring about the willingness of survey participants to participate in a 45-minute interview. Additional interview subjects were solicited through phone calls and in-person requests by the investigator and a graduate student researcher during business hours. Subjects were compensated with a $50 gift card for participation. All interviews were recorded on digital media and transcribed for subsequent analysis.

Download File
Inside-monopsony Download full working paper

This qualitative data collection is used to analyze how employers actively and uniquely construct their labor market practices in the face of institutional constraints such as wage mandates and prevailing industry norms. The interviews go beyond the survey results and seek to ascertain why a reported practice, such as investments in training, were chosen. In addition to the interviews, a web-based survey was conducted between July 1st and August 31st 2014 and collected a total of 104 valid responses. The survey consisted of 15 questions and was intended to gather detailed information on wage levels by occupation, training provided, skill requirements and educational attainment of workers. In addition the survey gathered background information on each restaurant such as market segment, average entrée price level, and number of seats available.

How the student debt crisis affects African Americans and Latinos

Posted on by David Hudson

A student hugs family during the procession at commencement ceremonies at Hampton University in Hampton, Virginia. (AP Photo/Steve Helber)

Our first Mapping Student Debt interactive released this past December revealed a striking negative relationship between income and delinquency across zip codes. Not surprisingly, we found that higher levels of income are associated with fewer problems with student loan delinquency. In this second installment of the Mapping Student Debt project, we document that the geography of delinquency is highly racialized.

Zip codes with higher shares of African Americans or Latinos show much higher delinquency. What’s more, our analysis finds that among minority student borrowers, those most adversely affected are the middle class—those who have taken out debt to go to college but who haven’t been able to find jobs or don’t have sufficient family wealth to pay it back.

Delinquency disproportionately affects minority communities

Our findings are stark. They show the strong relationship between a zip code’s minority population and its delinquency rate at both the city and national levels. In the Washington, D.C. metro region, for example, zip codes in the northeastern part of the District of Columbia and east of the Anacostia River and adjacent suburbs—all of which have the largest shares of African Americans and Latinos—also have delinquency rates that range from somewhat high to extremely high. The same pattern holds in Los Angeles, where areas with large African American or Latino populations, such as Compton, Linwood, and Huntington Park, are also where delinquency is highest. (See Figure 1.)

Figure 1

At the national level, too, we find that zip codes with higher shares of African Americans or Latinos have much higher delinquency rates. This relationship suggests that minority communities disproportionately suffer from student loan delinquency. (See Figures 2 and 3.)

Figure 2

Figure 3

Controlling for income

The geography of race and of income are similar, so a natural question that arises is whether race has an independent effect on delinquency, and, if so, what is it? The answer turns out to depend on income, but not in an obvious way.

In order to investigate the effect of race independent of income, we first ranked zip codes by median income and divided them into 100 groups of equal size. Within each of these income groups, median income levels are nearly identical, which means we can look at how delinquency varies across zip codes with different shares of African Americans or Latinos but otherwise very similar income levels. In Figures 4 and 5, we plot how an increase in the minority share of the zip code population changes the rate of delinquency. Points above zero mean that as a zip code’s minority population increases (relative to zip codes with a similar income), so does the share of delinquent loans in that zip code. Conversely, a negative number implies that zip codes with larger minority populations have lower loan delinquency rates.

Figure 4

Figure 5

In both Figures 4 and 5, the positive correlation between the share of minorities in a zip code and loan delinquency rates is highest for the middle of the income distribution. Among zip codes with a median income of about $20,000, for example, zip codes with a large share of Latinos and those without have approximately the same rates of delinquency. But among zip codes with a median income of around $60,000, those with large Latino share have much higher rates of loan delinquency than those without.

We see a similar pattern for the share of African American in zip codes, and there the effect is even more pronounced. For zip codes with median incomes above $60,000, the effect of race on delinquency either stays roughly constant or declines slightly.

Another interesting feature of the data is that among the zip codes with the poorest populations, an increase in the share of African Americans is associated with a decline in delinquency rate, whereas the share of the Latino population has no impact on delinquency. We do not think our data are rich enough to meaningfully address this particular fact, which merits further research.

The role of race in student loan delinquency

Minority populations disproportionately suffer from high delinquency, and, within minority populations, the middle class seems most adversely affected. What can we make of these findings? We believe that these two facts reflect the impact of structural racism in the U.S. higher education system, credit and labor markets, and distribution of wealth.

African Americans and Latinos are, on average, less likely than white students to complete college once they start. According to the National Center for Education Statistics, in 2013 roughly 57 percent of recent African American high school graduates and 60 percent of recent Latino high school graduates were enrolled in college compared to 69 percent of white students. Yet the National Center for Education Statistics reports that for the 2005 starting cohort of college students, about 21 percent of African Americans and 29 percent of Hispanics complete a four-year institution within four years compared to a four-year completion rate of 42 percent for white students.

The college enrollment gap between whites and minorities is narrowing, but the college completion gap is not. One likely explanation for higher student loan delinquency among African Americans and Latinos is that the borrowing is concentrated among those who either attended for-profit or other non-traditional institutions or who dropped out—exactly the population at the margin of attending college in the first place. Furthermore, we know that higher education is racially segregated, with minorities less likely to attend—or even consider applying to—selective institutions.

Even after controlling for key risk factors, African Americans and Latinos are disproportionately served by high-cost credit providers who provide less generous terms and more onerous repayment requirements, implying that discrimination occurs through market segmentation and sorting.

Another explanation for high delinquency rates among minorities is that after college, graduates still confront significant discrimination in labor markets, with minority applicants less likely to get job offers, even after factors such as education are taken into account. Even minority students who successfully complete college suffer from higher unemployment rates and lower earnings than their white counterparts. These disadvantages extend across college majors, occupations, and the type of higher education institution that these recent graduates attended. In combination, these factors leave minority students and their families substantially more vulnerable to delinquency than comparably situated white students and their families.

A closely related issue is that, holding income constant, African American and Latino households have substantially lower levels of wealth than do white households, including financial assets that can act as a buffer against student loan delinquency in the event of job loss or some other misfortune.

Middle-class minorities are hurt the most by student loan delinquency

Why are middle-class African American and Latino students and their families the most adversely affected by student debt delinquency? The poorest minority populations generally lack access to any kind of formal credit, instead relying on payday lending and other types of informal credit access. This means that they cannot be delinquent by our measure that is based on credit reports. What’s more, they rarely go to college, so in many cases, they do not acquire student debt.

The housing crisis revealed a similar dynamic in the late 2000s. The poorest minority households lost relatively little wealth because they didn’t have any to begin with, whereas somewhat richer minority households were among the biggest losers from the Great Recession. That was because they earned enough to have bought a house under the relatively generous terms available before the housing market crash, but then they were more likely to lose their jobs and less likely to have any cushion of family wealth. It is out of these very dynamics that persistent, multi-generational racial wealth gaps are born. And it seems likely that student debt is on the same path now—a signpost of relative economic success among minorities, but also a threat. Many young people of color have gone into debt to ascend to the middle class, and been supported by their families to do so, yet it’s not having the intended effect.

These data tell us that at least with respect to longstanding group and individual income and wealth gaps between minorities and the overall population, debt-financed higher education is not the solution, and may even be contributing to the problem. The fact that, among minorities, the middle class is most strongly affected implies the problem is structural racism, not poverty. Any solution to the student debt crisis has to recognize that.

Methodology

This geographic analysis uses two primary datasets: credit reporting data on student debt from Experian and income data from the American Community Survey.

The Experian data includes eight key student debt variables (see Figure 6) aggregated from household-level microdata to the zip code level. The underlying household data are a snapshot of the entire U.S. population at a single point in time—in this case, the autumn of 2015.

Figure 6

There are a number of caveats regarding the Experian data file that have guided our methodology for constructing variables and analyzing results:

  • The universe of households contains only those with “any type of credit” and which, therefore, have a credit report. Relative to the population as a whole, this likely excludes the poorest households without any official credit access whatsoever.
  • It is unclear how Experian constructs “households” since credit reports pertain to an individual’s credit history.
  • If the same student loan has more than one signatory, then the loan may be assigned to multiple households and hence to multiple zip codes or even counted more than once within the same household.
  • Experian claims that the universe of their geographically-aggregated data is all households with credit, but the levels of the data on loan balance and delinquency are more consistent with the idea that the universe is only households that have student loans. In other words, Experian claims their data include households that have credit but no outstanding student loans, but if that is the case the reported levels for average delinquency are much higher than other sources would suggest. Average delinquency rates, however, are comparable to reliable outside estimates if interpreted as delinquency among only those households with student debt.

For these reasons, we do not report any student loan data in rate amounts. Instead, we have used the Experian variable to construct an analog to relative delinquency.

To create the delinquency variable, we calculate a “delinquency rate” for each zip code by dividing the average number of student loans that are delinquent by 90 or more days per household by the average number of outstanding loans per household. Then, after winsorizing the top 1 percent of observations to the 99th percentile value, we project the “delinquency rate” onto a scale that ranges from 0 to 10.

For user-friendliness, we assign the student debt scale variable a qualitative category. If the delinquency reads “very low,” for example, it corresponds to a scale level between 0.067 and 0.091. Figure 7 summarizes the relationship between the delinquency scale variable’s levels and its qualitative descriptions.

Figure 7

Next, we merge zip code-level household median income with data from the 2013 American Community Survey on the share of African Americans and Latinos in those zip codes along with our imputed scaled delinquency variable in order to construct choropleth maps.

The actual map uses two different techniques to display the variables on a choropleth scale. For delinquency, we created 10 quantiles (or equal counts) to account for the right-skewed data. And for the two minority share variables, we used 10 jenks (or natural breaks in the data) to assign the color scale. Higher numbers and darker shading correspond to higher shares of outstanding loans that are delinquent by 90 or more days in the previous 24 months and higher shares of African Americans and Latinos in a zip code.

Additional reading

Adam Looney and Constantine Yannelis, “A crisis in student loans? How changes in the characteristics of borrowers and in the institutions they attended contributed to rising loan defaults.”

Benjamin Backes, Harry J. Holzer, and Erin Dunlop Velez, “Is It Worth It? Postsecondary Education and Labor Market Outcomes for the Disadvantaged.”

Caroline M. Hoxby and Sarah Turner, “What High-Achieving Low-Income Students Know about College,” American Economic Review.

Devah Pager, Bruce Western, and Bart Bonikowski, “Discrimination in a Low-Wage Labor Market: A Field Experiment.”

Fenaba R. Addo, Jason N. Houle, and Daniel Simon, “Young, Black, and (Still) in the Red: Parental Wealth, Race, and Student Loan Debt,” Race and Social Problems.

Janelle Jones and John Schmitt, “A College Degree is No Guarantee.”

Jeffrey P. Thompson and Gustavo A. Suarez, “Exploring the Racial Wealth Gap Using the Survey of Consumer Finances.”

Jess Bricker and others, “Changes in U.S. Family Finances from 2010 to 2013: Evidence from the Survey of Consumer Finances.”

John Schmitt and Heather Boushey, “The College Conundrum: Why the Benefits of a College Education May Not Be So Clear, Especially to Men.”

Joshua Angrist, David Autor, Sally Hudson, and Amanda Pallais, “Leveling Up: Early Results from a Randomized Evaluation of Post-Secondary Aid.”

Martha J. Bailey and Susan M. Dynarski, “Gains and Gaps: Changing Inequality in U.S. College Entry and Completion.”

Marianne Bertrand and Sendhil Mullainathan, “Are Emily and Greg More Employable Than Lakisha and Jamal? A Field Experiment on Labor Market Discrimination.”

Neil Bhutta, Paige Marta Skiba, and Jeremy Tobacman, “Payday Loan Choices and Consequences.”

Patrick Bayer, Fernando Ferreira, and Stephen L. Ross, “Race, Ethnicity and High-Cost Mortgage Lending.”

Rakesh Kochhar and Richard Fry, “Wealth inequality has widened along racial, ethnic lines since end of Great Recession.”

Stephanie Chapman, “Student Loans and the Labor Market: Evidence from Merit Aid Programs.”

Interactive: A new look at who earns what in the United States

Posted on by Austin Clemens

Discussions of how wages vary for different workers are often abstract. Most analyses focus on just wage levels, paying little attention to who the workers are, what they do, or other factors—such as gender and race—that play a critical role in shaping the wage distribution. This interactive offers a new and more concrete look at the wage distribution in the United States, using data from the Current Population Survey to reveal how a worker’s wage is connected to their job as well as their gender and race.

The interactive divides the U.S. workforce into “deciles”—10 groups of equal size—ordered from least paid to highest paid. The simplest version of the interactive shows the top hourly wage paid within each tenth of the workforce. The bottom tenth of workers all make less than $8.76 per hour. The next tenth of workers make more than $8.76 but less than $10.37, and so on until the top tenth, where we report only the minimum pay required to enter the tenth decile.

To give an idea of the kinds of workers in each wage group, we list the three most common occupations within each decile. These occupations are a broad description of the jobs that workers perform (cook, nurse, or lawyer, for example). To give a sense of how much different jobs pay within each wage group, we also list each occupation’s wage ranges.

The interactive further lets you look separately at wages and occupations across gender (men and women) and race (whites, African Americans, Hispanics/Latinos, and Asians) and see comparisons between these demographic groups.

How it works

To begin, let’s take a look at just the distribution for “all workers.” Here, you’ll learn that the lowest-paid workers (most commonly cashiers, waiters and waitresses, and retail salespersons) earn less than $8.76 per hour. Median-wage workers (such as first-line retail supervisors, drivers, and secretaries and administrative assistants) earn between $15.00 and $17.71 per hour. The highest-paid workers (managers, chief executives, and software developers, for example) make at least $42.13 per hour (and, in our data, up to well over $300 per hour).

You can also get more detail on the common occupations by clicking on a specific decile. When selecting the bottom decile, for instance, you’ll see that the three most common occupations in this lowest wage group—cashiers, waiters and waitresses, and retail salespersons—all have a wage range that rises above the bottom group. This is because the pay varies within each occupation, not just across occupations. Take the retail sales workers, for example. The lowest-paid retail workers earn $8.00 per hour, while the highest-paid salespersons can earn $28.00 per hour. Some occupations, such as retail salespersons, can span multiple wage groups. In fact, retail salespersons show up again as one of the most common occupations in the second and third deciles, as well.

Now, suppose you’re interested in seeing the wage distribution for women. The left-most dropdown menu in the interactive allows you to select “women,” or any other demographic group of interest, to find out what different wage groups get paid, what the most common occupations are in each wage group, and how pay varies across and within occupations.

You can even compare two demographic groups to each other. If you want to contrast the highest-paid white worker’s occupations to the highest-paid black worker’s occupations, for example, you can select “white” and “African American” respectively from each dropdown menu and click on the top decile to see just how much occupations and pay differ between the two groups.

Eager to start exploring the distribution all over again? Just hit the reset button at the top left of the interactive.

Methodology

The data behind this interactive is derived from the Center for Economic and Policy Research extracts of the Current Population Survey Outgoing Rotation Group. The CPS provides data on hourly wages, three-digit occupation categories, gender, and race and ethnicity, all of which were used to determine three key components:
1. A wage decile distribution

2. The top three occupations in each wage decile

3. The 10th, 50th, and 90th percentile hourly wage for each top occupation across the distribution

Each of these components is produced for all people, men, women, whites, African Americans, Hispanics/Latinos, and Asians, allowing us to compare the results across different demographic groups.

First, to ensure we had a sufficiently large sample size for all the demographic groups, we pooled together the 2011, 2012, 2013, and 2014 CPS survey results. We further limited our sample to working-age persons (age 16 to 64). Next, we assigned a wage decile to each observation in the dataset based on their real (2014 dollars) hourly wages; this hourly wage variable includes earnings from overtime work. Using the maximum hourly wage in each decile, we constructed a wage threshold distribution. We use wage thresholds because the CPS does not capture earnings at the very top well. Using the average of the wage deciles would, therefore, be misleading for the top wage decile.

In order to determine the top three occupations in each wage decile, we relied on a qualitative approach. To find the share of people in each three-digit occupation group by decile, we used a weighted frequency tabulation. We then manually sorted through these occupational shares to ascertain the top three largest occupations for each decile. Finally, for each occupation across the distribution, we calculated the 10th, 50th, and 90th percentile real hourly wage; these measures allow us to see the wage range of occupations that span multiple deciles.

What’s aging got to do with the U.S. labor market recovery?

Posted on by Bridget Ansel

The U.S. economy added 211,000 jobs in November, according to employment data released today by the U.S. Bureau of Labor Statistics. The employed share of the nation’s population in their prime working years (ages 25 to 54) jumped to 77.4 percent but still remains below healthy levels. One oft-cited explanation for the low overall employment rate—the aging of the U.S. population—does not satisfactorily explain the slow labor market recovery.

Hourly wages for private-sector workers in the United States grew by 2.3 percent at an annual rate last month—well below a goal of at least 3.5 percent, which is consistent with long-term productivity growth and inflation targets. As has been true every month this year, production and non-supervisory workers saw smaller pay increases, at an annual rate of 2.0 percent, suggesting rising wage inequality. Wages grew somewhat faster in retail industries (2.7 percent) and the leisure and hospitality sector (2.6 percent), which includes restaurants. Retail-sector and restaurant wage increases this year are primarily a consequence of minimum wage increases, political pressure, and an improving labor market.

Wage increases in restaurants have been running faster than price increases in that sector, as some commentators have noted. But even if pay raises for restaurant workers are fully passed onto consumers, we should still expect the price inflation resulting from these wage changes to be smaller than underlying pay increases because labor is not the only cost of business. In restaurants, labor makes up roughly one-third of total business costs, so a 10.0 percent increase in wages would only result in a 3.3 percent increase in prices. The final price increase could be closer to just 1.0 percent if the 10.0 percent wage increase was only for minimum-wage workers, since less than 30 percent of workers in the restaurant industry earn near the minimum wage.

Turning to employment, establishment survey data from the Bureau of Labor Statistics showed that, with the addition of 211,000 jobs in November, the U.S. economy has added an average of about 218,000 jobs over the past three months. Monthly job growth this year has averaged about 210,000, slower than the average of 260,000 last year. Most of the job growth last month occurred in construction (46,000) restaurants (33,000), and health care and social assistance (32,000). The overall unemployment rate remained at 5.0 percent, after falling by 0.7 percentage points so far this year. In a notable change last month, the employed share of the prime-age population (ages 25 to 54) rose 0.2 percentage points to 77.4 percent, but this is the first time the rate has moved above its value of 77.3 percent in February. In the absence of much higher employment rates for the prime-age population, we are unlikely to see sustained, broad-based wage growth.

Some argue that the failure of employment rates to return to their pre-recession levels despite the long-term drop in the unemployment rate is due to demographic changes, not the weakness of the recovery. As Baby Boomers age and retire, they are naturally less likely to be employed, pushing down the overall employment rate. But demographic changes are not a plausible explanation for the weak employment recovery following the Great Recession. Although the aging of the U.S. workforce does mechanically lower the overall employment rate, demographic changes play a relatively inconsequential role for the employment of prime-age individuals—a group that has also seen only a partial rebound in employment during the recovery. (See Figure 1.)

Figure 1

A simple way to determine how much demographic changes have lowered employment is to assume the U.S. population today has the same age, sex, and racial structure as it did in 2000, and to use each of these groups’ current employment rates. The overall employment-to-population ratio for those aged 16 and older fell from its high point of 64.5 percent in 2000 to 59.2 percent last year. Had demographics not changed since 2000, the rate would have fallen to 61.3 percent, so demographic changes account for about 40.1 percent of the fall in the overall employment rate.

Just as for the overall workforce, the unemployment rate for prime-age workers also fell sharply during the recovery. In late 2009, the unemployment rate for prime-age workers hit 9.0 percent but has since decreased by more than half to 4.3 percent as of last month. Even with this fall in unemployment, however, the employed share of prime-age workers (77.4 percent) is currently still below its trough of 78.6 percent after the 2001 recession.

Yet demographic changes have little to do with the fall in the employed share of the prime-age population. Aging is the most important demographic change for employment, and employment rates do not vary as much by age group for those in their prime working years. In fact, the prime-age employment rate in 2014 would have been less than 0.7 percentage points higher in the absence of changes in the population structure despite falling by nearly 4.8 percentage points since 2000. This means only 13.6 percent of the fall in the prime-age employment rate since 2000 is due to demographic changes.

Prime-age workers stopped participating in the labor force during a weak labor market, indicating that there is plenty of room for the labor market to improve before truly tightening.

Must-Read: Robert Lynch and Kavya Vaghul: Benefits and Costs of Investing in Early Childhood Education

Posted on by Brad DeLong

Must-Read: Robert Lynch and Kavya Vaghul: Benefits and Costs of Investing in Early Childhood Education: “A case for public investment in either a targeted or a universal prekindergarten program…

…can be made with the best policy depending in part on whether a higher value is placed on the ratio of benefits to costs (which are higher for a targeted program) or the total net benefits (which are higher for a universal program)…. If public funds are limited, a targeted program may be more attractive…. If a larger priority is placed on narrowing the achievement gap between children from low-income and upper-income families than on promoting economic growth, then the targeted program may be more effective…. A universal prekindergarten program… is likely to generate greater future economic growth…. In addition, children who are not eligible for a targeted program can benefit from high-quality pre-K, and targeted programs frequently fail to reach many of the children they are designed to serve…