The Atlanta Fed's macroblog provides commentary and analysis on economic topics including monetary policy, macroeconomic developments, inflation, labor economics, and financial issues.

Authors for macroblog are Dave Altig, John Robertson, and other Atlanta Fed economists and researchers.

January 16, 2020

Do Higher Wages Mean Higher Standards of Living?

Editor's note: We have updated macroblog's location on our website, although archival posts will remain at their original location. Readers who use RSS should update their feed's URL to https://www.frbatlanta.org/rss/macroblog.aspx. Also, we are implementing a new commenting system for posts.

A recent macroblog post used Atlanta Fed Wage Growth Tracker data to observe that the hourly wage of the lowest-paid workers has rebounded in recent years after declining for a decade. The chart below depicts this finding, showing the median hourly wage of the lowest-paid 25 percent of workers in the Tracker sample relative to the median for all workers.

Rate of Employer-to-Employer Transitions (12-month moving average)

Moreover, the post showed that this recovery was not just a story about states and localities increasing their minimum wages. It also appears that there has been a significant tightening in the labor market for unskilled or low-skilled jobs.

Taken at face value, this is good news for workers employed in low-wage jobs. But here's the rub: the median wage in the first quartile is still low—$11.50 in 2019, or 55 percent of the overall median wage. Moreover, these are hourly wages before taxes and transfers (we'll get back to this shortly). They don't represent what is happening to these workers' ability to make ends meet, which depends crucially on income after taxes and transfers.

For households at the bottom of the income distribution, means-tested transfers can play an especially important role. Means-tested transfers—cash payments and in-kind benefits from federal, state, and local governments designed to assist individuals and families with low incomes and few assets to meet their basic living needs—represent about 70 percent of income before taxes and transfers for households in the bottom quintile of the income distribution, according to a recent report by the Congressional Budget Office. However, the size of the transfers tends to decrease as earnings increase, and they stop altogether when a worker exceeds income- and asset-eligibility thresholds.

The interaction between changes in earnings and various means-tested public assistance programs is an important public policy issue, and it is one that staff at the Atlanta Fed are studying. In a March 2019 macroblog post, David Altig and Laurence Kotlikoff reported that this interaction results in low-income households facing a higher median effective marginal tax rate than high-income households. For low-income households with children, this effect can be especially severe because the presence of children increases the value of programs such as the Supplemental Nutrition Assistance Program (or SNAP, formerly known as the food stamp program) and the likelihood of enrollment in additional programs such as federally subsidized child care. (You can read further research on the effective or implicit marginal tax rates of low-income households at Congressional Budget Office (2016), Romich and Hill (2018), and Chien and Macartney (2019).)

To illustrate the point, the Atlanta Fed team studied the case of a hypothetical single mother with two young children who works in a near-minimum-wage, full-time job and whose basic living expenses are helped by various transfer programs. One avenue to improving her family's standard of living is if she were to return to school and pursue a higher-paying career as a nurse. Over the long term, the net gains from education and career advancement are unambiguous. However, the Atlanta Fed's analysis shows that as long as her children still require care, the reduction in payments from various benefit programs could partially or even completely offset the gains. Look for an Atlanta Fed paper discussing this very real dilemma coming soon on the Bank's Economic Mobility and Resilience webpage.

What do findings like this mean for interpreting the Wage Growth Tracker's evidence that people in the bottom part of the wage distribution are experiencing relatively larger wage gains? Perhaps there is a bit less to celebrate than meets the eye. Around 46 percent of these individuals are in households with children. To the extent that they also participate in means-tested public assistance programs, the relative increase in their family's standard of living could be much less than the size of their pay raise would suggest.

January 16, 2020 in Employment, Labor Markets, Wage Growth | Permalink


Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

December 16, 2019

Faster Wage Growth for the Lowest-Paid Workers

On November 25, Fed chair Jay Powell gave a speech titled "Building on the Gains from the Long Expansion," in which he observed that

Recent years' data paint a hopeful picture of more people in their prime years in the workforce and wages rising for low- and middle-income workers.

In making this point, Chair Powell used a cut of the Atlanta Fed's Wage Growth Tracker that looks at the median annual wage growth of workers in the lowest 25 percent of the wage distribution. As the following chart shows, the lowest-paid workers have been experiencing higher median wage growth (the blue line) in the last few years than workers overall (the green line). This reverses the pattern seen in the wake of the Great Recession, when median wage growth for lower-paid workers slowed by more than for workers overall.

Chart 1: Median Wage Growth

The faster median wage growth for lower-wage workers shown in chart above has also translated into an increase in the relative median wage level of these workers. To see this, the following chart shows the median wage level for those in the lowest wage quartile relative to the median for all workers in the Wage Growth Tracker dataset.

Chart 2: Relative Median Wage: Lowest Wage Quartile

The chart shows that for workers in lower-wage jobs, their relative median wage over the 2000s has deteriorated, and that erosion has reversed course only in the last few years. This reversal may reflect increasing tightness of the labor market for lower-wage jobs relative to other jobs over the last few years. The challenge of filling jobs requiring few skills is something we have been hearing about a lot recently from the businesses we talk to (for example, see here), and this sort of challenge could be behind higher wages for those workers. However, several state and local governments have increased the minimum wage in recent years, which would also push up the relative pay for those in the lowest-paid jobs.

Are the observations in the previous chart solely attributable to minimum wage increases? To get some idea, the next chart contrasts the relative median wage in states that increased their minimum wage at some point between 2014 and 2019 to those that did not. The blue line is the relative median wage of the lowest quartile in the 28 states that increased their minimum wage (23 states introduced new minimum wage levels, and five implemented increases legislated before 2014), and the green line is relative median wage for the states that did not increase their minimum wage.

Chart 3: Relative Median Wage: Lowest Wage Quartile

We would expect to see a rise in the relative median wage in the states that raised their minimum wage, and indeed we do. For the group of states that increased minimum wages (the blue line), the relative median wage is now closer to that of states that did not increase their minimum wage (the green line). Interestingly, though, even in the "no increase" states, the relative median wage has improved, suggesting that the increased tightness of labor markets, or some other factor than hikes in state minimum wages, is playing a role in pushing up the pay for those in lower-wage jobs. Consistent with the message of Chair Powell's speech, the good news is that there is scope to continue to build on the gains from the long and ongoing expansion for workers at the bottom end of the wage distribution.

December 16, 2019 in Employment, Wage Growth | Permalink


Blue counties account for around two-thirds of GDP. Higher minimum wages in blue counties can increase demand for labor in red counties, which should increase the price of labor in red counties too.

Posted by: Daniel Tobias | December 23, 2019 at 09:58 AM

Chart 3 indicates that relative median wages in states with a minimum hike have reached a new high, while in states that didn't raise minimum wages, median wages remain below pre-GFC levels. It appears that differences in overall pay structures may be affecting the differences in relative wages.

Posted by: jammer1297 | January 03, 2020 at 11:31 AM

Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

November 25, 2019

Is Job Switching on the Decline?

Here's a puzzle. Unemployment is at a historically low level, yet nominal wage growth is not even back to prerecession levels (see, for example, the Atlanta Fed's own Wage Growth Tracker). Why is wage growth not higher if the labor market is so tight? A recent article in the Wall Street Journal posited that the low rate of job-market churn likely explains slow wage growth. Switching jobs is typically lucrative because it tends to be going to a job that better uses the person's skills and hence offers higher pay. Job switchers can also help improve the bargaining position of job-stayers by inducing employers to pay more to retain them.

But is the job-switching rate really lower? A paper that Shigeru Fujita, Guiseppe Moscarini, and Fabien Postel-Vinay presented at the Atlanta Fed's 10th annual employment conference looked at a commonly used measure of employer-to-employer transitions. That measure, developed by Fed economists Bruce Fallick and Charles Fleischman in 2004, uses data from the Current Population Survey (CPS) on whether a person says that he or she has the same employer this month as last month. Job switchers are those reporting having a different employer. As the following chart shows, the Fallick and Fleischman data (the yellow line) support the Wall Street Journal story that the rate of job switching is much lower than it used to be.

Rate of Employer-to-Employer Transitions (12-month moving average)

Source Fujita et al. (2019)

However, Fujita and his coauthors discovered a potential problem with these data, noting that the CPS doesn't ask the same-employer question of all surveyed people who were employed in the prior month. Importantly, the incidence of missing answers has increased dramatically since the 2006, as the following chart shows.

Missing Answers to Question on Employer-to-Employer Transitions

Source Fujita et al. (2019)

If these missing answers were merely randomly distributed among job switchers and job stayers, then it wouldn't matter much for the Fallick-Fleischman measure. But Fujita et al. found that the missing answers were disproportionately from people who look more like job switchers in terms of observable characteristics such as age, marital status, education, industry, and occupation—making it likely that the Fallick-Fleischman measure undercounts job switchers.

The researchers developed a statistical adjustment to the Fallick-Fleischman measure to account for this bias (the blue line label our series in the following chart—this is the same as the first chart) that tells a somewhat different story than the original measure (yellow line).

Rate of Employer-to-Employer Transitions (12-month moving average)

Source Fujita et al. (2019)

In particular, the adjusted job-switching rate is only moderately lower than it was 20 years ago and has fully recovered from the decline experienced during the Great Recession. Although a decline in job switching might be a factor in the story behind low wage growth, based on this adjusted measure it doesn't seem like the dominant factor. The low wage growth puzzle remains a puzzle.

November 25, 2019 in Employment, Wage Growth | Permalink


Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

September 26, 2019

Digging into Older Americans’ Flat Participation Rate

The rate of labor force participation (LFP) by people age 55 and over had been rising during the decade leading up to the Great Recession. But more recently, as the following chart shows, the share of older individuals engaged in the labor market has barely budged. (We should note that for all the charts in this post, the data are from the Current Population Survey from the U.S. Bureau of Labor Statistics and the authors' calculations.) What behavioral and demographic factors could be underlying these trends?

We can use the Atlanta Fed's Labor Force Dynamics web page to explore why the 55-and-over population has a relatively flat LFP rate of late. One factor working to depress older Americans' LFP rate is the increase in the share of those 55 and older who are retired: from 46.9 percent in the second quarter of 2014 to 47.8 percent in the second quarter of 2019 (see the chart).

This change in the overall retirement rate of the 55-plus cohort is actually the result of the union of two opposite forces—one demographic and one behavioral. From the perspective of demographics, a greater share of the population is reaching the typical retirement age threshold of 65. For instance, the share of people aged 65 and older has increased from 18 percent in the second quarter of 2014 to 20.3 percent in the second quarter of 2019 (see the chart).

This demographic shift is important because the retirement rate is much higher for those 65 and older compared to those from 60 to 64 years old. For example, the retirement rate in 2019 for those 60–64 is around 25 percent, but it's 56 percent for those aged 65–69, and it's 81 percent for those 70 and older (see the chart).

However, the retirement rate among those 60–64 and those 65–69 has also declined in recent years. This change in behavior within older age groups is partly offsetting the downward pressure on participation coming from having a larger share of population over 65.

A second factor that has worked against the overall retirement effect and helped push up the LFP rate of the 55-plus population has been a decline in the share of older individuals saying they are not participating in the labor force because of disability or illness. This rate has decreased from 8.6 percent in the second quarter of 2014 to 8.1 percent in the second quarter of 2019 (see the chart).

As the chart below shows, this shift is largely a demographic effect. Nonparticipation because of disability/illness drops off significantly as people turn 65, so the fact that a greater share of people 55 and older are now 65 and older decreases the overall share of nonparticipation for disability/illness reasons as well. However, it's important to note that a lower disability/illness rate for those 65 and older doesn't mean that this older population is actually less disabled or sick. It is more that individuals in the data fall into only one category, and those 65 and older are more likely to say they are retired than to say they are disabled or ill.

A third force helping push up the LFP rate is the decline in the share of older individuals on the sidelines of the labor market—those who are not participating but nonetheless want a job—from 1.8 percent in the second quarter of 2014 to 1.6 percent in the second quarter of 2019 (see the chart).

This “shadow labor force” effect on participation is mostly the result of changes in behavior (that is, a reduced propensity to remain on the sidelines within age groups) rather than from demographic changes because the rate has declined within age groups, whereas the levels are roughly similar across age groups (see the chart).

To put these various pieces together, the following chart summarizes the overall contributions of demographic and behavioral forces on the LFP rate among the 55-plus population between the second quarter of 2014 and the second quarter of 2019. The chart shows that the contributions stemming from changes in demographics and behavior have largely offset each other.

As I've already described, the biggest demographic effects come from having more people at an age with two specific characteristics: a relatively high rate of nonparticipation because of retirement and a relatively low rate of nonparticipation because of disability or illness. But as the following chart shows, from the second quarter of 2014 to the second quarter of 2109, the retirement demographic dominates, so the overall demographic LFP effect is a negative one.

Conversely, the largest behavioral shifts are, first, a lower propensity within older age groups to stay out of the labor force because of retirement and, second, a lower share of older people wanting a job but not looking for one. As the following chart shows, from the second quarter of 2014 to the second quarter of 2019, these behavioral changes combine to push up LFP by enough to nearly offset the demographic shifts.

It seems reasonable to presume that the aging population will continue to be an important source of downward pressure on the LFP rate of older Americans over the next few years. What will be telling is whether or not the behavioral shifts we have seen will persist as strongly as they have up to this point and continue to provide a countervailing positive influence on participation.

The tools on the Atlanta Fed's Labor Force Dynamics web page are very useful for understanding what's behind changes in LFP for different demographic groups. In addition to cuts for different age groups, you can look at differences between men and women and among different racial and ethnic groups and levels of educational attainment. You can download the chart data—the charts in this blog are downloaded images from the web page—and even download the underlying Current Population Survey microdata from the Kansas City Fed's CADRE web page if you want to create your own cuts. Check it out!

September 26, 2019 in Employment | Permalink


Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

August 05, 2019

What the Wage Growth of Hourly Workers Is Telling Us

The Atlanta Fed's Wage Growth Tracker has shown an uptick during the past several months. The 12-month average reached 3.7 percent in June, up from 3.2 percent last year. But in 2016, it depicted acceleration that eventually reversed course. So is this recent increase real or illusory?

Although using a 12-month average quiets much of the noise in the monthly data, it is possible that the smoothed series still may exhibit some unwanted variation due to the way the Wage Growth Tracker is constructed. For example, how the monthly Current Population Survey reports individual earnings might be a factor introducing unwanted noise into the Tracker. Specifically, some people directly report their hourly rate of pay, and some report their earnings in terms of an amount per week, per month, or per year.

Relative to those paid an hourly rate, there are at least a couple of reasons why using the earnings of nonhourly workers might introduce additional variability into the Wage Growth Tracker's overall estimate of wage growth. First, reported nonhourly earnings include base pay as well as any overtime pay, tips, and commissions earned, and hence can vary over time even if the base rate of pay didn't change. For a worker paid at an hourly rate, reported earnings exclude overtime pay, tips, and commissions and so are not subject to this source of variation. Second, the method we use to convert nonhourly earnings to an hourly rate is likely subject to some margin of error since it involves using the person's recollection of how many hours they usually work. These two factors suggest that the earnings of workers paid at an hourly rate might be a somewhat cleaner measure of hourly earnings.

To investigate whether this distinction actually matters in practice, we created the following chart comparing the 12-month average Wage Growth Tracker since 2015 (depicted in the green line) with a version that uses only the earnings of those paid at an hourly rate (blue line).

Median wage growth, 12-month moving average

As the chart shows, the 12-month average of median wage growth for hourly workers generally tracks the overall series—both series are about a percentage point higher than at the beginning of 2015. However, the hourly series is a bit less variable, making the recent uptick in wage growth more noticeable in the hourly series than in the overall series. This observation suggests that as we monitor shifts in wage pressure, the hourly series could complement the overall series nicely. Versions of the Wage Growth Tracker series for both hourly and nonhourly workers are now available on the Wage Growth Tracker page of the Atlanta Fed's website.

If you would like to use the Wage Growth Tracker's underlying microdata to create your own versions (or to conduct other analysis), follow this link to explore the data on the Atlanta Fed's website. See this macroblog post, "Making Analysis of the Current Population Survey Easier," from my colleague Ellyn Terry to learn more about using this dataset.

August 5, 2019 in Employment, Labor Markets, Wage Growth | Permalink


Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

May 06, 2019

Improving Labor Force Participation

Without question, the U.S. labor market has tightened a lot over the last few years. But a shifting trend in labor force participation—and especially a rise in the propensity to seek employment by those in their prime working years—seems to be relieving some labor market pressure.

From the first quarter of 2015 to the first quarter of 2019, the labor force participation (LFP) rate among prime-age workers (those between 25 and 54 years old) increased by about 1.5 percentage points (see the chart below), adding about 2 million workers more than if the participation rate had not increased.

macroblog - May 6, 2019 - Chart 1: Labor Force Participation Rate: 25-54 years

Changes in the distribution of the prime-age population in terms of age, education, and race/ethnicity toward groups with higher participation rates and away from groups with lower rates accounts for about a third of the rise in the overall prime-age LFP rate. The other two thirds can be pinned on an increase in LFP rates within demographic groups—what we call "behavioral" effects.

Of the increased participation behavior within demographic groups, there has been a decline in the share of the prime-age population that say they want a job but are not actively looking for work at the moment. We refer to these individuals as the "shadow labor force" because even though they are not in the labor force this month, they have a relatively high propensity to have a job next month. Second, there's been a decline in the share of the prime-age population that are not participating because they are too sick or disabled to work. The contribution of the change in behavior in these two categories (as well as several others from the first quarter of 2015 to the first quarter of 2019) are shown in the following chart, which is taken from the Atlanta Fed's Labor Force Participation Dynamics tool.

macroblog - May 6, 2019 - Chart 2: Contributions to Change Due to Behavior Q1 2015-Q1 2019

In contrast, consider the period from the first quarter of 2008 through the first quarter of 2015, a time when the rate of prime-age LFP declined by almost 2 percentage points. During that period, even though slow-moving demographic changes were putting modest upward pressure on the prime-age participation rate, that support was more than swamped by negative changes in participation rates within demographic groups. The following chart shows the relative contributions of these behavioral changes.

macroblog - May 6, 2019 - Chart 3: Contributions to Change Due to Behavior Q1 2008-Q1 2015

Within demographic groups, the increased incidence of being too sick or disabled to work stands out as the largest contributor to the decline in prime-age labor force participation between 2008 and 2014.

Since 2014, prime-age LFP has benefited from the movement of both demographics and participation behavior. But so far, less than half of the overall behavioral decline between 2008 and 2014 has been reversed.

Though demographic trends are likely to remain positive, how much more participation behavior—especially as it is related to disability and illness—can shift as the labor market tightens remains unclear. The share of the prime-age population too sick or disabled to work had been on a rising trend for the decade prior to the last recession, suggesting that there may be some deeper and structural health-related issues that could keep the disability/illness rate elevated despite an increasingly tight labor market.

May 6, 2019 in Employment, Labor Markets | Permalink


Excellent article. How do you explain the extreme decline in the 16-19 year old component LFPR? And why does Fred not have a 20-24 year old segment? Thanks

Posted by: Doug KORTY | May 07, 2019 at 08:16 AM

I began tracking age-specific participation about a decade ago, checking whether (my hypothesis) early retirements were a major component. That turned out not to be the case – age 55-59 showed little change, and 60-64 actually rose more or less monotonically from 1994, when the series I use start. Ditto the 65-69 and 70-74 brackets and (very noisy) 75+. At the opposite end, participation age 20-24 dropped a lot, and reflecting a longer trend, teen LF participation went from 40-45% in 2000 to 25% in 2010.

I look forward to a future blog that puts the prime age bracket into wider context - I don't focus on such labor issues in my research so don't know the wider literature. But thanks for this post!

Posted by: Michael Smitka | May 08, 2019 at 09:15 AM

Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

March 26, 2019

Young Hispanic Women Investing More in Education: Good News for Labor Force Participation

In a recent recent macroblog post, my colleague John Robertson found that the recent rise in female prime-age (ages 25 to 54 years) labor force participation (LFP) over the last few years has been driven in large part by increased participation among Hispanic women. (Hispanic refers to people of Cuban, Mexican, Puerto Rican, South or Central American, or other Spanish culture or origin regardless of race.) Much of the LFP improvement among Hispanic women has come as they've shifted away from household duties.

To understand this development and determine whether it's a trend likely to continue, we look at trends in the activities of younger Hispanic women. In particular, we look at the so-called NEETs rate among women ages 16 to 24. The NEETs rate is the share of the youth population that is "Not Employed or pursing Education or Training." This group is sometimes referred to as "disconnected youth" or "opportunity youth" because they are generally less likely to be attached to the labor force as they move into their prime working years and are at higher risk of experiencing long-term unemployment, persistent poverty, poor health, and criminal behavior.

A look at the next chart shows substantial improvement in the NEETs rate among young Hispanic women over the last two decades. The gap has narrowed considerably and in recent years has tracked much more closely with black non-Hispanic women.

The declining NEETs rate for young Hispanic women primarily reflects shifting preferences toward more education and away from household responsibilities. As you can see in the next chart, the share of young Hispanic women who are in education or training has risen over the last two decades, up nearly 19 percentage points since 2000. Their share now more closely matches that of young black and white non-Hispanic women.

Mirroring the rise in educational activities has been a shrinking share of young Hispanic women who are not in the labor force because they are taking care of home or family, as the following chart shows.

Young Hispanic women have invested increased time in their education over the last two decades and as a result have higher average levels of educational attainment than earlier cohorts moving into their prime working years. To see this, the next chart shows the distribution of educational attainment over time for Hispanic women aged 25.

The higher levels of LFP in recent years among prime-age Hispanic women partly reflects the greater investment in education by younger Hispanic women. If this trend continues—and there is no obvious reason why it wouldn't—then it will help drive even higher labor force attachment for prime-age Hispanic women in the years to come.


March 26, 2019 in Education, Employment, Labor Markets | Permalink


Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

March 22, 2019

A Different Type of Tax Reform

Two interesting, and important, documents crossed our desk last week. The first was the 2019 edition of the Economic Report of the President. What particularly grabbed our attention was the following statement from Chapter 3:

Fundamentally, when people opt to neither work nor look for work it is an indication that the after-tax income they expect to receive in the workforce is below their "reservation wage"—that is, the minimum value they give to time spent on activities outside the formal labor market.

That does not strike us as a controversial proposition, which makes the second of last week's documents—actually a set of documents from the U.S. Department of Health and Human Services (HHS)—especially interesting.

In that series of documents, HHS's Nina Chien and Suzanne Macartney point out a couple of things that are particularly important when thinking about the effect of tax rates on after-tax income and the incentive to work. The first, which is generally appreciated, is that the tax rates that matter with respect to incentives to work are marginal tax rates—the amount that is ceded to the government on the next $1 of income received. The second, and less often explicitly recognized, is that the amount ceded to the government includes not only payments to the government (in the form of, for example, income taxes) but also losses in benefits received from the government (in the form of, for example, Medicaid or child care assistance payments).

The fact that effective marginal tax rates are all about the sum of explicit tax payments to the government and lost transfer payments from the government applies to us all. But it is especially true for those at the lower end of the income distribution. These are the folks (of working age, anyway) who disproportionately receive means-tested benefit payments. For low-wage workers, or individuals contemplating entering the workforce into low-wage jobs, the reduction of public support payments is by far the most significant factor in effective marginal tax rates and the consequent incentive to work and acquire skills.

The implication of losing benefits for an individual's effective marginal tax rate can be eye-popping. From Chien and Macartney (Brief #2 in the series):

Among households with children just above poverty, the median marginal tax rate is high (51 percent); rates remain high (never dipping below 45 percent) as incomes approach 200 percent of poverty.

Our own work confirms the essence of this message. Consider a representative set of households, with household heads aged 30–39, living in Florida. (Because both state and local taxes and certain transfer programs vary by state, geography matters.) Now think of calculating the wealth for each household—wealth being the sum of their lifetime earnings from working and the value of their assets net of liabilities—and grouping the households into wealth quintiles. (In other words, the first quintile would the 20 percent of households with the lowest wealth, the fifth quintile would be the 20 percent of households with the highest wealth.)

What follows are the median effective marginal tax rates that we calculate from this experiment:

Wealth percentile

Median Effective Marginal Tax Rate

Lowest quintile


Second quintile


Third quintile


Fourth quintile


Highest quintile


Note: The methodology used in these calculations is described here and here.
Source: 2016 Survey of Consumer Finances, the Fiscal Analyzer

Consistent with Chien and Macartney, the median effective marginal tax rates for the least wealthy are quite high. Perhaps more troubling, underlying this pattern of effective tax rates is one especially daunting challenge. The source of the relatively high effective rates for low-wealth individuals is the phase-out of transfer payments, some of which are so abrupt that they are referred to as benefits, or fiscal, cliffs. Because these payments differ widely across family structure, income levels within a quintile, and state law, the marginal tax rates faced by individuals in the lower quintiles are very disparate.

Note: The methodology used in these calculations is described here and here.
Source: 2016 Survey of Consumer Finances, the Fiscal Analyzer

The upshot of all of this is that "tax reform" aimed at reducing the disincentives to work at the lower end of the income scale is not straightforward. Without such reform, however, it is difficult to imagine a fully successful approach to (in the words of the Economic Report) "[increasing] the after-tax return to formal work, thereby increasing work incentives for potential entrants into the labor market."



March 22, 2019 in Employment, Fiscal Policy, Labor Markets, Taxes, Unemployment | Permalink


Many lower income workers are paid and transact their financial affairs in cash and pay no income or little income tax. Even if tax is reported, the governments inability to collect and debt foregiveness programs for lower income workers also has to be factored in. The actual rate (at least for income tax, not necessarily sales or property tax) is probably closer to something between 0 and 10%....

Posted by: Houston Tax Attorney | March 31, 2019 at 07:40 PM

Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

March 06, 2019

X Factor: Hispanic Women Drive the Labor-Force Comeback

The share of the prime-age population engaged in the U.S. labor market is on the rise, led by a sharp rebound in the labor force participation (LFP) rate by prime-age female workers (those ages 25–54). This point was highlighted in a recent Wall Street Journal article.

Since 2015 the LFP rate for prime-age women has increased by about 1.8 percentage points, reversing an almost 16-year slide. Using the data underlying the Atlanta Fed’s new Labor Force Participation Dynamics tool, some of the factors behind this increase become apparent. Of particular note is that Hispanics (people of Cuban, Mexican, Puerto Rican, South or Central American, or other Spanish culture or origin regardless of race) account for a bit less than one-fifth of the prime-age female population, but they accounted for almost two-thirds of the increase in female LFP during the last three years. This increase was the result of both a rising share of the population that is Hispanic and the rising LFP rate among Hispanic women. The share of the prime-age, Latina population increased by 1 percentage point between 2015 and 2018, and the LFP rate for this group increased 3 percentage points. (The reason why rising Hispanic LFP didn’t result in the overall female LFP rate increasing by more than 1.8 percentage points is because Hispanic women are still 8 percentage points less likely to be in the labor force than non-Hispanic women. But this participation gap is closing rapidly.)

The Atlanta Fed’s web tool also allows us to further explore what is behind the 3 percentage point LFP rate increase for prime-age Latinas in the last three years. (My Atlanta Fed colleague Ellyn Terry provides a longer-term view on Hispanic female labor force dynamics in this related macroblog post.) It’s particularly noteworthy that almost two-thirds of the recent increase is the result of a decline in family or household responsibilities keeping people out of the labor force (see the chart).

Contributions to Total Change by Nonparticipation Category: Women of prime ages of Hispanic descent with all education types from Q4 2015 to Q4 2018

This shift away from household duties is attributable to a combination of the shifting demographics of the Hispanic population (such as being more likely to have a college degree and thus obtaining a higher-wage job and being better able to afford child care) and a lower propensity to not participate for family reasons within Hispanic age and education groups.

The rebound in female LFP in the last three years is good news, with rising wages, particularly at the low end, and higher demand in traditionally female-dominated occupations contributing to the increase. But making the labor market a truly viable option for women still poses a number of challenges. The LFP rate of U.S. women has fallen behind that of many other countries, many of which have enacted family-friendly policies to help support women in the workplace.

March 6, 2019 in Employment, Labor Markets | Permalink


Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

February 14, 2019

Trends in Hispanic Labor Force Participation

Although the labor force participation (LFP) rate has fallen significantly for the overall population during the past two decades, the trends can differ a great deal depending on which demographic group you examine. One way to view these varied, ever-changing patterns is to use the Atlanta Fed's Labor Force Participation Dynamics tool. We recently redesigned the tool, adding a new interface and more options for understanding specific demographic groups' LFP. The tool also allows users to see what factors (such as disability/illness, being in school, retirement, or family responsibilities) influence changes in the LFP rate for different groups.

While the tool shows us many stories, a particularly interesting one is the experience of people of Cuban, Mexican, Puerto Rican, South or Central American, or other Spanish culture or origin regardless of race (henceforth referred to as Hispanic). Hispanics are a growing share of the U.S. population (the U.S. Bureau of Labor Statistics [BLS] projects that nearly a fifth of the people in the labor force will be Hispanic by 2024, up from a tenth in 1994), and therefore Hispanics' differences in attitudes and preferences for work will exert an increasingly great effect on headline LFP numbers.

Let's parse the LFP rate among Hispanics. First, the Hispanic population is more likely to engage in the labor force than non-Hispanics. (In 2017, the LFP rate of Hispanics was 66.1 percent, compared to 62.2 percent for non-Hispanics). Second, their LFP has fallen by less during the past two decades.

The pictures below come from the redesigned tool. The first two charts compare the decline in the LFP rate for all ethnicities (Chart 1) versus Hispanic (Chart 2) as the combination of six nonparticipation categories. Each colored bar represents how much a particular category of nonpartipation has changed since the fourth quarter of 1998. The red line shows the summation of the change in each nonparticipation category, or the net change in the LFP rate. For example, the LFP rate overall has declined 4.2 percentage points (ppts) during the past two decades. However, among Hispanics, it has fallen significantly less—just 1.0 ppt. A comparison of the size and direction of each of the nonparticipation categories between the two charts shows many differences in the factors affecting the decline in the LFP rate of each group.



Because differences across ethnicity could reflect differences in their age distributions—Hispanics are younger on average than the population as a whole—it is important to control for this difference. Using the tool, it's easy to narrow this comparison to look specifically at 26–55 year olds.

In particular, the LFP rate for women of all ethnicities from 26 to 55 years old has declined by 1.0 ppt since 1998. In sharp contrast, the LFP rate for Hispanic women 26 to 55 years old has actually increased by 3.8 ppts. Compared to 20 years ago, this group is less likely to say they don't want a job because of disability/illness (1.1 ppt) and family responsbilities (1.4 ppt). This group is also less likely to be part of the shadow labor force (1.4 ppt) compared to two decades ago. (The shadow labor force, as we define it, is made up of individuals who say they want a job but are not considered unemployed by the BLS.) This article from the BLS delves into more detail about Hispanics in the labor force.



The LFP tool allows you to explore many other labor force stories. Users can cut the LFP data by three education categories (less than a high school degree, high school or some college, or associate's degree or higher), two age groups (26–55 or all ages), three race/ethnicity categories (white non-Hispanic, black non-Hispanic, and Hispanic) and for men and women. One thing that the tool makes clear is that the factors that influence individual decisions to work, look for work, or to pursue other activities vary across demographic groups, and each group's experience contributes to our understanding of movements in the overall LFP rate.

February 14, 2019 in Employment, Labor Markets | Permalink


The new tool is an outstanding piece of work. Is there a deficiency in the underlying data which prevents an age partition for those older than 55? Early retirement and re-entry into the labor force may be significant indicators of economic conditions and certainty, especially in the period during and since the Great Recession. These factors are hard to see when the student-age and primary-age groups are included.

Posted by: Charles Smiler | February 14, 2019 at 03:17 PM

Thanks for your interest in the tool. You may want to look at the Trends Over Time tab of the web page, which shows how participation and nonparticipation categories, including retirement, have changed over time by age (and otherselected demographic characteristics).

Posted by: Ellyn Terry | February 26, 2019 at 10:51 AM

Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

Google Search

Recent Posts



Powered by TypePad