Blog postings by CEPR staff and updates on the latest briefings and activities at the Center for Economic and Policy Research.

In the summer of 1996, the Los Angeles Times was warning its readers about “the shadow of rising inflation” [1] looming over U.S. financial markets “like the mammoth alien spacecrafts of Independence Day,” the blockbuster movie of the time. The article suggested the danger of inflation was imminent and concluded that “for the shell-shocked bond market, there will be an assumption that either the Fed is going to tighten credit, or inflation is going to continue to rise, or both.”

The general consensus in 1996 was that the economy was at, or very close to full employment. Unemployment was at its lowest level in years, which was about 5.4 percent. In their outlook for 1996–2000, the Congressional Budget Office (CBO) estimated the non-accelerating inflation rate of unemployment (NAIRU) was 6 percent. The CBO also warned that “if rapid growth continues, inflationary pressures will mount.”

Martin Feldstein, then President of the National Bureau of Economic Research, expressed his concern with unemployment being too low as early as March 1995. Feldstein stated that while “people have argued that we can tolerate much lower interest rates than in past U.S. historical experience,” he did not “think that’s true.” Feldstein declared the unemployment rate of 5.4 percent as being “well below the unemployment rate that even the Congressional Budget Office (CBO) would call full employment” and urged the Fed to severely tighten monetary policy. Feldstein estimated NAIRU at 6.23 percent and concluded that the possibility of being wrong was “quite low.”

Add a comment

Jeff Hauser runs the Revolving Door Project, an effort to increase scrutiny on executive branch appointments and ensure that political appointees are focused on serving the public interest, rather than personal professional advancement.

As we hear of a settlement in the “Trump University” civil fraud case brought in part by New York State Attorney General and learn more and more about potential Treasury Secretary Steven Mnuchin, the phrase “personnel is policy” takes on an unfortunate new meaning.

Will Trump’s appointees to high government office ensure Donald Trump does not use control of the executive branch to enrich himself and his family?

Trump enriching himself as president is not an idle or libelous question. Trump himself raised the prospect in 2000 to Fortune Magazine, telling them that “[i]t’s very possible that I could be the first presidential candidate to run and make money on it.”

Add a comment

A common argument for the decline in employment in recent years is that more workers are dropping out of the labor force to live off public benefits, particularly Social Security Disability Insurance (SSDI). SSDI is a program that provides cash benefits and healthcare to former workers who become too disabled to continue at their jobs. The average benefit payment is about $15,000 per year, and workers must have an extensive work history in order to qualify.

While the percentage of the workforce receiving disability benefits has increased since 2000, much of this rise is simply due to demographics, most importantly the aging of the population.[1] Figure 1 shows the unadjusted SSDI beneficiary rate and a second rate that has been adjusted for the age and gender composition of the labor force. [2,3]

Figure 1

buffie 2016 11 16 1

SSDI is actually one of two major programs benefiting disabled workers. The other is Workers’ Compensation (WC), a privately-run insurance system for workers who get injured at their jobs and are unable to continue working. While there are some differences between WC and SSDI, the two programs are broadly similar. Moreover, previous CEPR research indicates that they function as substitutes: when enrollment goes up for WC, it goes down for SSDI, and vice versa.

Over the past twenty years, many states have made significant cuts to their WC programs. Benefits have declined, as have the number of injuries covered by WC. Not surprisingly, these cuts have coincided with a decline in the number of WC beneficiaries and a proportionate increase in the number of SSDI beneficiaries.[4]

If the number of beneficiaries in these two programs is combined, there is almost no change from 2000 to 2011 in the percentage of the workforce getting benefits. (Our data on WC recipients only go through 2011.) Furthermore, if we include the impact of changing demographics on the number of SSDI beneficiaries, the share of the workforce receiving benefits has actually declined, as shown in Figure 2.[5,6] (Note that the age- and sex-adjustment only applies to the number of workers receiving SSDI, as the data on WC beneficiaries do not include demographic breakdowns.)

Figure 2

buffie 2016 11 16 2

In short, there has been no increase in the share of the SSDI-eligible population living off some form of disability benefits over this period. Moreover, there has actually been a drop if we take into account the impact of changing demographics.

Finally, it is worth mapping the trend depicted in Figure 2 onto more recent years. Unfortunately, as stated earlier, our data on the number of WC recipients only go through 2011. Figure 3-1 shows the number of WC and SSDI beneficiaries as a share of the SSDI-eligible population from 2000 to 2015; for the years 2012 through 2015, it is assumed that either the number of workers receiving WC benefits (the dashed line) or the percentage of workers receiving WC benefits (the dotted line) hasn’t changed since 2011. If one of these assumptions is true, then the share of the workforce receiving some form of disability benefits would’ve fallen between 0.14 and 0.31 percentage-points between 2000 and 2015.[7]

Figure 3-1

buffie 2016 11 16 3

But the assumption made in Figure 3-1 is likely too generous; after all, the number of WC beneficiaries fell every single year between 2000 and 2011. Figure 3-2 shows the number of WC and SSDI beneficiaries as a percentage of the SSDI-eligible population under the assumption that the number (the dashed line) or share (the dotted line) of WC recipients continued falling at the same rate from 2011-2015 as during the previous 11 years. Using this set of assumptions, the share of the workforce receiving either WC or SSDI would’ve fallen 0.93 to 1.01 percentage-points.

Figure 3-2

buffie 2016 11 16 4

The number of workers receiving disability benefits has fallen over the past 15 years, though we can’t say by how much. The decrease may be comparable to a rounding error (0.14 percentage-points), or it may be rather large (1.01 percentage-points). But even if we don’t know the size of the decrease, we can be certain that at least some decrease has occurred – and this should put a real dent in the “SSDI recipients as takers” argument. Prominent conservative pundits have argued that the drop in employment since 2000 is driven in large part by the fact that more Americans are choosing to take disability benefits rather than work. Those who use this argument are engaging in serious cherry-picking: they are highlighting the program with rising enrollment (SSDI) while ignoring the program with declining enrollment (WC). When both programs are examined together, there is no apparent increase in the number of Americans receiving benefits.


[1] Another source of rising SSDI enrollment is the increase in Social Security’s “full retirement age”. When SSDI beneficiaries hit full retirement age, they stop receiving SSDI benefits and start receiving normal Social Security retirement benefits. This means that the one-year increase (from 65 to 66) in Social Security’s full retirement age has kept many disabled workers on SSDI for an extra year. Between 2000 and 2014, the number of 65-year-old SSDI beneficiaries went from zero to 467,000; this accounts for 11.9 percent of the increase in SSDI beneficiaries during this time.

[2] An extensive work history is required in order to become eligible for SSDI benefits. For a full description, see pg. 20-21 of this CEPR report, “Benefits Planner: Social Security Credits” at the Social Security Administration (SSA) website, and this SSA pamphlet. The SSDI beneficiary rate is calculated by dividing the number of SSDI recipients by the number of people eligible for benefits.

[3] The demographic adjustment is based on data presented in Table V.C5 on page 141 of the 2016 Social Security Trustees’ Report.

[4] To determine the number of WC beneficiaries, data are drawn from two sources: the Annual Statistical Bulletin published by the National Council on Compensation Insurance (NCCI), and various annual reports on WC published by the National Academy of Social Insurance (NASI). NCCI’s Annual Statistical Bulletin provides data, by state, on the number of WC beneficiaries per 100,000 covered workers. By combining NCCI’s data with NASI’s data on the number of covered workers, we are able to determine the number of WC beneficiaries in each state in any given year. However, because NCCI’s data do not cover North Dakota, Ohio, Washington (state), West Virginia, and Wyoming, it is assumed that the take-up rate among covered workers in those states is the same as the average take-up rate for workers in the other 45 states and DC.

[5] The number of workers taking some form of disability benefits is less than the sum of “WC beneficiaries plus SSDI beneficiaries”. This is because a small number of people – between 361,000 and 401,000 per year between 2000 and 2011 – actually receive benefits from both programs. In order to not double-count people benefiting from both WC and SSDI, the number of people receiving some form of benefits is calculated as follows: (WC Beneficiaries) + (SSDI Beneficiaries) – (Dual Beneficiaries) = (Total Number of Beneficiaries).

[6] Data on the number of dual beneficiaries for the years 2000-2002 are drawn from the 2001, 2002, and 2003 NASI reports on WC coverage. Unfortunately, beginning with NASI’s 2004 paper, the reported number of dual beneficiaries includes people utilizing a third (relatively minor) disability program known as “public disability benefits”. Therefore, post-2002 data come from the Social Security Administration’s Annual Statistical Reports on the Social Security Disability Insurance Program. The number of dual beneficiaries is drawn from Table 31. Because Table 31 includes SSDI dual beneficiaries whose second disability program can be either WC or public disability benefits, the number of WC-SSDI dual beneficiaries is calculated as follows:

All workers receiving both WC and SSDI (lines 7-12) are included;

All workers receiving both SSDI and public disability benefits (lines 13-16) are excluded;

All workers receiving WC, SSDI, and public disability benefits (lines 17-20) are included;

For workers listed in lines 21-23, it is assumed that the same percentage of these workers receive both SSDI and WC as was divined from lines 7-20;

All workers with pending WC or public disability benefit applications (line 24) are excluded, since they are not receiving benefits at the time.

This formula gives us a close approximation for the number of WC-SSDI dual beneficiaries for each year from 2005 to 2011. However, because estimates are not available for 2003-2004, there is a gap in the data on dual beneficiaries for those two years. The number of dual beneficiaries for both 2003 and 2004 was determined by a process of linear interpolation linking the 2002 NASI data with the 2005 Social Security Administration data.

[7] For the year 2012, the number of dual beneficiaries is calculated according to the methodology outlined in footnote number six.  However, beginning in 2013, the Social Security Administration changed how it presents data on the number of dual beneficiaries. For the years 2013-2015, the number of dual beneficiaries (as determined from Table 31) has been calculated as follows:

All workers receiving both WC and SSDI (lines 9-12) are included;

All workers receiving both SSDI and public disability benefits (lines 14-16) are excluded;

All workers receiving WC, SSDI, and public disability benefits (lines 17) are included;

For workers listed in lines 18 and 20, it is assumed that the same percentage of these workers receive both SSDI and WC as was divined from lines 9-17;

All workers with pending WC or public disability benefit applications (line 21) are excluded, since they are not receiving benefits at the time.

Add a comment

Private equity’s popularity among pension plans — still the largest source of investments in private equity funds — owes much to the fabled returns the industry earned in its glory days. Private equity (PE) firms claim that investing in their funds is the path to high returns and a secure retirement for pension plan beneficiaries. Returns over the last 10 years, however, have been disappointing. Research by leading finance economists demonstrates that the median private equity fund launched since 2006 has failed to beat the stock market. PE investors could have done as well investing in a stock market index fund without having to tie their money up for 10 years and without the lack of liquidity, lack of transparency and high fees of private equity.  

How does private equity manage to pull the wool over the eyes of supposedly sophisticated limited partner investors in PE funds — not just pension plans but insurance companies and university and foundation endowments? One part of the answer is that the private equity industry employs a flawed yardstick — the internal rate of return or IRR — to measure its performance. While this measure has been discredited by leading finance professors and discarded by them in favor of a truer measure of performance [1], the IRR has the virtue from the industry’s point of view that it exaggerates returns. This, however, is not its only virtue.

As PE firms have long known, the IRR is a mathematical algorithm that is easy to manipulate. For example, a PE fund that requires its highly indebted portfolio companies to sell junk bonds and take on even more debt in order to pay a dividend to investors in the fund in the first few years of the fund’s life will be rewarded with a boost to its IRR that bears little relation to what investors will ultimately receive. And that’s not the only trick PE firms use. Selling a profitable portfolio company early in the fund’s life will also goose returns as measured by the IRR even if holding onto the company and selling it later at a higher price would have increased the actual returns investors receive. But in the current low PE performance environment, these tricks may no longer raise the IRR sufficiently to make risky investments in PE look attractive.

Add a comment

Workers in the United States trail other high-income countries in the Organization for Economic Co-operation and Development (OECD) in terms of work-life balance. The U.S. tops the list in terms of yearly hours worked and falls significantly behind other countries when looking at efforts to improve work-life balance.

The European Union (EU) explicitly forbids an employer to hire workers for a regular workweek of more than 48 hours. There is no equivalent restriction in the United States. While this restriction doesn’t limit hours of workers who hold more than one job or are self-employed, it is indicative of the general attitude on work-life balance in the EU.

 

mb hours worked 2016 11 08 1

Add a comment

The following reports on labor market policy were recently released:


Center for American Progress

A Fair Shot for Millennial Women and Families
Sunny Frothingham


Institute for Women’s Policy Research

Supportive Services in Job Training and Education: A Research Review
Cynthia Hess, Yana Mayayeva, Lindsey Reichlin, Mala Thakur


Urban Institute

Validation of the Employment Retention Inventory
Jennifer Yahner, Ellen Paddock, Janeen Buck Willison

Add a comment

The October jobs report provides solid evidence that the labor market is improving for most of the labor force. While the pace of job growth slowed slightly to 161,000, there was an increase in the prime-age employment rate (ages 25-54) and an uptick in the pace of wage growth. It appears that workers are taking advantage of a tighter labor market to move into better paying jobs.

The employment-to-population (EPOP) ratio for prime age workers rose 0.2 percentage points to 78.2 percent. This is the highest it has been in the recovery, although it is still 2.0 percentage points below the pre-recession peak and 3.7 percentage points below its 2000 peak.

Other data in the report were clearly positive. Wage growth appears to have accelerated modestly. The average hourly wage over the last three months increased at a 2.9 percent annual rate compared to its average over the prior three months. Over the last year, the average hourly wage is up 2.8 percent.

This is consistent with a story where workers are looking for better paying jobs. In this respect, it is worth noting that the share of unemployment due to voluntary quits rose 0.9 percentage points to 12.1 percent. This is the highest level for the recovery and comparable to the pre-recession rates, although still more than 3.0 percentage points below the 2000 peaks.

Also consistent with this view is the fact that lower paying industries were not sources of big job gains in October. Restaurants added just 9,900 jobs in October, compared to an average of 21,000 over the last year. Retail actually lost 1,100 jobs in the month.

On the whole, this report indicates that the labor market is getting into a situation where most workers can benefit from the economy's growth.

Add a comment

With open enrollment for health insurance exchanges for 2017 starting soon, the news has been saturated with articles on exploding rates. However, after accounting for the subsidies individuals are eligible for through the Affordable Care Act, a close look at the actual cost for individuals purchasing insurance on the marketplace paints a very different picture.

The Kaiser foundation created a calculator that uses insurance premiums from exchanges in each state, along with available subsidies to estimate the actual costs for those buying insurance. These costs depend on age, family size, income, and location.

Add a comment

The following highlights CEPR’s latest research, publications, events, and much more.

Dean’s New Book

CEPR Co-Director Dean Baker’s new book, Rigged: How Globalization and the Rules of the Modern Economy Were Structured to Make the Rich Richer, is now available! His book shows that upward redistribution was not the result of globalization and the natural workings of the market. Rather, it was the result of conscious policies that were designed to put downward pressure on the wages of ordinary workers while protecting and enhancing the incomes of those at the top. Dean explains how rules on trade, patents, copyrights, corporate governance, and macroeconomic policy were rigged to make income flow upward.

Dean wrote a piece for the Institute for New Economic Thinking (INET)’s blog that elaborates on the central theme of the book, and the first chapter was featured in Naked Capitalism. He also wrote for PBS NewsHour how intellectual property rules help the rich and hurt the poor. Dean was interviewed on the Thom Hartman show twice, first by Alex Lawson (Social Security Works) and then by Thom himself, and Dean’s views were cited in this op-ed in the Toronto Star on the Comprehensive Economic and Trade Agreement (CETA).

CEPR on Trade

Dean wrote this op-ed on the Trans-Pacific Partnership (TPP) and “free” trade. Dean also wrote numerous posts for his blog, "Beat the Press," on the media’s coverage of trade issues, including this one that clarifies a piece on (“the usually excellent”) NPR show This American Life on NAFTA. Dean also took The New York Times to task for pushing NAFTA and for promoting protectionism for doctors, dentists, pharma and the entertainment industry. He also analyzed an NYT piece on the trends in trade.

Many of CEPR’s analyses of the TPP and other trade deals were written in the context of the presidential race. Dean wrote an op-ed for The Hankyoreh called “Apologies for Trump” that discussed Donald Trump’s position on trade, and he was cited in this Think Progress piece that looked at Trump’s proposals. In this post for Beat the Press, Dean says that Trump is not all wrong on trade, while in this one he notes that Trump is closer to the mark than CNN.

Add a comment

The following reports on labor market policy were recently released:


Economic Policy Institute

Black Workers’ Wages Have Been Harmed by both Widening Racial Wage Gaps and the Widening Productivity-Pay Gap
Valerie Wilson


The Century Foundation

Organizing’s Business Model Problem
Shayna Strom


National Employment Law Project

The Case for Phasing out Maine’s Subminimum Wage for Tipped Workers

Add a comment

The economy grew at a 2.9 percent annual rate, the strongest growth rate since the third quarter of 2014. Most forecasts had put growth for the quarter at just over 2.0 percent. While this number is better than expected, a big factor was an increase in inventory accumulation, which added 0.61 percentage points to growth. Accumulation was actually negative in the second quarter, so the rate of accumulation is likely to be even higher in the fourth quarter.

Most categories of final demand were relatively weak. Consumption grew at a modest 2.1 percent annual rate. Residential investment fell at a 6.2 percent annual rate, its second consecutive decline. Non-residential fixed investment grew at a 1.2 percent rate, roughly the same pace as in the second quarter. This follows two quarters of decline. The collapse of energy prices and the increase in the trade deficit in manufacturing are the major factors behind the weakness in this component.

Government expenditures increased at a 0.5 percent annual rate, with a 2.5 percent increase in federal spending offsetting a 0.7 percent decline at the state and local level. This is the second consecutive quarter of decline at the state and local level.

Exports were a source of strength in the quarter, rising at a 10.0 percent annual rate, the strongest performance since the fourth quarter of 2013. As a result of this increase, net exports added 0.83 percentage points to growth for the quarter.

One striking figure in this report is the slower pace of inflation shown in the core personal consumption expenditure deflator (PCE). This rose at just a 1.7 percent annual rate in the third quarter. The rate of inflation shown in the core PCE has been trailing off throughout the year, rising at a 2.1 percent annual pace in the first quarter and a 1.8 percent rate in the second quarter. While there are enough erratic movements in the quarterly data to avoid treating this as evidence of deceleration, it is certainly hard to make a case for acceleration with these data.

Add a comment

Japan is not exactly the country that comes to mind as a model of gender equality and high labor force participation for women. In 1995, the OECD estimated women’s employment rate in Japan at 56.5 percent, almost 10 percentage points lower than in the U.S., which had an employment rate of 66 percent for women at the time. Between 1995 and 1999, the employment rate for Japanese women grew by less than half a percentage point, while for women in the U.S. it grew by almost 3 percentage points.

However, since 2000 this trend completely reversed. Between 2000 and 2015, the employment rate for women in Japan increased by almost 14 percentage points, while in the U.S. it dropped by over 6 percentage points. Japan’s employment rate for women surpassed the U.S. in 2014. Currently, almost 65 percent of Japanese women are employed, while only about 63 percent of women in the US are employed.

In recent years, Japan has launched extensive campaigns to encourage labor force participation by women. The government took various steps such as increasing allowances given to new parents, subsidizing daycare, and ensuring both mothers and fathers benefit from paid parental leave.

Add a comment

The following reports on labor market policy were recently released:


Economic Policy Institute


What is the gender pay gap and is it real?
Elise Gould, Jessica Schieder, Kathleen Geier


Center for American Progress


A Progressive Agenda for Inclusive and Diverse Entrepreneurship
Kate Bahn, Regina Willensky, Annie McGrew


Center for Law and Social Policy

How States Can Protect Workers with Irregular Schedules from Losing SNAP Benefits
Nune Phillips









Add a comment

Government jobs are often thought to be highly desirable, with high levels of job security and generous benefits. In fact, research shows that the greater generosity of benefits is offset by lower pay. Adjusting for education and experience, compensation levels for government workers is comparable to compensation levels for private sector workers. In the last decade, there have been efforts at all levels of government to reduce the pay and benefits of government employees, which have been motivated in part by the argument that government workers are overpaid.

The evidence is that these efforts have accomplished their goal. The quit rates for government sector workers have risen substantially relative to the quit rates for their counterparts in the private sector.

Add a comment

Some may see evidence of domestic violence as a visible mark on the body — a bruised face, perhaps a broken arm, or much worse for many victims. However, what we may not see are the economic consequences suffered by those who have been abused: how many days of work a victim has missed due to domestic abuse, or how many jobs she or he may have lost due to their abuser’s actions. Domestic violence isn’t limited only to acts of physical violence; abuse may result in financial and economic consequences that take away a survivors autonomy. Public policy can help mitigate these devastating effects of domestic violence. A key policy that can help is paid sick days that cover time off to deal with legal and health consequences of abuse. If implemented, such paid sick days laws would positively impact all workers, and also benefit domestic violence survivors.

Paid sick days laws are starting to sweep the country there are now 37 jurisdictions that have paid sick day laws in effect or where such laws will be implemented soon. Paid sick days provide economic security for victims of domestic violence so that taking time off to deal with domestic violence issues court appearances, doctors appointments, meetings with social workers, or healing wouldn’t mean survivors have to forfeit income or put their employment in jeopardy. All five states that have passed paid sick days laws Connecticut, California, Massachusetts, Oregon, and Vermont include provisions where sick time can be used for specific “safe time” purposes. This allows workers to take time off for purposes related to domestic violence. However, not all city jurisdictions with paid sick days include this provision, an oversight that needs to be corrected. Paid sick days would allow victims time to seek lifesaving services from local domestic violence programs.

Add a comment

The following reports on labor market policy were recently released:


Economic Policy Institute

Black-white wage gaps expand with rising wage inequality
Valerie Wilson and William M. Rodgers III 


Urban Institute

Women’s Economic Empowerment: A Review of Evidence on Enablers and Barriers
H. Elizabeth Peters, Nan Marie Astone, Ammar A. Malik, Fenohasina Maret Rakotondrazaka, Caroline Heller


Institute for Women’s Policy Research

Job Growth Among Women Continues to Climb: 65 percent of Jobs Added in the 3rd Quarter of 2016 Went to Women


National Woman’s Law Center

Progress in the States for Equal Pay

Add a comment

In many narrow economic indicators, such as wealth, income, and housing, the United States ranks near the top of countries in the Organization for Economic Cooperation and Development (OECD). However, when it comes to broader measures focused of well-being, the U.S. does not perform very well. In terms of life-work balance, the OECD ranks the U.S. 30th out of the 38 countries included in its index. As also pointed out by the OECD, the U.S. is the only member country, which fails to offer any type of nationally mandated paid leave policies. While some states, or cities, provide paid family and sick leave, at a national level no such policy exists. Furthermore, most OECD members also mandate paid vacation, a concept foreign to many American workers.

The Washington Center for Equitable Growth used data from the Bureau of Labor Statistics Current Population Survey to show that the problem of overwork in America is present in all types of professions. Working over 40 hours a week is fairly common for workers regardless of education level or profession. For example, over 40 percent of those in management positions and over 30 percent of those in farming, fishing, and forestry work over 40 hours a week.

Add a comment

Last month, the Census Bureau released its annual report on income and poverty in the United States. The report indicated that although lower- and middle-income Americans had seen significant income growth in 2015, those at the top of the income distribution had seen the strongest growth over the past forty years.

The Census Bureau’s most widely-cited statistics on inequality come from data on household income. Various commentators have questioned the validity of citing household income, arguing that the definition of “household” doesn’t take into account differences in factors such as family size. If this is true, it could mean that the rise of economic inequality is more of a statistical aberration than a real phenomenon.

Add a comment

The Labor Department reported that the economy added 156,000 jobs in September, somewhat less than most economists had projected. The job growth figures were also on net revised down slightly for the prior two months, so that the average for the last three months stands at 192,000. The unemployment rate edged up to 5.0 percent, but this was due to a large number of people entering the workforce, as the employment-to-population ratio (EPOP) also rose by 0.1 percentage point to 59.8 percent. The EPOP for prime age workers reversed its decline last month and stood at 78.0 percent. However, this is still more than two full percentage points below its pre-recession peak.

Other news in the household survey was mostly positive. The number of people involuntarily working part-time hit a new low for the recovery. It is now down by more than 3.3 million from the recession peak, although it is still well above the pre-recession level. The duration measures of unemployment all fell, with the share of long-term unemployed dropping by 1.2 percentage points, a new low for the recovery. One negative was a modest decline in the share of unemployment due to voluntary job leavers, which remains more than a full percentage point below its pre-recession peak.

On the establishment side, professional business services, which added 67,000 jobs, construction, which added 23,000, and retail which added 22,000 were big job gainers, along with health care, 32,700 and restaurants, 29,700. Manufacturing lost 13,000 jobs and government lost 11,000. There was a modest increase in average weekly hours, reversing last month's decline. Wages grew at a 2.6 percent annual rate over the last three months, compared with the prior three, the same as their pace in August.

Add a comment

When the Bureau of Labor Statistics released its newest jobs report a little over three weeks ago, economic forecasters and analysts received some welcome news: the unemployment rate had held steady at 4.9 percent for the month of August. Not long after the release, several members of the Federal Reserve’s Federal Open Market Committee (FOMC) began talking about an imminent interest rate hike, noting that the economy was operating at close to full employment.

A recent report from CEPR challenges this narrative. The report — titled The Case for a Weak Labor Market argues that a large number of non-employed workers who want jobs simply aren’t being counted in the unemployment rate.

In 2007, there were approximately 7.1 million unemployed Americans. There were also an additional 4.7 million Americans not in the labor force the “labor force” being defined as the number of people who are either employed or unemployed who said they wanted to work. These 4.7 million people weren’t counted as “unemployed” because they hadn’t searched for work during the previous four weeks. But in total, there were about 11.8 million non-employed Americans who wanted to find jobs.

The most recent labor market data show that there are currently about 7.8 million unemployed workers. Given demographic shifts and moderate population growth, this isn’t dramatically out of line with the number of workers unemployed in 2007. However, there are now almost six million Americans out of the labor force who say they want a job meaning that the number of non-employed people looking for work is far higher than the unemployment rate implies.

The aging of the population does not explain this trend. In the year leading up to the recession, just 9.7 percent of prime-age (25–54) Americans not in the labor force said they wanted a job. Over the past year, it’s been 10.5 percent. The increase is nearly identical for older workers. Between September 2015 and August 2016, 3.0 percent of Americans aged 55 and older not in the labor force said they wanted a job, compared to just 2.2 percent in 2007. A far more likely culprit is the lingering economic weakness from the 2008 recession. Table 1 below shows the percentages of all people that want a job broken into three broad categories for 2007 compared to the past year.

buffie mcinnis 2016 10 table

While the economy has improved significantly over the past eight years, the experiences of being employed and unemployed today are much different than in 2007. Take for example the share of workers employed sufficient hours, which is down 1.7 percentage points relative to 2007.[1] This indicates that fewer workers are working the amount of hours they’d like to be. On the other hand, the share of workers who are underemployed, or working part-time involuntarily, is up nearly one percentage point. These workers are still counted as being employed; however, they are working part time hours because they are unable to find full time positions. In other words, these workers are still experiencing economic hardship.

In terms of the unemployed, long-term unemployment is up 0.5 percentage points and accounts for over 100 percent of the increase in unemployment, while short term unemployment is down relative to 2007. This indicates that being unemployed today is worse than being unemployed in 2007, since many individuals are staying out of a job much longer. Long-term unemployment brings a slew of hardships not necessarily experienced for those who are short-term unemployed, including depression, suicidal thoughts, and other problems. While the unemployment rate is near its 2007 average, those who are unemployed are experiencing more hardship than in 2007 since a larger share of them are experiencing long-term unemployment.

Looking at the bigger picture, “employed sufficient hours” and short-term unemployment are down nearly two percentage points, meaning the other four groups — the under-employed, the long-term unemployed, marginally attached workers, and other job wanters — are up nearly two percentage points. Job wanters include individuals who say they want a job and have searched for a job within the past 12 months, but not the past four weeks (also known as marginally attached workers) and people who say they want a job but haven’t searched within the past 12 months (also known as other job wanters). These individuals, who account for 3.6 percent of everyone who wants a job, are not captured in the unemployment rate. However, many of these people would likely return to work if there were more opportunities in the labor market.

For these reasons, the unemployment rate is not as accurate a measure of labor market tightness as it was in prior years.


[1] Those working full time or voluntarily working part time.

Add a comment

During tonight’s vice-presidential debate, Republican candidate Mike Pence will likely discuss his track record as governor of Indiana. However, Pence will likely leave out a few important facts about that record — including unsuccessful healthcare, tax, and labor market policies. Let’s look at each of these in turn.

Add a comment