New Census data show strong 2015 earnings growth across the board, with black and Hispanic workers seeing the fastest growth
Today’s Census Bureau report on income, poverty, and health insurance coverage in 2015 shows that median household incomes for all race and ethnic groups increased between 2014 and 2015. Encouragingly, groups that, by and large, had seen the worst losses in the years since the Great Recession saw the biggest earnings gains in 2015. Real incomes increased 6.1 percent (from $42,540 to $45,148) among Hispanics, 4.4 percent (from $60,325 to $62,950) among non-Hispanic whites, 4.1 percent (from $35,439 to $36,898) among African Americans and 3.7 percent (from $74,382 to $77,166) among Asians. While the increase in incomes was statistically significant for all groups except Asians, racial income gaps remained unchanged between 2014 and 2015. The median black household earned just 59 cents for every dollar of income the white median household earned, while the median Hispanic household earned just 71 cents. Meanwhile, households headed by persons who are foreign born saw an increase in incomes of 5.3 percent between 2014 and 2015 (from $49,649 to $52,295), compared to an increase of 4.4 percent (from $54,741 to $57,173) among households with a native-born household head.
Based on EPI’s imputed historical income values (see the note under Figure A for an explanation), real median household incomes for all groups, except Hispanics, remain well below their 2007 levels. Between 2007 and 2015, median household incomes declined by 6.8 percent (-$2,686) for African Americans, 3.2 percent (-$4,662) for whites and 5.4 percent (-$7,158) for Asians, but increased 5.4 percent ($2,310) for Hispanics. Asian households continue to have the highest median income, despite large income losses in the wake of the recession.
The primary driving force behind the slow return to pre-recession income levels has been stagnant wage growth. Real wages had been essentially flat since 2000, but wage growth received an added boost in 2015, as a result of low inflation. From the start of the recovery in 2009 through 2015, real earnings of men working full-time, full-year are up for all race and ethnic groups—white men (1.5 percent), Hispanic men (7.0 percent), and black men (3.4 percent). As a result, the black-white male earnings gap is unchanged, but the Hispanic-white male earnings gap narrowed. Black men earned 70 cents for every dollar earned by white men in 2015 (compared to 69 cents/dollar in 2009) and Hispanic men earned 63 cents on the dollar (compared to 60 cents/dollar in 2009).
Poverty declined in 2015 by all measures; government programs, once again, kept millions above the poverty line
The official poverty rate fell by 1.3 percentage points from 2014 to 2015, as annual earnings and household incomes rose significantly for the first time since 2007. Since 2010, the U.S. Census Bureau has also released an alternative to its official poverty measure known as the Supplemental Poverty Measure (SPM). As shown in Figure A, the SPM has consistently indicated that poverty in America is more extensive than the official poverty measure reports. The good news is that the SPM data do show a similar decline in poverty last year to that reported in the official poverty measure. This year’s SPM release reported that in 2015, 45.7 million people were in poverty—roughly 14.3 percent of Americans. Under the “official” poverty measure, 43.1 million people were in poverty, or 13.5 percent of all Americans.
Poverty rates, official and supplemental poverty measure (SPM), all people and children, 2009–2015
| SPM – all people | Official poverty – all people | SPM – children | Official poverty – children | |
|---|---|---|---|---|
| 2009 | 15.3% | 14.3% | 17.3% | 20.7% |
| 2010 | 16.0% | 15.1% | 18.0% | 22.0% |
| 2011 | 16.1% | 15.0% | 18.0% | 21.9% |
| 2012 | 16.0% | 15.0% | 18.0% | 21.8% |
| 2013 | 15.8% | 14.8% | 18.1% | 21.5% |
| 2014 | 15.3% | 14.8% | 16.7% | 21.1% |
| 2015 | 14.3% | 13.5% | 16.1% | 19.7% |
Note: 2013 values reflect the CPS ASEC redesigned income questions.
Source: EPI analysis of Current Population Survey Annual Social and Economic Supplement Historical Income Tables.
The SPM data also show a lower rate of child poverty than the official statistics, primarily as a result of the SPM’s inclusion of noncash income from government assistance in its calculations. In 2015, the official child poverty rate was 19.7 percent—a decline of 1.4 percentage points from 2014. Under the SPM, the child poverty rate was 16.1 percent, 0.6 percentage points lower than in 2014 although the Census Bureau reports that this reduction in the SPM child poverty rate was not statistically significant.
Income gains in 2015 don’t reverse long-run trend toward greater inequality
In recent decades, the vast majority of Americans have experienced disappointing growth in their living standards—despite economic growth that could have easily generated faster gains in their living standards had it been broadly shared. Today’s excellent news on family income growth over the past year doesn’t make up for this long legacy of rising inequality. It is certainly a good start. But, we’ll need a run of years like this to restore the income losses suffered during the Great Recession for most American families, let alone make up for a generation of income growth that lagged far behind the economy’s potential.
As with most economic analysis, assessing the growth of living standards for the vast majority requires specifying benchmarks against which to measure actual performance. I offer up two reasonable benchmarks. The first is how income growth differs for families at different parts of the income distribution. What we have seen since the last business cycle peak in 2007, before the Great Recession hit, is growing income inequality. Today’s news about income growth in 2015 is a welcome break from this trend, but does not yet overturn the general pattern that we have seen since 2007. The second benchmark I posit is income growth relative to that of earlier historical epochs. What this benchmark shows is that in the three decades following World War II, income growth was both much faster as well as more broadly shared than it has been since 1979.
By the Numbers: Income and Poverty, 2015
Key numbers from today’s new Census reports, Income and Poverty in the United States: 2015. All dollar values are adjusted for inflation (2015 dollars).
Earnings
Median earnings for men working full time rose 1.5 percent, to $51,212, in 2015. Men’s earnings are down 0.7 percent since 2007, and are still 0.1 percent lower than they were in 2000.
Median earnings for women working full time rose 2.7 percent, to $40,742, in 2015. Women’s earnings are up 1.5 percent since 2007, and are 7.8 percent higher than they were in 2000.
Median earnings for men working full time
- 2015: $51,212
- 2014–2015: 1.5%
- 2007–2015: -0.7%
- 2000–2015: -0.1%
Median earnings for women working full time
- 2015: $40,742
- 2014–2015: 2.7%
- 2007–2015: 1.5%
- 2000–2015: 7.8%
Superb income growth in 2015 nearly single-handedly restored incomes lost in the Great Recession
Today’s report from the Census Bureau shows impressive (and long-awaited) across-the-board improvements to household incomes over 2014–2015, as inflation-adjusted wages improved and unemployment fell (from 6.2 to 5.3 percent) while inflation was absent. Inflation-adjusted wages for women finally exceeded their pre-recession levels in 2015, rising 2.7 percent, while the 1.5 percent increase in men’s earnings leaves them just 0.7 percent down from 2007 levels. These findings reinforce the centrality of wage growth for reestablishing household income gains—as we argued in Raising America’s Pay—and the importance of lowering unemployment. This is for the simple reason that most households, including those with low incomes, rely on labor earnings for the vast majority of their income.
Despite gains in 2015, household incomes have still not fully recovered from the deep losses suffered in the Great Recession—the bottom 95 percent of households still had incomes in 2015 below those of 2007 (while those in the top five percent are now three percent ahead). One more year of modest growth will bring the broad middle class back to pre-recession incomes.
Non-elderly household incomes rebound
The Census data show that from 2014–2015, median household incomes for non-elderly households (those with a head of household younger than 65 years old) increased 4.6 percent, from $60,531 to $63,344. This increase is a superb and most-welcomed improvement. Median household income for non-elderly households in 2015 ($63,344) was 5.0 percent, or $3,304, below its level in 2007—roughly half the total loss that prevailed over the 2007–14 period. The disappointing trends of the Great Recession and its aftermath come on the heels of the weak labor market from 2000–2007, during which the median income of non-elderly households fell significantly, from $69,016 to $66,648, the first time in the post-war period that incomes failed to grow over a business cycle. Altogether, from 2000–2015, the median income for non-elderly households fell from $69,016 to $63,344, a decline of $5,672, or 8.2 percent.
What to watch for in the Census income and poverty data
Next Tuesday, the Census Bureau will release its data on income, poverty, and health insurance coverage for 2015, which will give us a better picture of how working families are—and are not—recovering from the Great Recession. Even in the full business cycle of 2000-2007, earnings and incomes never fully recovered to the pre-recession peaks reached in 2000, so when the Great Recession hit, the economic impacts were especially devastating for many. To the extent the data allow, next week’s release will let us see how much the recovery has improved the economic lot for typical Americans, paying particular attention to differences in the recovery across racial and ethnic groups.
First, EPI researchers will examine the data on median earnings, by gender, race and ethnicity. Hourly wage data for 2015 suggest decent growth between 2014 and 2015 across the board, driven mostly by unexpectedly low inflation, driven mostly by falling oil prices. Median hourly wages grew 1.7 percent between 2014 and 2015. Hand-in-hand with stable average weekly hours, this suggests an uptick in median annual earnings.
We’ll look at changes over the last year, as well as changes since before the Great Recession, and since 2000—the last business cycle peak that can be confidently associated with genuine full employment. Women have already exceeded their 2000 real earnings level; the hourly wage data indicate a return to 2007 prerecession levels of earnings for men in 2015. We’ll also analyze these changes by race and ethnicity to understand how the economy has treated demographic groups. Again, the hourly wage data is the best predictor of what we can expect for these sub-groups. (For a taste of these comparisons, check out EPI’s new State of Working America Data Library.) I expect the 2015 annual earnings data to show a slight decline in the gender wage gap, and the Hispanic-white wage gap, but a slight increase in the black-white wage gap.
Top H-1B employers use visa program for temporary labor—not as bridge to permanent immigration
Hoping to equate the H-1B temporary foreign worker program with permanent immigration, advocates often lump the H-1B visa together with lawful permanent residence (also known as “green card” status). But the H-1B is a temporary, nonimmigrant visa for guestworkers, not a permanent immigration status that would eventually put a migrant worker on a path to American citizenship.
The H-1B program can serve as a bridge to permanent immigration for many educated and skilled foreign workers; in fact, between 2010 and 2014, an average of 44,000 H-1B guestworkers became immigrants (i.e., lawful permanent residents) each year. The U.S. government approved an annual average of 115,000 new H-1B guestworkers over that same timeframe. The H-1B path to a green card is controlled by the employer. The employer—not the H-1B worker—has the discretion of applying for a green card, and as a result, employers hold a lot of power over the hundreds of thousands of H-1B workers here.
The first step an employer must take to put an H-1B nonimmigrant worker on the path to a green card is to file for permanent labor certification with the U.S. Department of Labor (DOL), to check if there are any U.S. workers available to fill the job that the H-1B worker is already doing. (These are sometimes referred to as “PERM” applications or the PERM process, which stands for Program Electronic Review Management, the name of the electronic system used by the DOL.) Public data are available showing which companies applied for permanent labor certification for their H-1B workers and for how many, and these data let us examine whether employers typically use the H-1B program as a bridge to permanent immigration—or not. As the table below shows, the top employers received large numbers of new H-1B workers in fiscal 2014 but applied for very few green cards for their H-1B workers.
Manufacturing job loss: the consequences of malign neglect of the dollar and Chinese overcapacity
Today’s jobs report from the BLS showed that the U.S. manufacturing sector lost 14,000 jobs in August and has now lost 57,000 jobs since January of this year. This job loss is, in part, a consequence of the sharp rise of the dollar in 2014 and 2015, which has gained nearly 20 percent on a broad, trade-weighted basis, as shown below. The rising dollar has reduced the cost of imports, increased the cost of U.S. exports resulting in growing trade deficits. Growing exports support U.S. employment, but growing imports cost U.S. jobs, so the manufacturing decline was entirely predictable from the expected increase in the U.S. trade deficit, which responds to changes in the dollar with a lag of one to two years. Yet the U.S. government continues to do nothing about destructive exchange rate movements, whether they are caused by intentional currency manipulation or more recent, market-driven misalignments.
Data for the U.S. trade deficit in July were also released this morning. The trade deficit in manufactured products (Exhibit 1S) increased 3.1 percent, year to date, relative to the same period last year, despite a decline in the overall U.S. trade deficit. U.S. imports of petroleum products declined sharply in this period, while the trade deficit in non-petroleum goods (which is dominated by trade in manufactures) increased sharply. The single largest cause of the growing manufacturing trade deficit is malign neglect of currency manipulation over the past 20 years by the U.S. government.
China, which has been the most important currency manipulator over the past two decades, was responsible for nearly two thirds (61.3 percent) of the U.S. trade deficit in manufactured goods in 2015. The trade deficit with China increased in July. China has also distorted trade by generating massive amounts of excess production capacity in a wide range of industries, including steel, aluminium, glass, paper and renewable energy products. China’s capacity growth has been fueled by illegal subsidies and other unfair trade practices. A new report from Duke University explores the impacts of overcapacity in China’s steel industry.
Looking under the hood of today’s jobs report
Today’s jobs report came in somewhat underwhelming. This morning, I compared payroll employment growth to weak tea and the labor market saw little to no improvement in other key measures. Yesterday, I urged readers to look under the hood of the headline jobs day numbers and see how well the economy is treating workers across various demographic groups. Today, I’m going to take one statistic from today’s report and see how various groups have fared.
According to the Bureau of Labor Statistics, the official unemployment rate is 4.9 percent. Let’s just remember for a moment that the unemployment rate only counts people actively looking for work, taken as a share of the labor force. So, this leaves out the estimated 2.2 million workers who we expect will return or join the labor force as job opportunities improve. With these missing workers, the unemployment rate would be 6.2 percent. It also leaves out workers who want to work full-time but could only find part-time work or those who might have looked in the last year, but not in the last month. Adding these in, the underemployment rate would be 9.7 percent.
Even with those caveats, I must admit the official unemployment rate is still quite a useful measure. And, along with nominal wage growth, it’s a key measure the Federal Reserve watches when deciding how to act on interest rates. At 4.9 percent, the unemployment rate is 0.3 percentage points higher than it was in 2007, before the recession began, and 0.9 percentage points higher than the last time the economy was at full employment (2000). In fact, for five months in 2000, the unemployment rate was below 4.0 percent, hitting a low of 3.8 percent in April 2000. Examined another way, the unemployment rate is 1.1 times higher today than in 2007 and 1.2 times higher than in 2000.
Looking at the latest wage data by education level
Earlier this week, I analyzed the latest wage data by percentile, which shows that inequality has grown since the last business cycle peak in 2007. Today, I’m going to discuss the latest wage data by education groups. The main takeaways are four-fold. First, women are consistently paid less than men across all education groups. Second, wages have increased more for those with a college or advanced degree than for those with lower levels of education, both in the past year and since 2007. Third, the increase in the college premium since 2007 is dwarfed by the growth in wage inequality generally. And fourth, much of the increase in wages in the last few years has been driven by historically low inflation, as opposed to strong or accelerating nominal wage growth.
The table below shows first half (FH) average wages for 2007, 2015, and 2016 by highest level of education attainment and by gender. You can see that at every level of education, men are paid more than women—illustrating the difficulty of women to educate their way out of the gender wage gap. In fact, the gap grows with increasing levels of education. One particular striking finding is that men with just a bachelor’s degree are paid more, on average, than women with an advanced degree.