Briefing Paper #393
Introduction and executive summary
After a year of solid job growth in 2014, the forecast for 2015 is for modest improvements in unemployment. The national unemployment rate is projected to fall from 5.7 percent in the fourth quarter of 2014 to 5.4 percent in the fourth quarter of 2015. While job growth is expected to continue at the same pace as the previous year, an increase in labor force participation is expected to result in a smaller drop in the unemployment rate than the year before (Moody’s Analytics 2015).
This issue brief examines how unemployment rates, employment-to-population (EPOP) ratios, and long-term unemployment (defined as out of work for six months or more) of whites, Latinos, and African Americans changed nationally and by state between 2013 and 2014, and projects unemployment rates for the fourth quarter of 2015. Together, these measures provide a more complete picture of labor market strength and recovery in the states. For example, in the aftermath of the Great Recession, changes in the unemployment rate have sometimes been distorted by workers leaving the labor market as opposed to more workers becoming employed. True labor market improvements are more likely in those states experiencing both unemployment declines and increases in the share of workers employed (the EPOP ratio). On the other hand, declining unemployment in those states without increasing shares of workers employed may suggest workers are simply dropping out of the labor force.
This report uses a unique analysis of Current Population Survey data and Local Area Unemployment Statistics program data from the Bureau of Labor Statistics to estimate state unemployment rates by race and ethnicity. Forecasts of overall state unemployment rates come from Moody’s Analytics. In order to estimate projected unemployment rates by race, we apply the ratio of each group’s state unemployment rate to the overall unemployment rate in that state (see Methodology).For Hispanics and blacks, this analysis is limited to states with sufficient sample size for reliable estimates. Due to a much smaller sample size, the analysis for Asians is more limited.
Key findings include:
- In the fourth quarter of 2014, nationwide unemployment rates were 4.5 percent for whites, 6.7 percent for Hispanics, 11.0 percent for blacks and 4.4 percent for Asians. These rates are projected to decline modestly through the end of 2015.
- Five years into recovery from the Great Recession, unemployment rates are finally nearing their 2007 levels, but the pace of recovery varies by state for different racial and ethnic groups. In the fourth quarter of 2014, the national white and Hispanic unemployment rates were each within 1 percentage point of prerecession levels while the black unemployment rate was 2.4 percentage points higher than it was at the end of 2007.
- The national black unemployment rate of 11 percent in the fourth quarter of 2014 is still higher than the national unemployment rate at the peak of the recession (9.9 percent in the fourth quarter of 2009, according to the Bureau of Labor Statistics). The modest overall improvement in the national unemployment rate over the next year is projected to bring the black unemployment rate down to 10.4 percent by the fourth quarter of 2015, nowhere close to its prerecession level (8.6 percent).
- Black unemployment rates are projected to fall significantly throughout 2015 in only two states—California and Illinois. Similarly, unemployment rates for Hispanics are projected to fall significantly only in Rhode Island. No states are currently projected to have significantly lower white unemployment rates by the end of 2015.
- In 2014, the annual white unemployment rate was highest in Nevada (7.0 percent), Oregon (6.5 percent), and West Virginia (6.5 percent). Between 2013 and 2014, the white unemployment rate significantly declined in 33 states, but the white employment-to-population (EPOP) ratio significantly increased in only six states.
- Northeastern states continue to have the highest Hispanic unemployment rates. In 2014, the annual Hispanic jobless rate was 16.2 percent in Rhode Island, 11 percent in Massachusetts, and 10.9 percent in Connecticut. In 14 states (data are available for 33 states), the Hispanic unemployment rate in 2014 was significantly lower than in 2013 and nine of those states also had significantly higher Hispanic EPOP ratios.
- In 2014, the annual black unemployment rate was highest in Wisconsin (19.9 percent), Nevada (16.1 percent), Michigan (15.8 percent), and the District of Columbia (15.7 percent). In 15 states (data are available for 30 states and the District of Columbia), the black unemployment rate significantly declined from 2013 to 2014 and in six of those states the black EPOP increased. Yet as seen in comparisons of the fourth quarter 2014 and 2007 unemployment rates, blacks have returned to pre-Great Recession unemployment rates in just two states—Connecticut and South Carolina—but in both cases, the fourth quarter 2007 unemployment rate was well above the national black unemployment rate in that quarter.
- Massachusetts (7.4 percent) had the highest annual Asian unemployment rate in 2014 and Hawaii (3.3 percent) had the lowest rate (among states with a sufficient sample size for reliable statistics).
- In 2014, the share of workers who were long-term unemployed was down significantly for all groups, compared with 2013. The rate was lowest among Hispanic workers (29.8 percent, down 4.8 percentage points), followed by whites (32.1 percent, down 4.1 percentage points), Asians (36.9 percent, down 4.3 percentage points) and African Americans (39.7 percent, down 3.8 percentage points).
White unemployment rates by state
In 2014, white unemployment was highest in Nevada (7.0 percent), Oregon (6.5 percent), and West Virginia (6.5 percent) (Table 1). The lowest white unemployment rates were in North Dakota (2.1 percent), Nebraska (2.6 percent), and South Dakota (2.7 percent).
EPI provides an analysis of unemployment rates by state on a quarterly basis. To view the interactive map allowing you to compare rates across racial and ethnic groups visit http://www.epi.org/resources/economic-indicators/state-jobs-by-race/
Between 2013 and 2014, the annual white unemployment rate declined most in Ohio (1.9 percentage points), Illinois (1.9 percentage points), and New Jersey (1.8 percentage points) (Figure A). In the aftermath of the Great Recession, changes in the unemployment rate have sometimes been distorted by workers leaving the labor market as opposed to more workers becoming employed. Of these three states, New Jersey was the only one in which the employment-to-population ratio for whites was statistically higher in 2014 than it was in 2013. The white unemployment rate significantly declined in 33 states between 2013 and 2014, but the white EPOP significantly increased in only six states. In two states—Oklahoma and New York—the white EPOP significantly declined along with the unemployment rate.
White unemployment rate, by state, annual 2014 and 4th quarter 2007, 2014, and 2015 (projected)
| Annual | Quarterly | |||
|---|---|---|---|---|
| 2014 | 2007Q4 | 2014Q4 | Projected 2015Q4 | |
| UNITED STATES | 4.9% | 4.0% | 4.5% | 4.2% |
| Nevada | 7.0% | 4.4% | 6.7% | 6.1% |
| Oregon | 6.5% | 5.3% | 6.6% | 6.2% |
| West Virginia | 6.5% | 4.1% | 6.3% | 5.9% |
| Rhode Island | 6.2% | 5.2% | 6.1% | 5.2% |
| California | 6.1% | 4.6% | 6.0% | 5.1% |
| Kentucky | 6.1% | 5.1% | 5.4% | 4.8% |
| Alaska | 5.7% | 4.5% | 4.8% | 4.1% |
| Michigan | 5.7% | 6.1% | 5.0% | 4.5% |
| Tennessee | 5.7% | 4.4% | 6.2% | 5.5% |
| Arizona | 5.5% | 3.1% | 5.2% | 4.5% |
| Arkansas | 5.5% | 4.6% | 4.9% | 4.3% |
| Maine | 5.5% | 4.7% | 5.4% | 5.0% |
| New Jersey | 5.5% | 3.9% | 5.1% | 4.7% |
| Illinois | 5.4% | 4.3% | 4.8% | 4.0% |
| Connecticut | 5.3% | 3.5% | 5.5% | 5.0% |
| Mississippi | 5.3% | 3.8% | 5.0% | 4.3% |
| Washington | 5.3% | 4.7% | 5.0% | 4.8% |
| Indiana | 5.2% | 4.2% | 4.7% | 4.4% |
| South Carolina | 5.1% | 3.6% | 5.4% | 5.0% |
| Missouri | 5.0% | 4.4% | 4.3% | 3.9% |
| North Carolina | 5.0% | 4.0% | 4.8% | 4.8% |
| Alabama | 4.9% | 3.0% | 4.7% | 4.1% |
| Florida | 4.9% | 3.8% | 4.6% | 4.5% |
| Pennsylvania | 4.8% | 4.1% | 4.2% | 4.5% |
| New York | 4.7% | 3.7% | 4.4% | 3.9% |
| Ohio | 4.7% | 4.6% | 4.0% | 3.9% |
| Delaware | 4.6% | 3.0% | 4.9% | 4.3% |
| Maryland | 4.6% | 2.7% | 4.0% | 3.4% |
| Massachusetts | 4.6% | 4.5% | 4.5% | 4.0% |
| Georgia | 4.5% | 3.3% | 4.9% | 4.3% |
| Hawaii | 4.5% | 3.9% | 4.2% | 3.7% |
| Virginia | 4.5% | 2.9% | 4.5% | 4.2% |
| New Mexico | 4.3% | 3.0% | 4.2% | 3.6% |
| Wisconsin | 4.3% | 4.0% | 3.9% | 3.1% |
| Idaho | 4.2% | 3.1% | 3.6% | 3.8% |
| Iowa | 4.2% | 3.5% | 3.9% | 3.4% |
| Louisiana | 4.2% | 2.2% | 4.3% | 3.8% |
| Montana | 4.2% | 3.3% | 3.7% | 3.7% |
| Vermont | 4.2% | 4.0% | 4.4% | 3.8% |
| Colorado | 4.1% | 3.5% | 3.3% | 3.0% |
| Kansas | 4.1% | 3.5% | 3.9% | 3.6% |
| New Hampshire | 4.1% | 3.4% | 4.0% | 3.3% |
| Wyoming | 4.1% | 2.4% | 4.0% | 4.0% |
| Oklahoma | 3.7% | 3.0% | 3.4% | 3.5% |
| Texas | 3.7% | 3.5% | 3.4% | 3.6% |
| Utah | 3.7% | 2.4% | 3.5% | 3.6% |
| Minnesota | 3.2% | 4.2% | 2.8% | 2.9% |
| District of Columbia | 2.8% | 1.7% | 2.9% | 2.8% |
| South Dakota | 2.7% | 2.0% | 2.6% | 2.3% |
| Nebraska | 2.6% | 2.6% | 2.5% | 2.3% |
| North Dakota | 2.1% | 2.0% | 2.1% | 2.2% |
Note: States are arranged from highest to lowest 2014 annual unemployment rate.
Source: EPI analysis of Current Population Survey data and unemployment projections from Moody's Analytics
In the fourth quarter of 2014, the white unemployment rate was 4.5 percent, nationally. This is very close to the prerecession rate of 4.0 percent for the fourth quarter of 2007 and we project that rate will remain fairly stable through the end of 2015. Given the likely margins of error in the state-race projection estimates, we have decided to focus on unemployment rate changes of 2 percentage points or more when discussing “significant” changes in projected unemployment rates (see Methodology section for details). Based on this criterion, no states are projected to have significantly different white unemployment rates by the end of 2015 (Table 1).
States with significant changes in the white unemployment rate and EPOP, 2013–2014
| Unemployment rate | EPOP | |
|---|---|---|
| Ohio | -1.9 | |
| Illinois | -1.9 | |
| New Jersey | -1.8 | 1.3 |
| Pennsylvania | -1.6 | |
| Indiana | -1.6 | 2.0 |
| Massachusetts | -1.5 | 1.2 |
| Kentucky | -1.5 | |
| Idaho | -1.5 | |
| Michigan | -1.4 | |
| Nevada | -1.3 | |
| Colorado | -1.3 | |
| California | -1.3 | |
| Washington | -1.2 | |
| North Carolina | -1.2 | |
| New York | -1.2 | -0.8 |
| Maine | -1.2 | |
| Georgia | -1.2 | |
| Wisconsin | -1.1 | |
| Connecticut | -1.1 | 2.2 |
| UNITED STATES | -1.1 | 0.2 |
| Texas | -1.0 | |
| Rhode Island | -1.0 | |
| New Hampshire | -1.0 | |
| Minnesota | -1.0 | |
| Delaware | -1.0 | 1.8 |
| Tennessee | -0.9 | |
| South Carolina | -0.9 | |
| Missouri | -0.9 | |
| District of Columbia | -0.9 | |
| Oregon | -0.8 | 1.3 |
| Oklahoma | -0.8 | -2.1 |
| Maryland | -0.8 | |
| Kansas | -0.8 | |
| Florida | -0.5 |
Note: The figure shows only states that had a statistically significant change in the white unemployment rate between 2013 and 2014. EPOP stands for the employment-to-population ratio.
Source: EPI analysis of Current Population Survey data
Another indicator of how much state labor markets have improved over the past year is the share of unemployed workers who have been out of work for at least six months, known as the “long-term unemployed.” Nationally, the share of long-term unemployed white workers declined 4.1 percentage points between 2013 and 2014, to 32.1 percent (Figure B). Similarly, 13 states saw significant declines in long-term unemployment among whites over the same period, with the biggest changes taking place in North Carolina (-9.6 percentage points), Oregon (-9.3 percentage points), and Florida (-8.7 percentage points). Long-term unemployment significantly increased for whites in Alabama (10.1 percentage points). The states with the highest shares of long-term unemployed white workers in 2014 were New Jersey (41.9 percent), Rhode Island (39 percent), Illinois (39 percent), and Nevada (38.6 percent). Of these, Rhode Island was the only state where the 2014 rate was statistically lower than the 2013 rate.
Share of whites who are long-term unemployed, by state, 2013–2014
| 2013 | 2014 | |
|---|---|---|
| New Jersey | 44.3% | 41.9% |
| Rhode Island* | 45.8% | 39.0% |
| Illinois | 39.2% | 39.0% |
| Nevada | 43.4% | 38.6% |
| Connecticut | 42.7% | 38.0% |
| Florida* | 46.6% | 37.9% |
| North Carolina* | 45.6% | 36.0% |
| Massachusetts | 35.4% | 36.0% |
| Alabama* | 25.7% | 35.8% |
| California* | 43.0% | 35.6% |
| New York* | 38.9% | 34.3% |
| Georgia | 37.2% | 34.1% |
| Maryland | 39.7% | 33.6% |
| Virginia | 36.5% | 33.0% |
| Michigan | 34.7% | 32.8% |
| UNITED STATES* | 36.2% | 32.1% |
| New Hampshire | 31.1% | 31.6% |
| West Virginia | 34.3% | 31.4% |
| Colorado* | 38.8% | 31.0% |
| Ohio | 34.5% | 31.0% |
| Pennsylvania* | 38.7% | 30.7% |
| Kentucky | 31.0% | 30.5% |
| South Carolina | 34.8% | 30.4% |
| Mississippi | 35.7% | 30.2% |
| Missouri* | 38.6% | 29.9% |
| Oregon* | 39.0% | 29.7% |
| Delaware* | 36.1% | 29.4% |
| Washington | 30.2% | 29.3% |
| Tennessee | 31.3% | 29.0% |
| Louisiana | 29.6% | 28.7% |
| Indiana | 26.2% | 28.2% |
| Texas | 28.9% | 27.8% |
| Arizona* | 35.1% | 27.1% |
| Wisconsin* | 32.9% | 26.9% |
| Minnesota | 26.3% | 26.9% |
| Idaho | 26.8% | 26.5% |
| Kansas | 28.8% | 25.7% |
| Maine* | 29.3% | 24.5% |
| Vermont | 23.4% | 24.3% |
| Arkansas | 25.2% | 22.6% |
| Oklahoma | 24.6% | 22.1% |
| Montana | 22.9% | 20.4% |
| Wyoming | 20.4% | 19.6% |
| Utah | 22.5% | 19.2% |
| Alaska | 22.7% | 18.5% |
| Iowa | 21.8% | 18.3% |
| South Dakota | 15.9% | 17.3% |
These trends in long-term unemployment among white workers also likely reflect differences in state unemployment insurance (UI) policies. Because receipt of UI benefits is contingent upon active job-searching, benefits can keep workers actively searching for jobs and thus classified as officially unemployed (instead of “out of the labor force”) while jobless. At the end of 2013, when Congress ended the federal UI extensions that provided benefits to workers unemployed beyond 26 weeks, access to benefits became increasingly skewed based on where workers live since states vary in the duration of regular UI benefits and the percentage of unemployed workers who receive them (i.e., the recipiency rate).
For example, in North Carolina and Florida where UI benefit duration was cut below 26 weeks and short-term (less than 26 weeks) white UI recipiency rates are among the lowest in the nation, there have been particularly large drops in long-term unemployment. This could reflect workers who have been unemployed for a long period and who now have lost some incentive to continue actively looking for work. This interpretation is strengthened by the fact that despite the large decline in the share of whites who are long-term unemployed and significant declines in the white unemployment rates in these states, neither state showed a significantly higher white EPOP ratio (see Figure A). Also, the white EPOP ratios in North Carolina and Florida were not significantly higher than the national white rate or rates in neighboring states. On the other hand, in New Jersey—a state that provides up to 26 weeks of benefits and has the highest short-term UI recipiency rate in the country (Kimball and McHugh 2015)—the share of long-term job seekers remains high although there were significant improvements in both the unemployment rate and the EPOP ratio (see Figure A).
Latino unemployment rates by state
Northeastern states continue to have the highest Hispanic unemployment rates among the states with a sufficient sample size for reliable statistics, as shown in Table 2. In 2014 the Hispanic unemployment rate was highest in Rhode Island (16.2 percent), Massachusetts (11 percent), and Connecticut (10.9 percent). Utah (4.4 percent), Texas (5.3 percent), and Kansas (5.4 percent) had the lowest Hispanic unemployment rates in 2014.
Hispanic unemployment rate, by state, annual 2014 and 4th quarter 2007, 2014, and 2015 (projected)
| Annual | Quarterly | |||
|---|---|---|---|---|
| 2014 | 2007Q4 | 2014Q4 | Projected 2015Q4 | |
| UNITED STATES | 7.4% | 5.9% | 6.7% | 6.4% |
| Rhode Island | 16.2% | 7.4% | 12.9% | 10.9% |
| Massachusetts | 11.0% | * | 12.5% | 11.0% |
| Connecticut | 10.9% | 8.3% | 10.1% | 9.2% |
| Alaska | 10.0% | * | * | * |
| Oregon | 9.7% | * | 9.6% | 9.0% |
| Pennsylvania | 9.6% | * | 7.8% | 8.3% |
| Delaware | 9.3% | * | * | * |
| Wisconsin | 9.1% | * | * | * |
| Washington | 9.0% | 5.6% | 8.7% | 8.2% |
| Michigan | 8.8% | * | * | * |
| Louisiana | 8.6% | * | * | * |
| California | 8.5% | 6.9% | 7.8% | 6.7% |
| Indiana | 8.5% | * | * | * |
| New York | 8.5% | 6.9% | 7.2% | 6.3% |
| Iowa | 8.4% | * | * | * |
| Arizona | 8.3% | 5.5% | 8.0% | 6.8% |
| Illinois | 8.1% | 5.7% | 7.7% | 6.5% |
| Idaho | 8.0% | * | 5.2% | 5.4% |
| New Jersey | 7.6% | 5.2% | 7.7% | 7.2% |
| Nevada | 7.5% | 6.7% | 6.0% | 5.4% |
| New Mexico | 7.5% | 4.2% | 6.5% | 5.7% |
| Minnesota | 7.0% | * | * | * |
| Hawaii | 6.7% | * | * | * |
| Florida | 6.6% | 5.6% | 6.3% | 6.0% |
| North Carolina | 6.6% | 6.3% | 4.7% | 4.7% |
| Colorado | 6.4% | 5.2% | 5.2% | 4.7% |
| Virginia | 6.1% | 3.6% | 5.6% | 5.3% |
| Georgia | 6.0% | 7.1% | 4.9% | 4.4% |
| Oklahoma | 6.0% | * | 5.7% | 5.8% |
| Nebraska | 5.8% | * | * | * |
| Maryland | 5.6% | 2.0% | 5.5% | 4.6% |
| Kansas | 5.4% | * | 4.5% | 4.2% |
| Texas | 5.3% | 4.6% | 5.2% | 5.5% |
| Utah | 4.4% | 3.7% | 4.0% | 4.1% |
Note: The * indicates missing values because data are not available due to insufficient sample size.
Source: EPI analysis of Current Population Survey data and unemployment projections from Moody's Analytics
Between 2013 and 2014, the annual Hispanic unemployment rate declined most in Wisconsin (5.5 percentage points), Rhode Island (4.7 percentage points), and Michigan (4.2 percentage points) (Figure C). In Wisconsin and Michigan, these lower unemployment rates were consistent with significantly higher Hispanic employment: The employment-to-population ratio rose 4 percentage points in Wisconsin and 6.4 percentage points in Michigan. Among states for which reliable estimates could be calculated, the 2014 Hispanic unemployment rate was significantly lower than the year before in 14 states. Nine of these states also had a significantly higher Hispanic EPOP ratios in 2014, indicating that improvements in the Hispanic unemployment rate were largely driven by increased employment rather than people leaving the labor force.
States with significant changes in the Hispanic unemployment rate and EPOP, 2013–2014
| Unemployment rate | EPOP | |
|---|---|---|
| Wisconsin | -5.5 | 4.0 |
| Rhode Island | -4.7 | |
| Michigan | -4.2 | 6.4 |
| Nevada | -3.6 | 3.0 |
| Colorado | -3.6 | 3.3 |
| Illinois | -3.1 | |
| North Carolina | -3.0 | |
| Massachusetts | -2.9 | 5.5 |
| Pennsylvania | -2.7 | |
| Kansas | -2.7 | 6.8 |
| New York | -2.5 | |
| California | -1.7 | 0.8 |
| UNITED STATES | -1.7 | 1.2 |
| Texas | -1.5 | 1.3 |
| Florida | -1.4 | 1.5 |
Note: The figure shows only states that had a statistically significant change in the Hispanic unemployment rate between 2013 and 2014.
Source: EPI analysis of Current Population Survey data
The national Hispanic unemployment rate at the end of 2015 is projected to be nearly unchanged—6.4 percent compared with 6.7 percent at the end of 2014. The prerecession Hispanic unemployment rate was 5.9 percent for the fourth quarter of 2007. Rhode Island is the only state expected to experience a significant decline in the Hispanic unemployment rate over the year (Table 2).
In 2014, the long-term unemployment share among Hispanic workers (29.8 percent) was the lowest of any racial or ethnic group—and down 4.8 percentage points from 2013 (Figure D). Among states with a large enough sample size for reliable estimates, the biggest improvements occurred in New Jersey (a change of -15.6 percentage points), Illinois (-11.4 percentage points), and Rhode Island (-10.7 percentage points). A total of five states saw a statistically significant decrease in Hispanic long-term unemployment between 2013 and 2014. Conversely, the rates were not statistically different between 2013 and 2014 in the states with the largest shares of long-term unemployed in 2014: Florida (43.6 percent), New Mexico (43.4 percent), and Connecticut (38.8 percent).
Share of Hispanics who are long-term unemployed, by state, 2013–2014
| 2013 | 2014 | |
|---|---|---|
| Florida | 42.8% | 43.6% |
| New Mexico | 47.0% | 43.4% |
| Connecticut | 42.7% | 38.8% |
| New York* | 47.9% | 38.6% |
| New Jersey* | 49.0% | 33.4% |
| Nevada | 32.2% | 33.4% |
| California* | 36.9% | 33.3% |
| Rhode Island* | 43.3% | 32.7% |
| Colorado | 34.0% | 30.8% |
| UNITED STATES* | 34.6% | 29.8% |
| Arizona | 27.4% | 26.9% |
| Illinois* | 36.8% | 25.4% |
| Texas | 24.9% | 21.6% |
African American unemployment rates by state
In 2014, black unemployment rates were highest in Wisconsin (19.9 percent), Nevada (16.1 percent), Michigan (15.8 percent), and the District of Columbia (15.7 percent) (among states with sufficient sample size for reliable statistics), as shown in Table 3. Virginia (8.0 percent), Maryland (8.8 percent), and Delaware (9.1 percent) had the lowest black unemployment rates. The lowest black unemployment rate, in Virginia, was higher than the highest white rate of 7.0 percent in Nevada.
Black unemployment rate, by state, annual 2014 and 4th quarter 2007, 2014, and 2015 (projected)
| Annual | Quarterly | |||
|---|---|---|---|---|
| 2014 | 2007Q4 | 2014Q4 | Projected 2015Q4 | |
| UNITED STATES | 11.4% | 8.6% | 11.0% | 10.4% |
| Wisconsin | 19.9% | * | * | * |
| Nevada | 16.1% | * | * | * |
| Michigan | 15.8% | 15.3% | 16.3% | 14.7% |
| District of Columbia | 15.7% | 9.7% | 15.1% | 14.6% |
| Iowa | 15.6% | * | * | * |
| Illinois | 14.7% | 12.2% | 13.6% | 11.4% |
| Missouri | 14.4% | 11.6% | 13.1% | 12.0% |
| Washington | 14.3% | * | * | * |
| California | 14.0% | 9.8% | 14.9% | 12.8% |
| Connecticut | 13.1% | 12.6% | 11.4% | 10.4% |
| Indiana | 13.1% | 11.3% | * | * |
| Georgia | 12.5% | 8.1% | 12.1% | 10.8% |
| Mississippi | 12.5% | 10.8% | 11.8% | 10.3% |
| Alabama | 12.3% | 5.3% | 9.1% | 7.9% |
| New Jersey | 12.0% | 8.4% | 11.2% | 10.5% |
| Ohio | 11.9% | 13.7% | 13.0% | 12.5% |
| Minnesota | 11.7% | * | * | * |
| Colorado | 11.5% | * | * | * |
| Rhode Island | 11.5% | * | * | * |
| Tennessee | 11.5% | 9.3% | 10.2% | 9.0% |
| New York | 11.1% | 7.7% | 11.1% | 9.7% |
| Florida | 10.9% | 6.1% | 9.6% | 9.3% |
| Pennsylvania | 10.8% | 6.9% | 11.1% | 11.8% |
| Massachusetts | 10.6% | * | * | * |
| Louisiana | 10.4% | 8.0% | 10.5% | 9.3% |
| Arkansas | 10.0% | 8.8% | 11.9% | 10.4% |
| North Carolina | 10.0% | 8.1% | 9.7% | 9.7% |
| South Carolina | 9.8% | 10.8% | 10.1% | 9.4% |
| Kentucky | 9.6% | * | * | * |
| Texas | 9.5% | 8.1% | 9.8% | 10.4% |
| Delaware | 9.1% | 5.0% | 8.6% | 7.7% |
| Maryland | 8.8% | 5.4% | 9.4% | 7.9% |
| Virginia | 8.0% | 5.1% | 7.5% | 7.0% |
Note: The * indicates missing values because data are not available due to insufficient sample size.
Source: EPI analysis of Current Population Survey data and unemployment projections from Moody's Analytics
Between 2013 and 2014, the annual black unemployment rate declined most in Arkansas (6.5 percentage points), Indiana (4.6 percentage points), and Tennessee (3.6 percentage points) (Figure E). Of these, only Arkansas had a significantly higher black employment-to-population ratio in 2014 (from 46.8 percent to 50.1 percent). Among states for which reliable estimates could be calculated, 15 states experienced a significant decline in the black unemployment rate between 2013 and 2014 and in six of those states the black EPOP increased. On the other hand, between 2013 and 2014 the black unemployment rate significantly increased in Missouri (3.2 percentage points) and Wisconsin (4.8 percentage points).
States with significant changes in the black unemployment rate and EPOP, 2013–2014
| Unemployment rate | EPOP | |
|---|---|---|
| Arkansas | -6.5 | 3.3 |
| Indiana | -4.6 | |
| Tennessee | -3.6 | |
| Pennsylvania | -3.5 | |
| Minnesota | -3.5 | |
| Ohio | -2.8 | 2.9 |
| North Carolina | -2.6 | |
| Illinois | -2.6 | 2.3 |
| Delaware | -2.2 | 3.4 |
| South Carolina | -2.0 | |
| Louisiana | -1.9 | 3.1 |
| California | -1.9 | 3.6 |
| UNITED STATES | -1.7 | 1.1 |
| New York | -1.6 | |
| Florida | -1.6 | |
| Maryland | -1.3 | |
| Missouri | 3.2 | |
| Wisconsin | 4.8 | -4.5 |
Note: The figure shows only states that had a statistically significant change in the black unemployment rate between 2013 and 2014.
Source: EPI analysis of Current Population Survey data
A national black unemployment rate of 10.4 percent is projected for the end of 2015, only slightly below the 11.0 percent rate at the end of 2014 and still nearly 2 percentage points above the prerecession rate of 8.6 percent for the fourth quarter of 2007. At the state level, significant reductions are expected in California and Illinois. The black unemployment rate is projected to be below 10 percent by the end of 2015 in 10 states (Table 3).
In 2014, long-term unemployment among African American workers (39.7 percent) was the highest of any racial or ethnic group, although it was down 3.7 percentage points from 2013 (Figure F). Among states with a large enough sample size for reliable estimates, only three had significant declines in long-term unemployment between 2013 and 2014: North Carolina (14.4 percentage points), Florida (10.0 percentage points), and Texas (8.2 percentage points). In 2014, the highest shares of long-term unemployed black workers were in the District of Columbia (56.3 percent), Illinois (52.7 percent), Alabama (48.9 percent), and New Jersey (48.6 percent). These trends in black long-term unemployment are similar to those of whites and have likely also been affected by differences in UI policies between states. In Maryland and Delaware, where the black unemployment rate is among the lowest in the nation, the share of black workers who are long-term unemployed increased between 2013 and 2014. This suggests that even in states where job prospects have improved, there are subgroups of workers who face ongoing challenges in finding work. (an increase of 9 percentage points from 2013 in both states).
Share of blacks who are long-term unemployed, by state, 2013–2014
| 2013 | 2014 | |
|---|---|---|
| District of Columbia | 53.4% | 56.3% |
| Illinois | 50.1% | 52.7% |
| Alabama | 43.6% | 48.9% |
| New Jersey | 51.0% | 48.6% |
| New York | 49.4% | 47.2% |
| Nevada | 49.5% | 46.6% |
| Maryland* | 36.9% | 45.9% |
| Georgia | 48.2% | 45.6% |
| Connecticut | 48.1% | 43.0% |
| Michigan | 44.2% | 41.9% |
| Delaware* | 31.0% | 40.0% |
| Florida* | 49.9% | 39.9% |
| UNITED STATES* | 43.5% | 39.7% |
| Mississippi | 47.3% | 39.7% |
| South Carolina | 37.5% | 39.1% |
| Pennsylvania | 44.2% | 39.0% |
| California | 43.2% | 38.2% |
| Ohio | 35.0% | 38.1% |
| Wisconsin | 46.0% | 37.7% |
| Louisiana | 41.6% | 36.4% |
| Virginia | 36.8% | 35.6% |
| Missouri | 42.5% | 34.3% |
| North Carolina* | 46.0% | 31.6% |
| Texas* | 37.5% | 29.4% |
| Tennessee | 34.1% | 26.7% |
Asian unemployment rates by state
Due to a much smaller population size, there is considerably less state level data available for the Asian population. Among states with sufficient sample size for reliable statistics, Massachusetts had the highest annual Asian unemployment rate in 2014 (7.4 percent) and Hawaii had the lowest rate (3.3 percent), as shown in Table 4. Long-term unemployment among Asian workers (36.9 percent) was the second highest of any racial or ethnic group studied here, and was down 4.3 percentage points from 2013. A national Asian unemployment rate of 4.2 percent is projected for the end of 2015, little changed from the 4.4 percent rate at the end of 2014, and within 1 percentage point of the prerecession rate of 3.5 percent for the fourth quarter of 2007.
Asian unemployment rate, by state, annual 2014 and 4th quarter 2007, 2014, and 2015 (projected)
| 2014 | 2007Q4 | 2014Q4 | Projected 2015Q4 | |
|---|---|---|---|---|
| UNITED STATES | 4.9% | 3.5% | 4.4% | 4.2% |
| Massachusetts | 7.4% | * | 7.6% | 6.7% |
| California | 5.9% | 4.7% | 5.2% | 4.4% |
| New York | 5.9% | 3.2% | 4.6% | 4.1% |
| Illinois | 5.3% | 3.4% | 5.2% | 4.4% |
| New Jersey | 5.0% | 2.3% | 4.7% | 4.4% |
| Texas | 4.0% | 3.0% | 3.6% | 3.8% |
| Hawaii | 3.3% | 2.4% | 3.1% | 2.8% |
Note: The * indicates missing values because data are not available due to insufficient sample size.
Source: EPI analysis of Current Population Survey data and unemployment projections from Moody's Analytics
Conclusion
Five years into recovery from the Great Recession, unemployment rates are finally nearing their 2007 levels, but the pace of recovery varies by state for different racial and ethnic groups. In the fourth quarter of 2014, the national white and Hispanic unemployment rates were each within 1 percentage point of prerecession levels while the black unemployment rate was 2.4 percentage points higher than it was at the end of 2007. Although long-term unemployment was down significantly for all groups in 2014, it remained above historic norms, revealing weaknesses in the labor market (Bivens and Shierholz 2014).
The economy continues to get closer to full recovery or pre-Great Recession labor market conditions each month. However, in some states and for some groups of workers, rates of unemployment are still unacceptably high. Nationally, this can be seen in the still-absent growth of nominal wages even as labor markets tighten on the employment side (Gould 2015). The lack of accelerating wage growth is a clear sign that large amounts of slack remain in the overall labor market, and this slack is almost surely much greater for particular groups in the workforce. Policies aimed at achieving full employment, as opposed to simply returning to the pre-Great Recession status quo, raise the bar to a point at which all who are willing to work at the prevailing wage rate are employed.
About the author
Valerie Wilson directs the Economic Policy Institute’s Program on Race, Ethnicity, and the Economy (PREE). PREE works to advance policies that enable people of color to participate fully in the American economy and benefit equitably from gains in prosperity. As director of PREE, Wilson oversees reports and policy analyses on the economic condition of America’s people of color. Prior to joining the Economic Policy Institute, Wilson was vice president of research at the National Urban League. She received her Ph.D. in economics from the University of North Carolina at Chapel Hill.
Methodology
The quarterly unemployment rate estimates in this issue brief are based on the Local Area Unemployment Statistics (LAUS) and the Current Population Survey (CPS) from the Bureau of Labor Statistics. The overall state unemployment rate is taken directly from the LAUS. CPS six-month ratios are applied to LAUS data to calculate the rates by race and ethnicity. For each state subgroup, we calculate the unemployment rate using the past six months of CPS data. We then find the ratio of this subgroup rate to the state unemployment rate using the same period of CPS data. This gives us an estimate of how the subgroup compares with the state overall.
While this methodology allows us to calculate unemployment-rate estimates at the state level by race by quarter, it is less precise at the national level than simply using the CPS. Thus, the national-level estimates may differ from direct CPS estimates.
For our projections for the fourth quarter of 2015, we use the same method but modify it slightly. We find the subgroup state ratios from the most recent six months of data, and then multiply this ratio by the projected state unemployment rate for a given quarter.
In many states, the sample size of these subgroups is not large enough to create an accurate estimate of their unemployment rate. We only report data for groups which had, on average, a sample size of at least 700 in the labor force for each six-month period.
Note that unemployment rate changes that would seem significant if we were reporting them nationally are not when reporting by race by state. As the sample size decreases, the margin of error increases. Projection data are also imprecise. For these reasons, in this paper, when we are discussing projected changes in unemployment rates by state and race and ethnicity between the fourth quarter of 2014 and the fourth quarter of 2015, we consider only changes of 2 percentage points or more to be significant.
References
Bivens, Josh, and Heidi Shierholz. 2014. Lagging Demand, Not Unemployability, Is Why Long-term Unemployment Remains So High. Economic Policy Institute Briefing Paper No. 381.
Current Population Survey basic monthly microdata. Various years. Survey conducted by the Bureau of the Census for the Bureau of Labor Statistics [machine-readable microdata file].
Gould, Elise. 2015. “Nominal Wage Growth Still Far Below Target.” Working Economics (Economic Policy Institute blog), February 6.
Kimball, Will, and Rick McHugh. 2015. How Low Can We Go? State Unemployment Insurance Programs Exclude Record Numbers of Jobless Workers. Economic Policy Institute Briefing Paper No. 392.
Local Area Unemployment Statistics. Various years. Data produced by the Bureau of Labor Statistics.
Moody’s Analytics. 2015. Moody’s Economy.com, MyEconomy.com [subscription-only database].
Wilson, Valerie. 2014. “Keep the Jobs Coming! People of Color Have Actually Benefited More from Job Growth This Year.” Working Economics (Economic Policy Institute blog), November 7.
