Last week, I posted Jobs recovery: a tale of two Georgias, detailing the mounting evidence that job creation in the state has been disproportionately fueled by the Atlanta metro area and a handful of other nearby metro areas.
That post used data from May’s survey of payroll establishments. That’s the data used on the national level to determine job growth (or loss) each month. The unemployment rate is determined by a different set of data: a survey of households to see who claims to be working, looking for work, etc. Those numbers for May were released this past Thursday by the Georgia Department of Labor.
Sometimes the two data sets diverge dramatically. The payroll jobs estimate shows Georgia gaining a net 34,000 jobs from May 2011 to May 2012, while the household survey shows almost 63,000 more employed persons. The divergence is in part due to the broader definition of “work” in the household survey, which includes agricultural work, unpaid work, self-employment, and other categories that aren’t counted in the payroll survey.
Over time, the month to month inconsistencies in the surveys tend to level out.
Assessing the data is all the more difficult because of seasonal adjustments. Employment follows clear seasonal trends, which economists smooth out with adjustments. The Georgia Department of Labor uses seasonal adjustments for the statewide unemployment rate, but not for the rates for individual metro areas or cities.
Anyway . . . on to Georgia’s data for May from the survey of households, on which the unemployment rate is based. I’m going to use numbers that are NOT seasonally adjusted. If you want to look at some of the numbers yourself, start at the landing page for Georgia’s most recent data.
Georgia’s civilian labor force increased by 31,208 from May 2011 to May 2012, considerably less than we would expect given current population growth. The declining labor force participation rate is a bad sign, although it could in part be due to increased retirements and to more workers returning to school. But 62,902 more Georgians aged 16 and over reported themselves to be working compared to a year ago, a decent increase. The two numbers combined to cut the state’s NSA unemployment rate from 9.6 percent in May 2011 to 8.9 percent in May 2012.
The following metros saw an increase in the number of employed persons compared to a year ago:
- Albany +879
- Athens-Clarke County +4,042
- Atlanta +40,702
- Brunswick +1,020
- Columbus +2,663
- Gainesville +6,074 (amazingly good growth)
- Macon +2,601
- Valdosta + 1,543
- Warner Robins +402
Warner Robins and Albany showed job losses in the payroll data, so this is encouraging news. The employment picture also looks more vigorous here for the rest of these metro areas. Let’s hope that in the coming months the payroll data will catch up to these numbers.
The following metros saw a decrease in the number of employed persons in May 2012 compared to May 2011:
- Augusta-Richmond County -5,509
- Dalton -3,391
- Hinesville-Ft. Stewart – 168
- Rome -511
- Savannah -2,048
By contrast, Atlanta saw the civilian labor force increase by 20,085 (less than we’d expect given population growth). But the increase in employed persons was twice as great, bringing the Atlanta metro unemployment rate from 9.5 percent in May 2011 to 8.6 percent in May 2012.
Georgia’s a big state, so it’s not surprising that some areas would be showing some real gains while others are still shedding jobs. But I’d sure like to see forthright discussion about those differences and the possible implications for public policy.