Jobs numbers fly fast and furious in election campaigns. VTDigger has parsed the details to help you keep facts straight and figure out what they mean.
Following are fact-checks of prominent claims about job creation and loss made by the incumbent Democratic Gov. Peter Shumlin and the Vermont Republican Party.
Vermont GOP claims
True.
This reference comes from monthly employment estimates based on household surveys (not an actual count of jobs). This is the only source of monthly employment data — all other data take longer to produce. But, the expedience costs some accuracy. Numbers that take longer to crunch often differ from the monthly estimates.
From what is known so far, for three months in a row, the seasonally adjusted figures rose in unemployment rates, declined in the number of Vermonters who report they’re working, and increased in the number who say they’re unemployed.
Between April and May, the employment level rose from 339,500 to 339,650. Since then, it’s fallen each month, reaching 335,800 in August.
Unemployment follows a converse path: From 12,100 in March, the number fell to 11,500 in April, where it stayed through May. Then the number of unemployed rose each month, reach 14,400 in August.
”Today, Vermont’s workforce is an astonishing 9,750 workers smaller than it was when Governor Shumlin took office in January 2011,” Sunderland said in the news release, which is posted on the party’s website.
True-ish.
As noted above, the advantage of monthly employment estimates’ quick turnaround often is offset by their lack of accuracy. Seasonal adjustments and statistical benchmarking help correct for this, but the figures remain estimates.
According to the seasonally adjusted data, the state’s labor force totaled 359,800 in December 2010, the month before Shumlin took office. As of August 2014, that number fell to 350,200 — a loss of 9,600 people in the workforce. (If measuring back to Shumlin’s first month, January 2011, the loss does equal 9,750.)
Remember, that’s not the number of people working. That’s the number of people who respond to a household survey that they are ready, willing and able to work. See below for figures of how many people are employed vs. unemployed.
The seasonality of Vermont’s economy, however, means month-to-month employment comparisons are always appropriate.
Sunderland also said the August data also show there are 2,600 fewer Vermonters working today than when Governor Shumlin was first inaugurated.
Maybe not.
The monthly labor report estimates 335,800 Vermonters were employed in August 2014, compared to 337,950 in December 2010. That’s a loss of 2,150. (If measuring back to Shumlin’s first month, January 2011, the loss does equal 2,600.)
The seasonality of Vermont’s economy, however, means month-to-month employment comparisons are always appropriate.
A more accurate — though still not perfect — count of jobs is the Quarterly Census of Employment and Wages. This doesn’t come out as often, but it’s more reliable than the self-reporting from a sample of the population. Contrary to job losses, the census shows job creation ranging from about 6,000 to 9,000 positions, as noted below.
The state’s unemployment rate rose to 4.1 percent in August and is now at its highest level of 2014, Sunderland said.
True.
Vermont started 2014 with an estimated unemployment rate of 4.0 percent in January, which fell to 3.3 percent in April and May. It has since increased to 4.1 percent, the highest rate of the year.
”Most shockingly, if Vermont’s workforce stayed level since the Governor first took office, the small number of working Vermonters today would produce an unemployment rate of 6.9 percent,” the news release said. “That is nearly a full percentage point higher than it was when Governor Shumlin was inaugurated in January of 2011.”
A fair point.
This hypothetical scenario would peg the unemployment rate today at 6.7 percent, not 6.9 percent.
The labor force in January 2011 was 359,950, and unemployment was 6.0 percent. As of August, the labor force dropped to 350,200, when unemployment reached 4.1 percent. If the labor force today were 359,950, then today’s number of employed people (335,800) would imply 24,150 unemployed, which in turn would lead to an unemployment rate of 6.7 percent.
Sunderland said the point remains that, if only no one had left the workforce and no one lost their jobs, the hypothetical rate would be higher than the 6 percent unemployment of Shumlin’s first month on the job.
The scenario illustrates the importance of the size of the labor force on statistical estimates of unemployment. It’s possible for unemployment rates to drop not as a result of new people getting jobs, but instead because the workforce shrinks (when people retire, move out of state or give up on finding a job, for example).
Gov. Peter Shumlin’s claims
True.
The official unemployment rate was 4.1 percent in August, which takes third place as the lowest in the country (behind North Dakota, with 2.8 percent and a three-way tie between Nebraska, South Dakota and Utah with 3.6 percent).
A more inclusive estimate of unemployment, which accounts for discouraged and underemployed workers, measures 9.3 percent. That’s significantly higher than the official rate, but still low nationwide.
Critics say it’s not job creation, but a shrunken workforce (as noted above) that’s driven the rate down.
In the same debate, Shumlin said his administration has overseen the creation of 9,200 new jobs. Shumlin makes a similar claim on his campaign website.
Hard to say.
The Quarterly Census of Employment and Wages is the most reliable source of jobs numbers, capturing an estimated 98 percent of all American jobs, according to the Bureau of Labor Statistics, which prepares the numbers. But no data are perfect barometers. Direct comparisons are not always appropriate, due to coding changes at the federal level. Nonetheless, here’s what the numbers show.
In 2011, statewide annual employment averaged 295,512, compared to 301,586 in 2013. (Data for 2014 is not available.) That’s 6,074 more jobs in 2013, on average, than there were in 2011. Which is about one-third less than Shumlin took credit for in the debate.
Year-over-year changes often are recommended as most appropriate for comparison of economic changes, but the employment census numbers are available quarterly, too.
In the first quarter of 2011, statewide employment averaged 291,846, compared to 301,227 in the first quarter of 2014 (more recent data are not yet available). That’s creation of 9,381 jobs, which slightly eclipses the figure Shumlin cited.
Shumlin’s campaign manager, Scott Coriell, said they used a different data source to come up with their job creation number. The Current Employment Statistics Program is produced by the Vermont Department of Labor in conjunction with the U.S. Bureau of Labor Statistics. Similar to monthly unemployment rate estimates, the CES exchanges accuracy for timeliness.
According to the survey data, employers reported about 8,300 more jobs in August 2014 than they did in December 2010, before Shumlin took office. But given Vermont’s extreme seasonality, that’s a bit like comparing summer squash to ski poles.
In a later debate on Vermont Public Radio, Shumlin said Vermont “has more jobs than at peak in 2005 …”
Half true.
According to the Quarterly Census of Employment and Wages, it’s true that Vermont averaged more jobs in 2013 than it did in 2005: 301,477 compared to 300,919 — an increase of 558 jobs.
However, 2005 did not mark Vermont’s recent employment peak — whether you look at the quarterly employment census or the monthly jobs estimate. The distinction of “peak” belongs to 2007, when the Quarterly Census of Employment and Wages shows an annual average of 303,448 jobs in Vermont. Compared to 2013, that’s a loss of 1,971.
