Two labor market surveys for Connecticut pointed in dramatically different directions for July, leaving labor market experts puzzled about which way employment rates are trending in the state.
The state’s business establishment survey, which estimates payroll jobs in businesses that are covered by unemployment insurance, showed that Connecticut added 5,100 positions in July, “a decent performance for recent times.”
At the same time, the state’s household survey—a measurement of household employment, unemployment, and the state’s unemployment rate—showed a sharp and dramatic decrease in employment, an increase in unemployment and a discouraging increase in the unemployment rate.
Based on the household survey, the estimate of unemployed people, seasonally adjusted, was 5.6 percent higher from June 2012 to in July, and the unemployment rate climbed higher four-tenths of a percentage point to 8.5 percent, back to the same level measured for October 2011.
The July 2012 U.S unemployment rate was 8.3 percent, up one-tenth of a percentage point from the previous month.
If the household survey numbers are true, Connecticut is now back above the national unemployment rate.
"Both can't be true"
“They both can’t be true,” said Andy Condon, the director of the Office of Research at the Connecticut Department of Labor, of the establishment survey and the household survey.
“These programs have been vital to our understanding of economic conditions in Connecticut, but we will have to wait until more data comes in to see if July’s results were a statistical anomaly or an early indicator of a turning point in the economy, as yet uncorroborated by other data.”
Condon, on a teleconference call with reporters Thursday morning, said he has not ever seen a discrepancy this sharp between the numbers. He said explaining the discrepancy is difficult, if not impossible.
“There are two possibilities. One is that these are different surveys and different populations measuring the same thing. Since they are both sample-based surveys, it’s possible that one or the other is anomalous this month. That isn’t very likely, but it’s possible,” he said.
“The second set of possibilities is that you can’t discount the fact that the household survey is giving us an early indicator of unemployment that we’re not seeing elsewhere. Only time will tell us whether that is true.”
Locally, unemployment numbers for Madison ticked up in June to 6.0 percent, from 5.7 percent in May and 4.7 percent in April. That is the highest it has been this year for the town, but lower than June 2011, when it was 6.2 percent.
In June 2012 there were 593 people listed as being unemployed in town. Those numbers, provided by the state Department of Labor, are not seasonally adjusted and are the latest available for the town. See this list from the Connecticut Department of Labor for June numbers for other towns.
"They typically move in similar directions"
Condon said one survey is not considered more reliable than the other. “They typically move in similar directions, this is unusual,” he said. He said he did not think other states are seeing similar discrepancies, but that in some cases in the past, numbers have had to be revised in subsequent months.
The business establishment survey, which showed that Connecticut added 5,100 positions in July, showed those gains across the board in a wide variety of jobs, Condon said, including manufacturing, business services, and health care.
Gov. Dannel P. Malloy, in a prepared statement, said he was “skeptical” about the numbers.
"A gap of this magnitude has never happened"
“I am skeptical about the jobs numbers released today ... A gap of this magnitude between these two surveys has never happened in the 22 years they’ve been conducted.”
“To buy into the household survey number you’d have to believe that Connecticut lost 503 jobs every day during the month of July, and there’s just no evidence to suggest that happened. Unemployment claims have drifted upward, but not at a rate that justifies the household survey numbers. Hence the skepticism.”
Malloy said it is clear that people do come flooding back into the job market once a recovery is underway, since they are more optimistic they can find work, causing the unemployment rate to go up. Also, he said, it takes more than 20 months to turn around a state economy that “failed, on a net basis, to grow jobs for more than 20 years.”
State's recovery affected by external factors
Finally, he noted, the state’s recovery is affected by external factors, including the struggling national recovery and uncertainty in Europe.
“The phrase ‘worst economic downturn since the Great Depression’ is used so often, we’ve become immune to the words. But they’re true. So, it should come as a surprise to no one that pulling the country and our state out of that downturn is hard. Really hard,” he said.
“But let me be clear: we are making progress, and I have every confidence we’ll continue to make progress. Just yesterday, we announced an agreement that will retain more than 1,200 high paying jobs with good benefits and create up to 1,000 more in our state. It’s also important to remember that the private sector in Connecticut has created more than 23,000 jobs since January of 2011, according to the employer survey.”
The details as reported by the state Department of Labor are in a press release posted with this story.
Looking for a job, check out these resources from the state Department of Labor.