December’s ugly employment numbers, which painted a picture of a region that’s shedding jobs and beset with the highest unemployment levels since the mid-1980s, make it look like the sky is falling on the Buffalo Niagara labor market. But local economists say they believe the picture is much brighter than what statistics from the state Labor Department show.

They think the local job market is stagnant – at worst – and possibly growing slowly. And they think unemployment, while still stubbornly high, isn’t quite as bad as the more-than-two-decade high of 8.6 percent that was reported for December.

The reason is a glitch in the data that the Labor Department uses to compile the job and unemployment numbers, which economists said shows the region has suffered through the loss of 4,000 jobs – a 14 percent plunge – in a job category that includes temporary jobs available through employment agencies.

“I don’t think the numbers are as bad as we’re seeing,” said Gary Keith, the chief economist at M&T Bank. “I can’t imagine that 4,000 jobs just vanished like that, and we also don’t feel it in things like retail sales and home sales.”

John Slenker, the Labor Department’s regional economist, also agrees that something seems amiss in the job data, and he expects revised figures that are scheduled to be released in early March to show that the local employment market is in significantly better shape than the bleak December numbers showed.

Those December numbers don’t paint a pretty picture. The local unemployment rate jumped to a three-year high of 8.6 percent last month, the Labor Department said Tuesday. And last week, the department reported that the region has lost 2,100 jobs over the past year – the region’s fifth straight month of year-over-year job losses at a time when the local economy should be adding jobs with the recovery from the Great Recession now in its third year.

So why are economists, like Slenker and Keith, skeptical of the numbers?

A big part of it is how the figures are compiled in the first place.

The Labor Department bases its monthly jobs report on a survey that covers about 3 percent of all local establishments.

The department’s economists then analyze the information and make projections based on broader – but older – Census data that serves as the benchmark for the preliminary employment estimates that it releases each month.

This year, Keith speculates that one, or possibly more, temporary help agencies that are part of the monthly survey are reporting a significant drop in employment. The reason could be as simple as an agency that’s part of the monthly job survey lost a big contract to provide workers to another temporary help firm that isn’t part of the survey.

And when that drop is extrapolated over the entire job market, it results in a big decline in employment – roughly one of every seven jobs that had been reported in the statistical category that includes temporary help jobs.

The passage of time puts the figures on ever shakier ground. That’s because the Census numbers that the monthly employment report is benchmarked against are from last spring, and their relevance to the current job market wanes with each passing month, Keith said.

“They are really at the far end of their ability to project outward with their small samples,” Keith said. “There’s just chaos in these numbers.”

Beyond that, new federal rules that took effect about two years ago to prevent the improper manipulation of employment data have limited the ability of local labor market economists, like Slenker, to make adjustments to regional survey data when it doesn’t appear to jibe with what actually is happening in a local market.

For instance, Slenker notes that the job losses over the past five months haven’t been accompanied by a spike in first-time applications for unemployment benefits.

Keith said he hasn’t seen the decline in retail sales or home sales that would be expected to result from the loss of more than 2,000 jobs across the region.

“We don’t see any of the other things that would cause us to be worried,” Slenker said.

The reported decline in temporary jobs runs contrary to the findings of a quarterly survey of employers conducted locally by the Manpower temporary help agency. That survey found that the hiring plans of those local employers for the first quarter of this year have strengthened, although they are not nearly as robust as they are nationally.

“Hiring activity is expected to increase,” said Kelly M. Scott, a Manpower spokeswoman. “Employers expect improved employment prospects compared with one year ago.”

There are other signs of hidden strength in the local job market. The Census data, which form the basis for the Labor Department’s projections, already have been updated through the end of June, and those numbers show that the Buffalo Niagara job market at the end of the second quarter was about 600 jobs stronger than the initial Labor Department figures showed.

Private sector employment, which excludes government jobs, was about 1,800 jobs better than the Labor Department figures showed, and the category that includes temp jobs was up by 150 positions according to the Census data at the end of June, rather than being down by nearly 4,000 jobs in the state figures.

Slenker sees those figures as a sign that the revised job data due out in March – which is based on those same Census figures – will show a stronger Buffalo Niagara job market than the current reports are showing.

Just how much stronger, however, is open to debate.

At worst, Slenker thinks the revised figures could show a stagnant local job market, although he believes there is a good chance the upcoming data will show modest job growth.

“I know that we have more jobs in June than we originally stated because the revisions are based on the [Census] data,” he said. “We could even be up slightly.”

Keith also expects improvement in the revised data. “It’s hard to predict how much better it will be,” he said.

“It’s frustrating,” Keith said, noting that the latest bleak job numbers – because they indicate a significant weakening in the local economy – could cause local businesses to delay investments or expansion efforts, even though the actual state of the region’s job market may be far less dire.

“They may base decisions off that, and if it’s a statistical quirk, we’re doing a disservice to ourselves,” he said.