The Maine Department of Labor’s chief economist says his department often clashes with federal labor officials over estimates of job growth in the state, including figures for 2014 that were cut by more than half in a revision this year.

The issue arose this week when the federal Bureau of Economic Analysis released its gross domestic product report for 2014 and said Maine’s output of goods and services barely moved in 2014, increasing just 0.2 percent from $54.6 billion to $55.8 billion. The bureau also cut the figure for Maine’s 2013 GDP from a preliminary growth estimate of 0.9 percent to a revised 0.5 percent.

Economists and state officials were surprised by the low numbers, which reflected slow growth in 2013 and a virtually flat economy in 2014.

One of the reasons the low growth figure caught people off guard is that another federal agency, the Bureau of Labor Statistics, had forecast last year that Maine would add 5,100 jobs during the year, a figure that, if it had been realized, would have likely led to more robust economic figures.

The Bureau of Labor Statistics revised its jobs figures in March and said the Maine economy created just 1,900 jobs during the year, a sharp decrease that helped explain why Maine’s economy was so lackluster.

Glenn Mills, the Maine Department of Labor’s top economist, said his agency often disagrees with Bureau of Labor Statistics economists over job forecasts and urges changes that are not always adopted.

“They don’t do a great job with the estimates and we are aware when they’ve gotten too high or too low,” Mills said.

The Bureau of Labor Statistics relies heavily on its survey of 2,000 Maine employers (out of about 47,000 total) to get its estimates, Mills said. But that survey is often swayed by seasonal factors, he said, along with a weighting formula that the federal agency applies to businesses it surveys. That can often result in a small number of employers – or even one employer – affecting the preliminary estimates by dozens or hundreds of jobs based on a short-term increase or decrease in hiring, he said.

For instance, it could survey a hotel in May that would reflect dozens of new hires as it prepares for the summer tourist season – jobs that don’t exist come the next March. Or it could survey a manufacturer that hires workers to staff a new production line, but then the contract for the product falls through and the workers are laid off.

Mills said Maine’s labor department offers input to the Bureau of Labor Statistics, but the federal economists rely heavily on their formula for job estimates and are reluctant to change their numbers. Maine labor economists are largely responsible for revisions, he said, which they make when actual job figures are reported in state corporate filings.

“We were aware that their (2014 job) estimates were way too high,” Mills said, but federal officials largely stuck with their formula, resulting in the estimate of 5,100 new jobs for the state in 2014. When the actual numbers came in, the jobs that were added were less than half of the estimates.

Mills said any kind of economic projections are speculative and can be greatly affected by unanticipated events, such as a big layoff or sudden surge in sales. But he said state officials know the Maine job market best, adding the Bureau of Labor Statistics would be well-served by taking more state input. That, he said, could result in better estimates and less drastic revisions.