Recent jobs data have stymied analysts and labor officials, with the state”™s unemployment rate rising to 9.1 percent in August from 8.3 percent a year earlier as the U.S. unemployment rate declined a full percentage point over the same period.
As economists and bureaucrats argue over the numbers, businesses will likely be forced to wait until March 2013 for more clarity, at which point the state Department of Labor will release revisions to its payroll and employment data for the previous 18 to 24 months.
One economist predicted the statewide unemployment rate would be revised down a full percentage point in March, but state officials said they have yet to make any projections.
Two primary measures of employment, both conducted by the U.S. Bureau of Labor Statistics (BLS) on behalf of the U.S. Department of Labor with the data then provided to each state”™s labor officials, portray a contradictory picture of the jobs recovery in New York state.
A monthly survey of private businesses that forms the basis for payroll employment estimates shows the state added 278,000 jobs between July 2010 and July 2012, including 24,800 in the three-county region comprising Westchester, Putnam and Rockland.
However, a survey of individual households, which is used in the calculation of unemployment rate estimates, suggests the state has lost a net 36,900 jobs over that same period, and that the three-county region including Westchester has added just 9,600 jobs.
In fact, between July 2010 and July 2012, no state had a larger unemployment rate increase than New York.
The difference is “really very stark, and you don”™t see that in any other state,” said James A. Parrott, deputy director and chief economist for the Fiscal Policy Institute, a nonpartisan think tank with locations in Albany and New York City.
“New York state has actually had employment growth over the last two years pretty much in line with the national average, and yet our unemployment rate has been going up while the national rate has been going down,” he said.
Parrott rejected the notion that the disparities are the result of a small sample size for the household and payroll surveys, arguing instead that they are being caused by a systematic problem with the formula BLS, which is a division of the Census Bureau, uses to arrive at unemployment figures specific to New York state.
“Some economists have tried to explain the disparity between the two surveys in New York by pointing to the relatively small sample size in the current household survey, and it”™s true that the sample size is not as great as we would like … (but) there”™s been no change in the sample size for decades in New York,” he said. “My sense is that new population weights related to the 2010 Census are what”™s thrown off the overall estimates.”
According to the state Labor Department, the unemployment rate is based primarily on the results of a telephone survey of 3,100 New York state households, while the monthly payroll estimates are based on a survey of 18,000 statewide employers.
Parrott predicted the unemployment rate would be revised closer to 8 percent when the state issues its adjustments in early 2013.
For the time being, he said, the confusion isn”™t helping business owners.
“I think the most telling factor facing businesses is what”™s happening to the volume of their own business … that said, it certainly doesn”™t help confidence and the outlook if you”™ve got an unemployment rate that is rising the way it is, at least as it is being reported,” Parrott said.
A state official said the process of revising the previous 18 to 24 months”™ worth of employment data begins in January, with any revisions typically released in early March.
As part of the process, the initial employment and payroll estimates released by the state on a monthly basis are benchmarked against data collected from unemployment insurance tax records.
The official said that historically, there have been some large upward and large downward revisions. However, the official added that the initial estimates released by the state have proved to be increasingly accurate over the past several years.
The official said BLS is constantly fine-tuning its calculation methods, adding that it is possible changes resulting from the 2010 Census are impacting the current estimates, but saying that won”™t be known until the benchmarked data are released.
