Yesterday I wrote about the economic drag caused by squeezed state and local budgets, posting a figure that showed the loss to overall GDP caused by state/local contractions. Today, I’ve added state and local job losses to that picture.
The figure below just plots the same data as in the last figure—the annual percentage point contribution or subtraction to real GDP growth from the state and local sectors. But it adds annual gains or losses in jobs.
Sources: BEA, BLS
As you can see, it’s a very tight fit. Last year, state and local squeeze shaved about 0.3% off of GDP and cost 266,000 jobs. A simple regression of state/local job losses on the GDP contribution finds that for every point of growth that the states and locals take off of GDP, employment in the sectors falls around 700,000.
We generally recognize that GDP losses map onto job losses but the fit is not usually this tight—there are lags in the generalized relationship between growth and jobs and lots of other moving parts. But that’s less the case in state and local governments. Here, the chain of events is pretty obvious and pretty clear. You squeeze their budgets, it shows up quickly and directly in growth and jobs.
Conversely, and here’s the policy part, were we to use federal stimulus to help relieve their budgets, we could get this relationship running in a better direction.
Update: The always righteous Larry Mishel (president of the Economic Policy Institute) points out that the job losses I’m citing above are only part of the story. States and cities buy private services and contract with private firms. Ethan Pollack writes: “For each dollar of budget cuts, over half of the jobs and economic activity lost are likely to be in the private sector.”
Though this is a federal case, Larry’s point reminds me of when the Federal Aviation Administration closed for two weeks due to Congressional gridlock on renewing their budget. About 4,000 federal workers were furloughed but about 70,000 private sector workers on FAA contracts were laid off.
[See also CBPP’s analysis of this problem.)