Counting Jobs Created or Saved by the "Stimulus" Bill, Part 4:
Jobs 'created or saved' becomes jobs 'funded'
1 Feb 2010 in Regulatory Economics
Washington Post staff writer Ed O'Keefe says the "Obama administration's economic stimulus program created nearly 600,000 jobs in the final three months of 2009."
These figures are analogous to those reported three months ago and which caused significant controversy. Initial reporting was rife with errors and relied on a system that impeded error correction.
They are different, however, in ways that make them incomparable with the figures initially reported.
Recovery.Gov now has an explanation for how the numbers were calculated for the second quarterly reporting period:
HOW RECOVERY AWARD RECIPIENTS CALCULATE JOBS
Recipients of Recovery Act contracts, grants, and loans are required to report quarterly on the number of jobs paid for with Recovery funds. The method or formula for calculating jobs was simplified after the first round of recipient reporting.
How did the job estimate guidance change?
The initial guidance captured jobs for a period of time longer than a quarter (February through September). The new guidance captures jobs for a single quarter (e.g., October through December). In addition, the new guidance eliminates the distinction between a job created and a job retained. Jobs are now simply based on the number of hours worked in a quarter that were paid for by Recovery funds.
In December 2009, the Office of Management and Budget (OMB) issued the changes in guidance to align with the recommendations of the Government Accountability Office. OMB believes the new formula will help improve the overall quality of recipient reporting.
It does not matter if the hours were worked by a person who was newly hired, a person whose job was saved by the Recovery Act, or a person who is in an existing position that is now being funded by the Recovery Act.
The Administration has abandoned the effort to estimate the numbers of jobs that previously did not exist (i.e., "created") or otherwise would have been terminated (i.e., "saved"). Recipients now report the number of full-time equivalent jobs paid for by the stimulus bill. According to Recovery.Gov, the "number of hours worked" that were "funded under the Recovery Act" during October--December 2009 works out to 599,108 full-time equivalent jobs. The interactive map at the top of the home page says that during this period, "funds awarded" totaled $41 billion. Dividing the two yields $61,435 per job funded, but it is not clear that job totals for this period are the result of of funds awarded during the same period.
The Council of Economic Advisers also calculates a figure for employment effects. CEA takes gross federal stimulus spending and applies a multiplier of 1.35 or 1.44 (for quarters #3 and #4 after enactment, respectively) How the CEA model handles transfer payments to State and local governments isn't clear.
The Recovery.Gov and CEA methods differ. The CEA method assumes that October--December 2009 stimulus federal spending had multiplier effects of 35%. New OMB guidance for counting Recovery Act effects "simplifies the formula used to calculate job estimates" by removing a previous requirement to calculate cumulative totals and expanding the definition of a countable job:
- Definitions of jobs considered to be created or retained:
- A job created is a new position created and filled, or an existing unfilled position that is filled, that is funded by the Recovery Act;
- A job retained is an existing position that is now funded by the Recovery Act.
In principle, "jobs retained" is analogous to "jobs saved." However, the definition for "jobs retained" is closer to "jobs funded.," which is much broader and less interesting. "Jobs funded" includes employment effects that are strictly transient; they go away once the federal spigot is turned off.
As before, no documentation is required to support the figures reported:
- This Guidance does not establish specific requirements for documentation or other written proof to support reported estimates on jobs created or retained; however, recipients should be prepared to justify their estimates. Recipients must use reasonable judgment in determining how best to estimate the job impact of Recovery dollars, including the appropriate sources of information used to generate such estimate. Where such written evidence exists, it can be an important resource for validating the job estimates reported.
There are three practical consequences of these changes. First, figures reported under the new guidance cannot be added to those reported under the old guidance. The thing being counted has changed.
Second, the new guidance should yield larger numbers. The number of "jobs funded" is larger than the number of "jobs saved."
Third, the absence of any requirement for documentation will invite skepticism similar to what accompanied the initial round of reporting. Hardly anyone doubts that federal stimulus spending funds jobs. Whether it creates or saves jobs is the interesting question, and it was the advertised purpose of the Recovery Act.
