The facts haven’t changed; so far we haven’t found anything seriously wrong with the data that we’re using except a little that came in after the October 10 deadline. What is changing are the methods used to turn raw data into “jobs”. We are now reporting 1594 direct jobs.

 

A number that we do understand is the number of weeks of work paid for directly by ARRA funds which flowed through the State government in programs which are reportable under Section 1512 of the American Recovery and Reinvestment Act (how’s that for a mouthful?). Approximately 11,400 weeks were worked in Vermont by persons supported by ARRA resources. This is not something we report to the feds but we do want to report it to you.

 

This is the period when the federal agencies like the Federal Highway Administration and the Department of Education look at the reports we filed with the Office of Management and Budget (OMB) and see whether they have any problem with them. Most of their problems are minor things – whether or not there should be dashes in certain fields and how much explanation there is in text fields – we make these changes and move on, although they do take time we’d like to be spending on other things. OMB had some general guidelines but agencies were allowed to modify them for “their” programs; so, although we report the same data for every program, we report and calculate it differently depending on the program.

 

 

How to calculate “jobs created or retained” is one important difference in guidance between agencies – and in our understanding of the guidance when we did the calculations and the various agencies interpretation of the guidance now. As explained here, we are only calculating “direct” jobs – people employed by the state and its prime recipients as a result of stimulus funding. But even this isn’t simple.

 

The job count is of “full time equivalents” (FTEs, in our jargon). For a forty hour a week job, two people each working twenty hours are one FTE; one person working sixty hours is 1.5 FTEs. That’s the easy part; it gets complicated when you ask how many weeks are we talking about.

 

The October 30 report is for the period beginning in February when the President signed the American Recovery and Reinvestment Act (ARRA) into law and ending on September 30. The next report will be for the quarter which began October 1. If you know that there were 2000 hours of work paid for by a particular award, do you divide this by the 1280 working hours since ARRA was signed and decide that 1.56 jobs were created? This is the average jobs attributable to the award since the beginning of ARRA. Or do you say the award was in the third quarter so we’ll divide by 520 hours in the quarter and report that there were actually 3.85 jobs – the average for the quarter. What if work only started in the last week of the quarter? Do you then divide by 40 and report 50 jobs? This number is likely to be close to the number of people who actually got employed.

 

Strangely, none of these answers ranging from 1.56 to 50 are wrong; they’re all based on the same data. One is the average number of jobs attributable to the award since ARRA began; the second is the average number of jobs in the quarter; the third is the average number of people working on the award since work on it began. A mathematician would argue that only numbers covering the same period can properly be added together, even if all the numbers are otherwise correct. But mathematicians didn’t make the rules we use to report. Most agencies want jobs reported as hours worked divided by working hours since the award date. Others want hours since the beginning of ARRA. Most of the time, we went with the award date or project start date when there were separate projects in an award in an attempt to be consistent across awards. So we have to recalculate in some cases where the federal agency wants to do it differently.

 

This is the first time through this kind of reporting both for the Federal agencies and for us. They’ve done a good job with automated tools. State agencies have done a great job responding to us in the Office of Economic Stimulus and Recovery even as the rules change. Hopefully OMB will be able to enforce uniformity on the agencies going forward so that the numbers will have more meaning.

 

Vermont and other states have suggested to the OMB and agencies that, in the future, they report “weeks worked” as we are doing in Vermont rather than a mathematically-challenged and hard to understand concept of “jobs”. We will continue to report this number to you going forward.

 

Tom Evslin