No one knows how much time and money is being diverted from taxpayers’ pockets into the coffers of public employee unions through a practice called release time. This allows government employees who are also union officials to be released from their regular jobs to perform union work, while still being paid full salary and benefits by their government employers.
Release time is a widespread practice among states, cities and school districts. How widespread is not known. There has been almost no research into the topic of union release time at the state and local level.
To get some answers, the Goldwater Institute sent public records requests to the state departments of corrections, the capital cities and the largest school districts in every state. The requests sought data on the total hours and cost of government-paid union release time that each jurisdiction allowed in the last full fiscal year. Also requested was documentation on individuals who had used more than 500 hours of union release time, as well as copies of collective bargaining agreements where release time provisions are normally spelled out.
The Goldwater Institute documented more than 400,000 hours of government-paid union release time in its survey of 150 jurisdictions. That is a conservative estimate based on actual disclosures and analyses of contracts, and likely undercounts the total hours allowed.
Only 82 of the jurisdictions were able to provide data on the hours they allowed, including 44 that reported allowing zero hours. The rest either could not or would not provide data.
If an entity reported actual data on the hours and cost of release time, that was taken at face value as an accurate figure. There were a couple of exceptions in which information reported by government officials was wildly inconsistent with the provisions of the union contracts. Columbus City Schools, for instance, insisted union officials do not receive district-paid release time, yet its contracts allow unions more than 3,000 paid hours annually. In that case, the hours allowed in the contract were counted.
Where governments did not provide actual data, the Goldwater Institute estimated permitted release time by analyzing that jurisdiction’s union contracts. This only identifies defined hours, such as government-donated time to employee hour banks the unions can use, or full-time union release positions which are counted as 2,080 hours, the standard work year. Most contracts also have undefined provisions that cannot be estimated, such as allowing union stewards unlimited amounts of time to represent employees in grievance and disciplinary matters.
Among the jurisdictions that were unable to provide data, most acknowledged they do not track union release time. Some claimed they could, or at least might, be able to provide data, but that it would take many hours and cost hundreds if not thousands of dollars to research.
The Montana Department of Corrections, for example, said it could produce the data, maybe, but only if the Goldwater Institute agreed to pay $1,500 up front for the hours of research it would take. The Fairfax County, Virginia, school district responded that it would take 12 hours to do the research.
For the most part, the Goldwater Institute declined to pay research fees, and counted those jurisdictions as not tracking release time. While cost was a factor in a few instances, such as the Montana corrections department, the main reason was that an important part of the survey was to determine whether agencies bothered to track union release time in any meaningful way, whether for the benefit of the public or policy makers; or simply because of good fiscal management practices.
For agencies that were able to produce complete data, finding it usually involved a simple computer search of their payroll databases, where union release time is tracked with a unique time code. If an agency required many hours of research, costing thousands of dollars, then that was taken to mean it does not routinely track union release time and has no easy way to do so.
The Goldwater Institute did agree to pay standard production and copying fees in the states where those are allowed by public records laws before an agency is obligated to process a request.
Several jurisdictions invoked state residency requirements that allowed them to reject records requests filed by non-residents. In those instances, the Goldwater Institute enlisted the aid of other nonprofit public policy organizations in those states to file the requests on its behalf.
Since the initial public records requests were filed in January, all jurisdictions had ample time to deliver the data, even with the months-long government shutdowns because of the Corona virus that followed. Many agencies that meticulously track the hours and costs of release time were able to produce the data within weeks. Others, such as the New York Department of Corrections, are still sending letters saying they will need more time.
The records request specifically sought government-paid union release time. Unpaid release time, time that is reimbursed by the unions, or time that was donated by employees was specifically not sought, except to the extent the government paid benefits.
