Are ARPA funds for states and localities being spent successfully?

December 14, 2021

By Amanda Kass, Associate Director at the Government Finance Research Center and Philip Rocco, Associate Professor of Political Science at Marquette University

The American Rescue Plan Act (ARPA) was a historically important response to a massive economic crisis induced by the COVID-19 pandemic. One of the most significant pieces of the legislation allocated $350 billion in aid to state and local governments U.S. Territories, Tribal Governments, and D.C. to cover a combination of plummeting revenues and staggering service demands. But given the unprecedented nature of the program––and the unprecedented problems it was designed to respond to––how will we know if the program has been a success?

This question is harder to answer than it seems. Experts in program evaluation argue that this effort benefits dramatically when there is a relatively clear, limited set of policy goals. Yet the Coronavirus State and Local Fiscal Recovery Funds (SLFRF) address a wide variety of goals, including some that are in conflict with one another.

The program’s flexibility gives it a family resemblance to the long-defunct system of General Revenue Sharing program, in which the federal government transferred funds to state and local governments which could be used for any purpose at all. Yet SLFRF are not fully discretionary--meaning recipient governments do not have complete autonomy on how to use the aid.

Based on the guidance provided for the program by the Treasury, SLFRF aid can be used for four broad categories. First, governments can use the funds to respond to the economic and health effects of COVID-19. Second, they can replace revenues lost as a consequence of the pandemic. Third, they can provide premium pay for essential workers. And finally, governments can use funds to support water, sewer, and broadband infrastructure projects. Governments choose how to allocate the SLFRF aid among these broad categories and decide the specific projects to fund within these categories. In determining how to use the money, and in-line with President Biden’s larger “Build Back Better” agenda, the Treasury Department encourages governments to use the funds in ways that will “promote strong, equitable growth, including racial equity.”

Because the program is designed to be tailored to state and local needs, data collected for evaluations cannot fit in the one-size-fits-all category. The information governments must provide to the Treasury Department varies based on jurisdiction size and what they use the money for. States and very large cities and counties have to submit three different types of annual and quarterly reports, while non-entitlement units (generally cities and counties with populations less than 50,000) only have to submit an annual project and expenditure report. The Recovery Plan Performance Reports, which states, and very large cities and counties must submit, are more than just accounting documents that capture line-item spending. In these reports, governments must explain the targeted outcomes of their spending programs and how those outcomes are going to be “achieved in an effective, efficient, and equitable manner.”

The revenue replacement category provides governments the greatest latitude and as a result, the success of this spending may be the most difficult to evaluate. Under the regulations, governments calculate their COVID related revenue losses using a uniform formula. They then spend the money on general government services, and, unlike the other categories, those services do not have to be COVID related.

In essence, the revenue replacement category allows governments to commingle SLFRF aid with other revenue sources and treat it as general revenue. But doing so also means the SLFRF aid cannot be neatly tied to specific spending items or programs, the success of which are more amenable to evaluation. This contrasts with the other categories in which states and large cities and counties have to report to the Treasury Department how they spent the money at the project level as well as the programmatic outcomes of that spending. Since the revenue replacement category is less administratively burdensome than other spending, governments may be incentivized to use this category to the maximum extent possible.

In addition to ways in which the program’s design and regulations may shape spending, how the SLFRF plays out is also subject to the decisions by state and local officials and what they see as the pressing needs of their communities and political priorities. There’s also sharp partisan politics surrounding the program with some elected officials questioning the very premise of its necessity. Further, in some instances, elected and appointed officials have chosen to use the aid for purposes that are explicitly prohibited by ARPA. Thirteen Republican Attorneys General, for example, won a lawsuit to strike down a provision that prohibited using the APRA aid to pay for tax cuts.

Returning to our original question: how will we know if the SLFRF was a success or failure? Evaluating the program in a holistic way is going to be challenging along several dimensions: 1) lack of singular policy aim from architects and proponents; 2) national partisan politics over it; and 3) tension between what’s politically useful (measurable outcomes; allocating aid to specific projects) to what’s administratively least burdensome.