• Block grants and the “flypaper” effect

    I may not be dug out from the snow, but I am dug out from grant reviews. So, I’m returning to considerations of substance use treatment and prevention. My focus, at the moment, is trying to understand the public funding landscape for such programs. The 2003 paper by Cowell, McCarty, and Woodward, “Impact of Federal Substance Abuse Block Grants on State Substance Abuse Spending: Literature and Data Review” (pdf), is a great way to get started.

    It examines several studies pertaining to federal block grants to states for substance use treatment and prevention.* An important question considered by the studies reviewed is the existence and size of the “flypaper” effect. What’s that?

    Put simply, the flypaper effect occurs when grants “stick where they hit.” If states use federal funds designed to address substance use for that purpose, then the funds have stuck to their target, like a fly to flypaper. If, on the other hand, states find ways to redirect the funding to other activities then the flypaper is not very sticky.

    Federal-to-state substance use prevention and treatment block grants (SUPTBGs) are a big deal, accounting for about 40 percent of all public financing for such programs.

    The first question one should consider is, what flexibility do states really have in directing substance use prevention and treatment block grant (SUPTBG) funds? The authors explain,

    First, although states are required to report their use of funds, the agency responsible for administering the block grant, SAMHSA [Substance Abuse and Mental Health Services Administration], is not authorized to enforce the conditions or prosecute violations of the conditions. Rather, SAMHSA’s mandate is simply to monitor the use of block grant funds. Second, […] monitoring of the funds is imperfect because states are able to disguise actual state spending in ways that would be difficult to detect by SAMHSA’s monitoring efforts. To demonstrate that they meet maintenance-ofeffort requirements, for example, some states may count Medicaid-funded services as state-funded services. Given the lack of enforcement of conditions and the imperfect monitoring, states likely understand that block grant funds are fungible.

    That’s good context and speculation. But what do studies of actual data reveal about the flypaper effect? Do the funds stick to their target or not? If so, to what extent? Four studies are summarized and reviewed by the authors. Below I list the studies, provide excerpts of their abstracts, and include a few passages from the Cowell, McCarty, and Woodward paper as commentary.

    [1] Federal Block Grants and State Spending: The Alcohol, Drug Abuse, and Mental Health Block Grant and State Agency Behavior, by Jacobsen and McGuire:

    A central question about the block grant form of intergovernmental aid is whether states spend the funds on the intended services or use budgetary strategies to appear to be in compliance with maintenance-of-effort provisions but then reallocate block grant funds from the targeted program. We studied the effect of the Alcohol, Drug Abuse, and Mental Health block grant program on state substance abuse expenditures by analyzing spending data from the fifty states between fiscal years 1987 and 1992. Our findings suggest that this block grant has stimulated state spending, but this effect may be relevant only since 1990 and, in fact, differs among states.

    Cowell, McCarty, and Woodward paper commentary:

    Jacobsen and McGuire found that states generally spent block grant dollars as Congress intended. [Other commentary on this paper applies to the following one as well.]

    [2] Monitoring and Enforcement in Federal Alcohol and Drug Abuse Block Grants, by Ma, McGuire, and Weng:

    Our results are broadly consistent with earlier findings. State [spending] did respond to block grants, and this magnitude increased after 1989. Our results, however, indicate that technical reviews and waiver applications do not appear to explain states’ responses. In other words, these specific enforcement mechanisms, after the step-up in enforcement has been controlled for, do not appear to lead to increased spending. There are two possible interpretations of these results. First, the effectiveness of general enforcement does not seem to be enhanced by the particular mechanisms under investigation. Second, states’ compliance may depend more on a continuing relationship with the federal government or reputation rather than ex post auditing or inspection.

    Cowell, McCarty, and Woodward paper commentary:

    Ma et al. supported Jacobsen and McGuire’s findings. […]

    Neither Jacobsen and McGuire nor Ma et al. addressed a potential aspect of mis-specification in the equations. Both studies assumed that if the estimated effect of the block grant on state spending was greater than the estimated effect of personal income on state spending then this was evidence of the flypaper effect. It was then reasoned that the difference between the block grant effect and the personal income effect is the magnitude of the flypaper effect. However, Fisher shows this reasoning is incorrect. Deriving the empirical flypaper effect from theory, Fisher shows that the empirical specification should also include the per capita share of taxes. […] Given Fisher’s critique, it is debatable whether Jacobsen and McGuire and Ma et al. actually found a flypaper effect. Therefore it is unclear that there is a positive association between the substance abuse block grant and state substance abuse spending.

    Other potential sources of specification bias should be considered and addressed when examining the influence of block grants on state spending. These include: accounting for the influence of neighboring states or states with similar characteristics; properly specifying the production function or cost function of public goods; and specifying the functional form of the model. [Citations omitted.]

    [3] The Impact of Federal Alcohol and Drug Abuse Block Grants on State and Local Government Substance Abuse Program Expenditures: The Role of Federal Oversight, by Gamkhar and Sim:

    Are the federal Alcohol and Drug Abuse (ADA) block grant funds substituting for or supplementing state and local government spending on substance abuse? Using panel data on state and local government substance abuse programs, this study explores the fiscal effects of the ADA block grant money and the increased enforcement (after 1989) of federal restrictions on state spending of ADA block grants. The findings here reveal that for the current period, the federal ADA grant has no statistically significant effect on state and local government substance abuse spending both before and after 1989, and the increased enforcement of federal restrictions on the ADA grants after 1989 does not change this result. An additional finding is that lagged ADA grants have had a large effect on substance abuse spending both before and after 1989—a feature of the program not considered in previous studies.

    Cowell, McCarty, and Woodward paper commentary:

    Gamkhar and Sim noted that the SAPTBG funds allocated in a particular fiscal year are typically available for use by states for two fiscal years, so states have some flexibility when they use their block grant funds. Failing to include these lags could give a biased estimated of the effect of the substance abuse block grant. […] In contrast to the findings of Jacobsen and McGuire and Ma et al., Gamkhar and Sim found that current period substance abuse block grant funding has no statistically significant effect on state or local substance abuse spending either before or after 1989. Current period block grant funding, however, had a positive effect on state expenditure in the following period.

    [4] National and state spending on specialty alcoholism treatment: 1979 and 1989, by Huber, Pope, and Dayhoff:

    The National Drug and Alcohol Treatment Unit Survey was used to measure changes in specialty alcoholism treatment spending between 1979 and 1989 nationally and by state. National spending more than doubled from $1.6 billion to $3.8 billion in 1989 dollars. Private spending increased more rapidly than public spending, although most clients continue to be publicly funded. Dramatic differences across states in public funding growth were partially explained by differential increases in per capita income and in federal substance abuse block grants. Access to treatment continues to vary widely across the states.

    Cowell, McCarty, and Woodward paper commentary:

    Although the fourth study that examined the impact of the SAPTBG does not address flypaper effects directly, it found that state substance abuse spending is positively associated with substance abuse block grants. Huber et al. […] found for every $1.00 increase in state SAPTBG funding, the state administered spending on alcoholism treatment increased by $0.80.

    Clearly the evidence is mixed on the size of the flypaper effect for SAPTBG funds, or such was the case at the time Cowell, McCarty, and Woodward wrote their paper (2003). I’m still reviewing the literature in this area and may find more recent work pertaining to this question.

    * Cowell, McCarty, and Woodward also comment on certain aspects of available data for studies of the type reviewed and make suggestions for future research in this area. I will not cover those parts of their paper in this post.

    Comments closed