What the heck is the MVPF?
Why Oregonians, especially elected officials, should know about this new way to allocate taxpayer dollars.
In an era of data, analytics, and statistical regression, politicians cannot get by with hunches and gut feelings. We can and should expect that our elected officials are consulting new research on what makes good policy. In particular, mayors, county commissioners, and state legislators should be reviewing the Marginal Value of Public Funds, or MVPF.
According to Opportunity Insights, the MVPF measures long-run policy effectiveness. Unlike a traditional cost-benefit analysis, the MVPF uses a longer time horizon to assess the effectiveness of a program. An early childhood education program had a MVPF of 12, which means that the enrolled children received nearly $12 in benefits for each $1 the of long-run cost to the government.
Oregonians would benefit from the MVPF becoming a core part of our policymaking. Unlike a traditional cost-benefit analysis, the MVPF allows policymakers to consider how the benefits of a program may produce societal returns that result in the government saving money or increasing its revenue.
For instance, some policies in the “Policy Impacts Library” have an infinite MVPF because the increase in income generated by additional schooling results in more tax revenue for the government—so much additional revenue that the program costs are entirely covered. That library, compiled by Opportunity Insights, should be bookmarked by every policymaker in the state.
While politicians like to think that they are policy innovators, the real innovation would be to lean into this metric as a guide for how to maximize the returns on public dollars. Relying more on the MVPF would uncover policy solutions that might otherwise sit on the backburner. Community college tuition changes in Texas, for instance, generated a MVPF of 349.51—in other words, $349.51 in benefits to students for every $1 in program costs. Medicaid expansions to young children recorded an infinite MVPF, so did helping low-income individuals move to areas with higher rates of social mobility.
Use of the MVPF will also help policymakers avoid “solutions” that tend not to have much bang for the buck. A housing voucher program in Chicago offers a case study in what not to do, only returning $0.65 in benefits for every dollar in long-term program costs.
Of course, not all benefits can be measured and some costs are always worth incurring. Society, for instance, should make sure that public buildings are accessible and designed for all, regardless of the MVPF on that project. But the days of policymakers leaning only on their guts or party platform to support or oppose a policy should be over.
Opportunity Insights is working on adding more and more policies to its library. Not all of those policies will be especially relevant to Oregon. For instance, the MVPF of a policy in New York City may not mean much for evaluating even an identical policy meant for Roseburg or Burns. Still, there should be some evidence that policymakers in Oregon have consulted similar policies in the library to make sure they’re headed in the right direction.
The stakes are too high for our politics to be dictated by party platforms. Evidence-based policymaking may not excite your political juices, but we need to stop treating politics like some weird version of fantasy sports. Who “wins” and “loses” in the political game does not matter when lives are at stake and our quality of life is in question.
Kevin Frazier edits The Oregon Way. His “day job” is attending school at the UC Berkeley School of Law and Harvard Kennedy School. A born-and-raised Oregonian, he looks forward to finding his way home after graduation.
"Silence, Texting, and the Pursuit of Knowledge" by hjl is licensed under CC BY-NC 2.0
Kevin, isn't this just a more refined method for computing the benefit side of a cost-benefit analysis? A good thing for sure, but not necessarily that much different from what most policy shops do now. Plus, I'd like to see some caution with the projection of results from current populations based on prior results. Better to see a range of outcomes, low to high, allowing for the vagaries of human experience. I think that would add some credibility to the projections. When one sees returns quantified down to the penny, it occasions some skepticism. A little predictive humility would be more convincing.