Weekly Economic Report: An Experiment with Education Savings Accounts

by Wyoming Liberty Group Staff

The State of Arizona has an ESA program (Empowerment Scholarship Account) that is often discussed in the national debate over education choice. What would an ESA program look like in Wyoming?

Main finding: A Wyoming version of the Arizona ESA, with universal eligibility for the K-12 student population in the state, could generate $40 million in spending on education choice. Under alternative assumptions that number could grow to $500 million.

---

In 2011 the State of Arizona opened its Empowerment Scholarship Account program. That year, 115 students participated at a cost of $1.4 million for the state. By 2017 participation had grown to 4,102 students with total expenditures reaching $37,000,000. Spending per ESA-participating student started at $12,174 in 2011 and was initially volatile with a peak at $17,219 in 2013. It then gradually dropped to $9,020 in 2017, while the participation rate as percent of eligible students increased from 0.1 percent to 1.64 percent.

On average, the number of students participating in the ESA program has doubled since 2013. One reason is probably the expansion of eligibility. Initially, 115,000 students were eligible, approximately ten percent of the elementary and secondary student population in Arizona's public schools.* In 2015 eligibility expanded to 250,000 students, equal to approximately 22 percent of elementary and secondary public-school students.

As an experiment, this Report applies the Arizona ESA program to Wyoming. The purpose is not to give a full account of what an ESA program would mean to our state's education system, but rather to provide some reference points for further discussion of the economic implications of an ESA program. For this purpose, the legal side of the ESA model are disregarded. It is assumed that home-schooling parents are equally eligible, while maintaining their freedom to decline the ESA entirely on the same terms as they decline public-school enrollment for their children.

The experiment is based on historic data, a method best known as "retroactive forecasting". It provides a reasonable comparison of a reform to actual historic events, under a strict assumption that all other variables are kept constant. While not an immediate basis for extrapolation or conclusions regarding the effects of said reform, retroactive forecasting explains the impact of a reform in a tangible context, suitable for more detailed research into the reform.

The first step in this experiment is to establish an ESA funding rate. In Arizona, average per-student ESA funding has exceeded average per-student spending in public schools by 76 percent: for every $1 spent on an elementary or secondary student in Arizona public schools, the ESA program has paid out $1.76 per participant. This is not a realistic ratio for Wyoming: in 2015, per-student spending in Arizona public schools amounted to $7,772 while the same amount in Wyoming was $16,431.

A more reasonable approach than to import the ratio for ESA spending to public-school spending is to simply import the actual amount of ESA per-student spending. Thus, it is assumed that a Wyoming ESA program, started in 2011, offered $12,174 per participating student. This amount equals 76 percent of per-student spending in Wyoming public schools of that year. For simplicity, it is assumed that the Wyoming ESA keeps its per-student spending locked at a 76-percent ratio to the same figure for public schools. Under this assumption Wyoming ESA per-student spending would evolve as follows:

The next step is to define eligibility. In Arizona, as mentioned, the initial eligibility cap was set at approximately ten percent, subsequently raised to equal about 22 percent of public school enrollees. Since participation among the eligible student population is scant, only reaching 1.64 percent in 2017, it is not reasonable for this experiment to impose such strict caps on eligibility. Instead, it is assumed here that the entire K-12 student body is eligible.

If the Arizona participation rates are applied, the number of students participating in the Wyoming ESA would grow as follows:

If these students were enrolled in private schools, enrollment there would increase noticeably, from an extra three percent in 2011 to an extra 46 percent in 2015 (the latest year with reliable private-school enrollment data).

Implications of the findings

An education-choice market with more than $40 million to be spent by parents is a substantial change to both the K-12 school system in Wyoming and to the market for private education. However, there are a couple of significant uncertainties in this experiment.

The most important uncertain variable is the participation rate. If one percent of Wyoming parents choose to participate initially, and if enrollment then would grow as it did in Arizona, education-choice funding through the ESA would exceed $500 million within eight years. This is a formidable market potential for educators outside of the public-school system.

Another uncertain variable is the public-school funding formula upon which ESA appropriations are based in this experiment. If the ESA program would motivate a large amount of parents to leave public school for other alternatives, there is a possibility that public-school funding would be cut as a result. Wyoming public schools have an unusually high ratio of non-instructional staff to teachers, creating a substantial overhead. With declining enrollment it is likely that school districts will have to reconsider their funding models.

If the ESA funding model were independent of per-student funding of public schools, the program would be financially more stable. Its funding would also be more in line with how the Arizona ESA operates.

There are macroeconomic implications of this ESA experiment, primarily for the private-education industry. However, any estimates of the ESA's impact in terms of new jobs and growth in local economies will have to wait until the aforementioned uncertainties regarding a Wyoming ESA have been sorted out. Since no such effort is currently under way, the results reported here will remain as a stand-alone experiment in the finances of one education-choice model in Wyoming.

---

*) Enrollment and per-student funding figures are obtained from the U.S. Department of Education, National Center for Education Statistics (NCES). Since there is no comparable account of home-schooled students in the NCES database, any figures on ESA user rates reported here are to be considered in this context.