Pre-School (Untargeted Programs)

In order to compute the program’s costs, the author adds up three components: i) marginal government outlays in the preschool program, ii) the present value of income taxes collected on extra earnings caused by program participation, and iii) the fiscal savings from substitutions across public programs.

Cascio (2020) estimates government outlays on preschool by adopting two potential strategies. In the first strategy they assume that per-pupil program outlays are equal to per-pupil K-12 outlays (upper bound). In the second strategy they assume that outlays are equal to per-pupil outlays in Head Start.

In order to estimate the impact of pre-school on government tax revenue, the author takes the impact of pre-school on test scores and translates those gains into increased lifetime earnings and, subsequently, increased tax revenue. The paper follows a forecasting approach from Kline and Walters (2016) who use Chetty et al. (2011)‘s result that a one standard deviation increase in test scores yields a 10 percent increase in earnings. To get to the dollar value of increased lifetime earnings, the author multiplies this percentage impact by the total discounted lifetime earnings at age 12 of $522,000 in 2010 dollars from Chetty et al. (2011). To transform the value into lifetime earnings at age 4 the author assumes a 3 percent discount rate. Cascio (2020) then multiplies those earnings gains an average tax rate of 20% to arrive at a figure for increased tax collection.

In order to estimate the impact of the program on spending across other government programs, the author assumes that pupils would attend Head Start if they were not enrolled in a public pre-K program.

The net cost calculations together with the WTP values result in a MVPF of 1.96 for universal pre-K programs, with a standard error of 1.39, implying a 95% confidence interval of to -0.76 to 4.68. Assuming more conservatively that costs are equal to Head Start costs, the MVPF rises to 4.27.