Volume Output Measures and Price Indices of the United States Tertiary Education Services
AbstractThe System of National Accounts and the National Academies of Sciences recommend that output of education services be measured by volume output rather than cost of production as currently done in national accounts for services provided by government entities and non-profit institutions. Measuring education services by volume is important as such output measures allow researchers to consider productivity of the education sector. This paper constructs volume measures of tertiary education services in the United States using three different metrics: the number of students enrolled, the number of credit hours taken, and the number of degrees awarded. I then compare the growth of volume output measures to the published NIPA numbers in higher education: government consumption expenditures and household consumption expenditures. Between 2004 and 2015, the volume output of full-time-equivalent students and credit hours increased at average annual rates of 2.5% and 2.3%, respectively, which grew slower than the government consumption expenditures in higher education (2.7%). On the other hand, household consumption expenditures on higher education grew little in those years (less than 0.1%) presenting discrepancies between the official statistics and what universities actually produce.
The source of discrepancies seems to lie in the price indices used to deflate higher education services. I construct price indices from tuition revenues of universities and net tuition payments by students after discounts and grants are taken into consideration. My estimates show that the GDP deflator used to deflate government services is lower than the price deflator calculated from tuition revenues of public universities while the PCE deflator for higher education substantially overstates the inflation rate of household spending on tuitions. This research points to further areas of possible improvements in how we measure output of education services and price index for it, which in turn could lead to changes in GDP measurement.