Publication Number: IB99-1 March 5, 1999
When University of Missouri policy has established a price index based mechanism to increase educational fees, it has turned to the Higher Education Price Index (HEPI) because HEPI arguably best reflects the increasing cost of providing education.
From the perspective of a provider of higher education, HEPI is a logical and rational way to reflect increasing prices for higher education's goods and services. From the perspective of the purchaser of higher education, the initial cost to attend and the rate at which cost increases may be more clearly reflected in the Consumer Price Index (CPI) because CPI estimates changes in cost-of-living.
If rate of change in these two indices had coincided, then the consumers' and the producers' perspectives would be congruent. That has not been the case. Over the last 25 years, HEPI has outpaced CPI 15 times and all 15 were in the period 1982 to present. These sequential differences have accumulated and have produced a pattern of expanding difference. The purpose of this section is describe this trend and illustrate some of the components behind the patterns.
In a nutshell, CPI and HEPI are price indexes. Both result from the accumulated prices of a fixed "basket" of goods and services expressed as an index. If the basket cost $100 in 1983 and $110 in 1984, then indexes for 1983 and 1984 would be 100 and 110 using 1983 as a base. Inflation would be 10%. This is of course an over-simplification and use of either index or both should be done with much care. For example, CPI reflects prices paid by urban consumers and the basket of goods has not been rigidly fixed. HEPI components are apparently fixed and reflect higher education spending patterns in 1972. Please note that while a description of the Research and Development Price Index is included, it is very nearly perfectly correlated and equal in magnitude to HEPI due the high degree of overlap in components. In other words, it is redundant as a higher education index.
Figure 1 (PDF 4 KB) and Table 1 (PDF 5 KB) describe the relationship between median Missouri family income and the University of Missouri's educational and required fees (E&RFs) for a resident undergraduate. From these 24 year displays, it is very clear that the University's E&RFs are now a greater percentage of family income in spite of the fact that median family income has increased by about 20% in this same period. After a long period where the relationship between median family income and UM E&RFs was 3% to 4% for a Missouri family of four, it has rapidly increased over the most recent decade to over 8%.
That said, median income for a Missouri family of four is a robust statistic that has high public acceptance and frankly, any statistic has similar limitations when government programs, modern family structures, and parental financial responsibility patterns are considered. The income figure for a Missouri family of four is derived from the Current Population Survey (Bureau of the Census) and per capita personal income estimates (Bureau of Economic Analysis).
Table 2 (PDF 4 KB) and the two associated graphs Figure 2 (PDF 4 KB) and Figure 3 (PDF 5 KB). Table 2 (PDF 4 KB)and Figure 2 (PDF 4 KB)and Figure 3 (PDF 5 KB) illustrate use of the "standard" 1983 base and a 1974 base to show that base-year computation is important. There is no a priori reason to use 1983, it is simply and commonly accepted practice. The danger in using one base-year or another is that differences can be exaggerated or reduced. Figure 2 (PDF 4 KB) and Figure 3 (PDF 5 KB) are consistent in showing the recent pattern of HEPI increasing more rapidly than CPI but the 1974 base HEPI better shows the period of the late 1970s and early 1980s when CPI was increasing faster than HEPI. Table 3 (PDF 6 KB) and displayed as Figure 4 (PDF 7 KB) (1983 base) and Figure 5 (PDF 7 KB) (1974 base) along with CPI and the CPI medical care component. Because HEPI and outpaced CPI over recent history, some or all components of HEPI must be outpacing CPI. (Note that a base 1983 CPI was 165 in 1998.) Both graphs show that the HEPI components library acquisitions (276) and fringe benefits (237) have increased far more rapidly than the other components of HEPI. The more volatile pattern of utilities makes its relative price either more or less than other components depending on the base year used. Using the oil embargo years of the mid 1970s results in relatively low current costs. The next component in order of those that outpaced inflation is of special importance due to its magnitude, professionals' salaries (194). Fully 50% of current HEPI is derived from professionals' salaries and probably about an additional 10% from their fringe benefits. Of this category, faculty salaries is the largest component and constitutes about 37% of HEPI with an additional 7% in their fringe benefits. The graph and table also display CPI and the Medical Care component of CPI for purpose of comparison. Clearly, CPI Medical Care and HEPI benefits have followed a similar course and very highly correlated. In sum, the observed increase in Medical Care as a component of CPI is very similar in shape to the Benefits component of HEPI. To two differ, however, in the effect that they exert on the relative overall index. Benefits is a much large component of HEPI than Medical Care is of CPI owing to the personnel cost driven nature of HEPI. Last, it is important to note that the Higher Education Amendments of 1998 require study of higher education delivery costs and presumably will result in a new HEPI-like index. The College Cost Study is described in a draft Statement of Work document currently available as http://ocfo.ed.gov/gophroot/5contracts/DocsOnLine/. The results of this study are to be delivered to Congress by September 30, 2002. Figure 6, (PDF 5 KB), national faculty salaries in 1998 are higher than any year but, the first year shown, 1974. Whether faculty salaries have increased more or much more than CPI depends upon the base year comparison -- More if 74 is the base and much more if 83 is the base. A rough computation suggests that more than 7% of the 20% difference in 1998's CPI and HEPI can be attributed to faculty salaries increasing more rapidly than CPI. It is also important to note that Figure 4 (PDF 7 KB) and Figure 5 (PDF 7 KB) (HEPI Components) showed salaries paid professional outpacing CPI but the salaries of non-professionals and service employees did not. So, while faculty are better paid so are professionals generally. Figure 7 (PDF 5 KB) shows the average 9 month salary paid Missouri faculty over nearly 25 years. It is readily apparent that the pattern is very similar to that of AAUP professors at public doctoral institutions and that the recent pattern of rapid increase in educational fees has also been a period of rapid increase in faculty salaries. For context, Figure 7 (PDF 5 KB) also shows median income of a Missouri family of four. The median family income trend line shows much variation about its average of about $49,000 (98 $'s). The faculty salary line only recently equaled the median family income line as faculty are experiencing their highest earnings in 25 or more years. The average unweighted faculty salary is about $45,000 (98 $'s). Please note that the relationship between faculty salary and family income is that of an individual to a family of 4, many of which are two earner families. The figure is offered to place salaries within an "industry" in a broader context.
Appendix Table A (PDF 8 KB) provides greater detail of data illustrated in previously referenced tables and figures.
Description of CPI (Including a Discussion of the Boskin Commission Report)
Discussion of College and University Price Indexes
