RETURNS TO FIELD OF STUDY VERSUS SCHOOL QUALITY.docx

《RETURNS TO FIELD OF STUDY VERSUS SCHOOL QUALITY.docx》由会员分享，可在线阅读，更多相关《RETURNS TO FIELD OF STUDY VERSUS SCHOOL QUALITY.docx（21页珍藏版）》请在得力文库 - 分享文档赚钱的网站上搜索。

1、RETURNS TO FIELD OF STUDY VERSUS SCHOOL QUALITY: MBA SELECTION ON OBSERVED AND UNOBSERVED HETEROGENEITY WAYNE A. GROVE and ANDREW HUSSEY While a substantial literature has established returns to college major and to school quality, we offer the first such estimates for Masters of Business Administra

2、tion (MBAs). To control for their nonrandom selection of fields, we estimate the returns to MBA concentrations using both ordinary least squares (OLS) with detailed control variables and including individual fixed effects. We find approximately 7% returns for most MBAs but roughly double that for fi

3、nance and management information systems (MIS). Thus, MBA area of study can matter as much or more than program quality: only attending a top 10, but not 11-25, MBA program trumped studying finance and MIS at a nontop 25 program. (JEL I21, J30, J24) I. INTRODUCTION Given the importance of human capi

4、tal investment, prospective college students and parents must often choose between higher- quality-and-cost and lower-quality-and-cost schools. The considerable literature that has examined this trade-off identifies not only school quality as a principal driver of postbaccalaureate earnings but also

5、 students choice over fields of study.1 Despite the attention lavished on school rankings, James et al. (1989), for example, con- clude that . college experience variables (especially major) explain more of the variance (in earnings) than measured family background, ability, and college characterist

6、ics combined (252).2 That is, a prospective baccalaureate *We thank Mark Montgomery, John Robst, Vincent Hevern, and participants at the George Mason University and Le Moyne College seminars. Grove: Professor, Economics Department, Le Moyne Col- lege, Syracuse, NY 13214. Phone 315 445 4262, Fax:315

7、445 4540, E-mail grovewalemoyne.edu Hussey: Assistant Professor, Department of Economics, Fogelman College of Business & Economics, University of Memphis, Memphis, TN 38152. Phone 901 678 1487, Fax 901 678 2685, E-mail ajhusseymemphis.edu 1. See, for example, Altonji (1993), Berger (1988), Day- mont

8、 and Andrisanti (1984), Grogger and Eide (1995), James et al. (1989), Loury (1997), Loury and Garman (1995), Monks (2000), Fitzgerald (2000), Arcidiacono (2004), McDonald and Thornton (2007). 2. Similarly, Fitzgerald (2000) finds that higher educa- tion experiences, namely, grade point averages (GPA

9、s) and major fields of study, explain more of the variation in 1991 male earnings for the 1980 High School and Beyond sam- ple than do institutional characteristics but that both sets of factors are roughly equally important for females. student can expect strong future earnings not only by attendin

10、g a highly ranked institution but also by majoring in a highly rewarded field, such as engineering, the sciences, or business, at a lesser ranked school. Surprisingly, this impor- tant conclusion about the trade-off between con- tent and quality has not been generalized to any other postsecondary de

11、gree programs, such as associate degrees at community colleges or mas- ters of education, Masters of Business Admin- istration (MBA), law, medical, or PhD pro- grams. The goal of this article is to fill this void in the literature by estimating the returns to both program quality and fields of speci

12、alization in the context of graduate education. Specifically, we estimate these returns for the third most commonly earned postsecondary degree, the MBA (Morgeson and Nahrgang 2008). Prior research has estimated a large drop-off in returns to an MBA beyond the nations elite programs (Arcidiacono et

13、al. 2008). Because of this, and because MBAs are likely to face significant opportunity costs of moving to Economic Inquiry (ISSN 0095-2583) Vol. 49, No. 3, July 2011, 730749 730 doi:10.1111/j.1465-7295.2010.00292.x Online Early publication April 27, 2010 2010 Western Economic Association Internatio

14、nal ABBREVIATIONS AACSB: Association of Advance Collegiate Schools or Business GMAC: Graduate Management Admissions Council GMAT: Graduate Management Admission Test GPA: Grade Point Average MBA: Masters of Business Administration MIS: Management Information Systems OLS: Ordinary Least Squares GROVE

15、& HUSSEY: RETURNS TO FIELD OF STUDY VERSUS SCHOOL QUALITY 731 attend a more highly ranked program than what may be available locally, estimating the trade- offs between returns to program quality and to fields of specialization is especially important for MBAs. In the data set used in this study, th

16、e average MBA student had about 6 yr of full-time work experience, half were married, and 27% had children. However, if family and job ties restrict potential MBA students from enrolling in more highly ranked programs, might they nonetheless be able to earn high wage pre- miums by specializing their

17、 studies in fields with higher returns? Our study focuses on MBA areas of con- centration. In response to competitive pressures and consumer demand, many business schools have offered the opportunity for students to go beyond the core MBA curriculum by tailoring additional coursework toward a partic

18、ular area.3 In this way, MBA concentrations mirror the role of undergraduate majors. Unfortunately, to our knowledge, no entity such as the Associa- tion of Advance Collegiate Schools or Business (AACSB) or the Graduate Management Admis- sions Council (GMAC) systematically collects information about

19、 the specific courses that con- stitute areas of concentration by MBA program. The structure and actual courses included in con- centrations within MBA programs vary from school to school, akin to differences among undergraduate business programs.4 However, MBA concentrations typically consist of fo

20、ur to five topical courses, but range as high as seven or more, according to Dierdorff and Rubins (2008) survey of 576 of the 621 programs in the United States accredited by the AACSB.5 Perhaps the main distinction between undergrad- uate and MBA fields of study is that while most U.S. undergraduate

21、 business programs offer the 3. According to Dierdorff and Rubin (2008), 56% of accredited conventional MBA programs now offer at least one concentration and 93% of those offer at least two. An additional 9% of MBA programs offer entirely specialized degrees. 4. For illustrative purposes only, we of

22、fer two current examples of how concentrations are organized. At the Kel- logg MBA program (at Northwestern University), a finance concentrator takes Finance II (Finance I is a Core Curricu- lum requirement) and then three additional finance courses out of six choices. At Dukes Fuqua School of Busin

23、ess, each MBA concentration includes a set of electives from which one chooses six courses four courses in a focal area (e.g., finance) and two courses from other areas that support the focal concentration (referred to as a 42 structure). 5. According to Dierdorff and Rubin (2008), the average numbe

24、r of required courses per concentration varies as fol- lows: accounting, 5.31; finance, 5.2; MIS, 7.55; international business, 3.93; and management, 4.09 (p. 21). full array of standard business fields, that is, general management, finance, accounting, mar- keting, management information systems (M

25、IS), and international business, MBA areas of con- centration, despite their increasing prevalence, are more selectively offered by schools. Fur- thermore, the presence of areas of concentration does not appear to be correlated with overall school quality, suggesting that the quality of pro- gram at

26、tended and the choice of focus of study may be two independent drivers of increased earnings among graduates.6 Besides replicating the returns to quality- versus-content for an increasingly important and diverse advanced degree, doing so for MBAs offers one chief advantage over existing esti- mates

27、for collegiate degrees. Among other post- secondary degree earners, only MBA students typically have substantial full-time work expe- rience. The existence of pre- and post-MBA earnings helps solve a perennial obstacle in estimating the financial returns to college edu- cation, where the lack of pre

28、collegiate full- time earnings prohibits a direct comparison of net returns. Thus, some part of the observed relationship between educational choice and attainment and postgraduation earnings may result from unobserved individual characteristics (Brewer and Ehrenberg 1996; Heckman 1979).7 Consequent

29、ly, researchers have used four strate- gies to identify the wage effect of schooling, that is, to separate the returns to schooling from the effect of observed and unobserved attributes on educational choice and attainment: exclusion restrictions,8 instrumental variables,9 sibling and twin data sets

30、,10 and controlling for selection 6. For example, in 1993 among 25 leading business programs, the following numbers did not offer the following areas of concentration: accounting, 4; finance, 2; MIS, 10; international business, 11; and marketing, 2 (Segev et al. 1999, 555). 7. Some researchers have

31、attempted to account for self- selection concerns by explicitly modeling the students choice of the type of institution of higher education to attend (Brewer, Eide, and Ehrenberg 1999; Montgomery 2002, for full- versus part-time MBA programs) or students choice of field (Arcidiacono 2004; Paglin and

32、 Rufolo 1990). 8. Willis and Rosen (1979) rely on exclusion restrictions in a structural model, using income elasticity estimates for selectivity bias to predict the income associated with each field of study for all students. 9. Other investigators have relied on instrumental vari- ables, for examp

33、le, proximity to colleges or date of birth, to identify the effect of education on earnings (Angrist and Krueger 1991; Kane and Rouse 1995). 10. Twin studies estimate the value of an additional year of education, controlling for family background and common genetic influences (Ashenfelter and Rouse

34、1998; Berhman and Taubman 1989; Berhman et al. 1994, 1996). 732 ECONOMIC INQUIRY with lots of observables.11 In the latter approach, to obtain sufficiently detailed individual infor- mation over time, researchers use a variety of nationally representative longitudinal data sets on labor market outco

35、mes of distinct cohorts of college graduates.12 We offer fixed effects as a fifth approach to identify the wage effect of field of study, as Arcidiacono et al. (2008) did for returns to MBA quality. More specifically, we use two strategies to control for selection into both MBA concen- trations and

36、program quality categories. First, we use a selection-on-observables approach by including in the analysis a nationally repre- sentative longitudinal data set with a partic- ularly rich set of variables observable to the econometrician. Our second strategy exploits the existence of both pre- and pos

37、t-MBA earn- ings13 an anomaly among higher education students, because undergraduate, graduate, doc- toral, and professional degree seekers typically go directly from one educational program to another.14 This important feature of the data allows for the use of individual fixed effects in earning re

38、gressions, which eliminates time- invariant, individual-specific heterogeneity as reflected in an individuals wages. This strat- egy may be considered an improvement over the selection-on-observables approach, in that observable covariates, however numerous they may be, imperfectly proxy for the act

39、ual factors contributing to both educational decisions and education-independent labor market outcomes. Consider, for example, the comparison of person A, who has more innate ability (or motivation, ambition, etc.) and interest in finance, versus person B, who is otherwise observationally 11. Black,

40、 Sanders, and Taylor (2003), for example, identify wage differences associated with college majors by comparing workers with identical demographic characteris- tics (namely, age, race, and ethnicity, based on data from the 1993 National Survey of College Graduates, NSCG). 12. Examples include the Na

41、tional Longitudinal Survey of the (High School) Class of 1972 (NLS-72) cohort (Arcidiacono 2004; Grogger and Eide 1995; James et al. 1989), the High School and Beyond Longitudinal Study of 1980 Sophomores (H&B-So:1980/1992) cohort (Fitzgerald 2000), or the Baccalaureate and Beyond study (B&B: 93/97)

42、 cohort (Thomas and Zhang 2005). 13. At the time of GMAT registration, average work experience among our sample of eventual MBA students exceeded 5.5 yr and over 75% had at least 2 yr of full-time work experience. 14. Boudarbat (2008) examines a rare exception, where 43% of the students in his study

43、 of Canadian community college major choice had prior full-time work experience; work experience is coded as a dichotomous variable rather than by quantifying experience or using earnings. identical but has less such aptitude and prefer- ences for fields of study. Person A is both more likely to sel

44、ect a finance concentration and to achieve greater earnings independent of choos- ing finance, so a simple cross-sectional compar- ison (or the use of ordinary least squares OLS) would lead to upward biased estimates of returns to the finance field. The fixed-effect specifica- tion moves beyond this

45、 comparison and instead investigates the within-individual variation, not requiring a control group of non-MBAs (or nonfinance concentrators) to identify the effect of studying finance on those who obtain an MBA and choose finance as a concentration.15 The data come from the GMAT Registrant Survey,

46、a longitudinal survey of registrants for the Graduate Management Admission Test (GMAT), a standardized examination meant to assess an individuals readiness or propensity for advanced business and management training, which is required by most MBA programs for admission. The survey occurred in four w

47、aves from 1990 to 1998, whether or not the registrant ultimately obtained an MBA. For several reasons, the GMAT Registrant survey is a good source of data to evaluate the returns to MBA fields. First, GMAT test tak- ers comprise a relatively homogeneous group in terms of human capital and career goa

48、ls. Sec- ond, we have good information about scholas- tic aptitude and quality of education because the survey data are linked to test scores and other data from GMAC records. The surveys also provide a wealth of additional information about individuals, including work experience, earnings, and nonc

49、ognitive information about individuals, such as self-assessed soft skills that may proxy for self-confidence. Thus, these data provide extensive information observable to the researcher about worker heterogeneity. Our estimates of the return to MBA areas of concentration suggest average earning gains for most fields of study of around 7% but wage premiums of twice that amount for MIS and finance. Although attending a top 10 program 15. That is, the use of fixed effects allows us, i