- Email: [email protected]
Assessing the economic return of specializing in orthodontics or oral and maxillofacial surgery
E
N
D
S
ABSTRACT A
P
methods—the internal rate of return, or IRR, and net present value, or NPV—to determine whether or not positive economic returns are realized from additional training in these specialties. The findings have a twofold impact. First, they are relevant for the recent dental school graduate contemplating entry into one of the specialties, since the results predict the economic feasibility of entering these fields. Second,
A
IO N
I
rospective candidates for the medical professions enter these professions primarily because of their interest in improving patients’ health. Although health care is the main impetus for entrance, students realize that their livelihood will depend in large part on the particular profession chosen. Therefore, candidates for the health professions and their specialties may need to balance the goal of benevolence with the The positive economic realities before investing in financial the advanced professional training that returns is a prerequisite for these fields. The purpose of our study was to brought by determine the economic viability of specialization entering orthodontics or oral and maxindicate that illofacial surgery vs. practicing as a genthe demand for eral dentist. We measured this viability additional by examining the marginal returns from dental additional human capital formation education (that is, the additional revenue gained from an investment of additional should resources in education). To this end, we continue. used two standard capital investment
T
DAVID W. CORDES, D.M.D., M.D.S.; NEVILLE DOHERTY, Ph.D.; RENE LOPEZ, M.P.H.
CON
Assessing the economic return of specializing in orthodontics or oral and maxillofacial surgery
D
J Background. The ✷ ✷ ® authors determined the economic returns from an educational investN ment in the specialties of A U I N G E D U C 1 R orthodontics and oral and TICLE maxillofacial surgery. They also addressed problems found in previous studies. Methods. The marginal return for specializing in orthodontics or oral and maxillofacial surgery was determined using net present value and internal rate of return, or IRR, with the income of the general dentist serving as the common opportunity cost. Extreme scenario, threshold and one-way sensitivity analyses were used to account for variation in the data. Results. The median group of orthodontists broke even 5.9 years after specialty training and had a working lifetime net return of $271,536 above that of general dentists; the IRR for them was 10.36 percent. The median group of oral and maxillofacial surgeons broke even 2.3 years after training and had a working lifetime net return of $587,563 above that of general dentists; the IRR for them was 25.30 percent. Conclusions. Under the most likely conditions, the authors found a positive economic return to dentists in both specialties from their additional dental training. Practice Implications. The positive financial returns brought by specialization indicate that the demand for additional dental education should continue within an environment of increased educational investment costs. T
R
A
T
the findings are relevant for dental educators, since they influence the demand for various levels of dental education and the profession’s supply of manpower. PREVIOUS RESEARCH
Several studies have shown a relationship between expected economic returns and the demand for additional profes-
JADA, Vol. 132, December 2001 Copyright ©1998-2001 American Dental Association. All rights reserved.
1679
T R E N D S
sional training. Wilkinson1 showed that while difgeneral practitioners, and found a higher rate of ferences in ability and tastes, market imperfecreturn for the specialist group. More recently, tions and changes in demand for types of occupaWeeks and colleagues9 determined the return on tions may account for part of the differences an educational investment to various profesbetween enrollments, discounted earnings—or sionals, including general dentists and all dental NPV of returns—are a significant factor in a specialists. They reported that general dentists person’s choice of occupation. In looking at the had an IRR of 20.73 percent, whereas all specialdeterminants of demand for medical education, ists (grouped together) had an IRR of 27.06 perSloan2 demonstrated that career decisions are cent, when adjusted for hours worked. responsive to interoccupational differences in In both studies,3,9 individual specialties cannot expected financial returns. He found, for example, be distinguished, and the earnings of a highthat a $10 increase in physicians’ annual income school graduate were used as the opportunity correlated with a range of four to 13 people being cost. NPV and IRR require a common opportunity attracted to medicine. His study also showed that cost. Although the authors attempted to provide expected long-run returns are only one determione by using only high-school graduates who purnant of demand. sued professions, the investigators may have Short-run costs. Short-run incurred costs selected an educational stage that was too early. are another factor. With each annual dollar Subsequent measurements would not be proporincrease in the direct cost of medical tionately different between profeseducation, six to 14 applicants are sions, but the absolute values of the discouraged from entering the field. Most of the economic measurements would be lower if Likewise, for a dental career, several research done to date college graduates were the comparauthors have shown that students’ ison group. has focused on decision to apply to dental schools is Finally, although the authors returns to general relative to alternative investments attempted to account for additional dentists, with little (such as investment in government dental specialty training time, cerattention given to bonds) and is positively associated tain generalizations limit the appliwith the dental rate of return on cability of the results to any one specialists. human capital investment.3-6 specialty. The median income for all of the specialties was used; direct Possible methodological probeducational costs of tuition for spelems. However, each of these studies cialty residents were excluded; and the analysis may have had methodological difficulties. included only two years of specialty training time, Maurizi3,4 and the American Dental Association’s while, in fact, specialty training can range from Bureau of Economic and Behavioral Research5 two to six years (for example, surgical training). possibly overestimated the rate of return by using We examined the specialties of orthodontics noncomparable groups, while Dunlevy and and oral and maxillofacial surgery separately and Niessen6 did not adjust for the different number of hours worked at different stages of a dental compared them with general dentistry, using the career, or for any possible differences in the income of the general dentist to provide the number of hours worked between professions. opportunity cost. By accounting for differences in These authors also included living expenses as a training time, educational costs and hours component of the direct costs of financing a dental worked, the study addresses the possible probeducation. This is somewhat surprising since lems faced by previous studies and provides a Friedman7 and Stigler8 previously had shown that methodologically sound basis for comparison. living expenses are common to any educational MATERIALS AND METHODS pursuit or occupation and should not be considered part of an analysis of returns on investment. We used NPV and IRR to determine the marginal Most of the economic research done to date in return on an educational investment for dentists the dental sector has focused on returns to genspecializing in orthodontics or oral and maxilloeral dentists, with little attention given to spefacial surgery. The use of these two methods in cialists. In 1969, however, Maurizi3 compared the conjunction with one another follows the recom1961 rates of return for orthodontists and nonmendation of Nas.10 orthodontic specialists as a group with those for NPV. The NPV, in human capital terms, com1680
JADA, Vol. 132, December 2001 Copyright ©1998-2001 American Dental Association. All rights reserved.
T R E N D S
pares in present dollars the future wage difference between two occupations, after deducting the additional training expenses required by one of the occupations. In this study, the NPV was defined as the current value of the future wage difference between a dental school graduate and an age-matched graduate with further training in one of the two specialties, after all educational expenses from the additional training were subtracted. n
NPV = ∑ FR j j = 0 (1 + r) j+1 FRj is the sum of the annual values for the hourly flow of returns to a specialist, j is the number of years from 0 to the nth year and r is the discounted rate of an alternative investment. IRR. The IRR was defined as the discount rate for which the NPV becomes zero (in the equation below, π is the discount rate). n
IRR = ∑ j=0
FR j = 0 (1 + π) j+1
To account for possible differences in work time, we adjusted the number of hours of labor in each area of dentistry.9,11 The resulting measures are based on hourly wages and expressed as hours-adjusted NPV and hours-adjusted IRR. There are two major potential limitations associated with using the IRR alone, which are controlled for by using the NPV along with it. These are that an investment may have more than one IRR if the net stream of returns changes signs (that is, positive and negative returns) more than once during the life of the investment,1 and that because it is a percentage, IRR is not sensitive to magnitude and may be misleading when investments of different sizes are compared. The NPV eliminates these problems by its summative nature and use of the present dollar as its unit of measure. Study protocol. Following the protocol established by Weeks and colleagues,9 we assumed that students entered and completed a four-year dental education program immediately after attending a four-year baccalaureate program. This established the age of a dental school graduate as 26 years. We then assumed that the dental school graduates proceeded directly into specialty training, without an interim period of
advanced general dental training or general practice. Thus, our investment analysis began at the time of dental school graduation, with the cost of becoming a general practitioner regarded as part of the minimum necessary expense common to both general dentists and specialists. We then established a working lifetime from dental school graduation to age 65 years to provide a finite number of years during which we calculated specialists’ marginal returns. We assembled cross-sectional data in regard to length of training, educational costs and stipends associated with U.S. residency programs,12 as well as incomes and hours worked for general dentists, orthodontists and surgeons. Data regarding incomes and hours worked for general dentists, orthodontists and surgeons were obtained from “The 1993 Survey of Dental Practice: Income From the Private Practice of Dentistry”13 and “The 1993 Survey of Dental Practice: Specialists in Private Practice.”14 We assumed that residents worked as many hours during their educational training period as did the specialty-matched practitioners.9 Cost and income data were expressed in 1992 dollars, and the future flow of returns was discounted using the interest rate on longterm government securities (7.52 percent). Despite the vast sample size derived from the ADA’s 1993 “Survey of Dental Practice: Specialists in Private Practice,”14 the age interval data for “under five years since graduation” was insufficient. Therefore, we designed a mail survey to obtain the information for this subsample. We sent the survey to 100 percent of dentists who graduated from residency programs from 1988 through 1991. (Their names were drawn from the 1995 Directory of Members for the American Association of Orthodontists15 and the 1996 Directory of Members for the American Association of Oral and Maxillofacial Surgeons.16) The survey contained some questions that were identical to those on the ADA survey and used the same inclusion criteria to determine the annual number of hours spent in practice, as well as to assess the direct educational costs. To account for variation in the data and to measure the robustness of the results according to the categorization of Briggs and colleagues, we used three types of sensitivity analyses: extreme scenario, threshold and one-way.17,18 RESULTS
This study used NPV and IRR to determine the
JADA, Vol. 132, December 2001 Copyright ©1998-2001 American Dental Association. All rights reserved.
1681
CUMULATIVE NET PRESENT VALUE ($ PER HOUR)
T R E N D S
350 300 250 200 150 100
■ ■
50 0 -50
■
■
■ ■
■ ●● ■ ● ■ ● ■ ● ● ■● ● ■
●●
●
●
■
●
■
●
■
●
■
●
■
●
■
●
■
●
■
●
■
■■
■
■■ ■ ■ ■ ■■ ■ ■ ■ ■ ■ ■ ■
●● ●● ●● ● ● ● ●
● ● ● ● ● ● ●●
●
Orthodontists
■
Oral and Maxillofacial Surgeons
-100 27 29 31 33 35 37 39 41 43 45 4749 51 53 55 57 59 61 63
AGE (YEARS)
Figure. Hours-adjusted net present value of educational investment for orthodontists and oral and maxillofacial surgeons, discounted at 7.52 percent with central values for parameters.
marginal return on an educational investment for dentists specializing in orthodontics or oral and maxillofacial surgery. To provide a methodologically sound comparison between specialists and general dentists, we used general dentists’ income to calculate opportunity cost, and addressed differences in training time, educational costs and hours worked. NPV and IRR for orthodontists. Using general dentists’ income as the opportunity cost resulted in general dentists having an NPV of $0.00, regardless of the discount rate, since the period of investment commenced at the end of dental school. In comparison, the working lifetime cumulative NPV, or NPVc, for orthodontists was $271,536 when unadjusted for hours of labor. In other words, an orthodontist could expect to earn, in present value terms, $271,536 more than a general practitioner. When averaged over a working lifetime, this amount translates to an additional $7,146 per year. In terms of an IRR, the investment had a marginal annual yield of 10.36 percent over the orthodontist’s working lifetime (that is, the cost of investing in specialty training yielded a return of 10.36 percent annually over the working lifetime). When adjusted for hours of labor, the average NPV, or NPVa, was $6.10 per hour. This figure represents the current value of the future additional hourly wages from specializing in orthodontics. The hours-adjusted IRR of 16.62 percent was a 6.26 percent increase over the unadjusted rate. This increase is 1682
attributable to the fact that orthodontists worked an average of 243.9 fewer hours per year and 9,268 fewer hours per working lifetime than did general dentists. NPV and IRR for surgeons. Surgeons experienced an even higher return on their educational investment than did orthodontists. The surgeons’ NPVc was $587,563, which is more than twice the amount for orthodontists. The annual NPVa was $15,462. When expressed as an hourly wage, surgeons earned an additional $8.40 per hour from their investment. In terms of a percentage yield, the unadjusted IRR was 25.30 percent and, when adjusted for hours worked, the IRR increased to 26.83 percent. Although surgeons spent twice the number of years in training and 198 hours more per year in practice than did orthodontists, the break-even point for surgeons occurred 11⁄2 years earlier. (The break-even point is the point at which a specialist begins to earn more than a general dentist after the additional educational costs have been taken into account.) Surgeons broke even at age 32 years, recovering their investment 2.3 years after completing residency training. Orthodontists, on the other hand, broke even at age 33 years, 5.9 years after completing their residency training (Figure). With respect to the rate of increase in the positive flow of returns, up to age 39 years, the surgeons’ investment grew more rapidly than did the orthodontists’ investment. At age 39 years, however, the rate for surgeons slowed, narrowing the gap between the specialties in terms of NPVc. The orthodontists’ rate of increase slowed at age 47 years and never reached the level obtained by the surgeons (Figure). The positive findings and significant results, however, were not robust enough to withstand several of the variations in sensitivity analyses. At the lowest incomes (for example, first quartile), there were no differences in the NPV between specialists and general dentists. DISCUSSION
Using NPV and IRR, we measured the marginal return from an educational investment in orthodontics and oral and maxillofacial surgery. The results indicated that under the most likely conditions (such as a median number of years of specialty training), there was a positive economic return to dentists in both specialties from the human capital formation beyond that of a general dental degree. In other words, under most eco-
JADA, Vol. 132, December 2001 Copyright ©1998-2001 American Dental Association. All rights reserved.
T R E N D S
nomic conditions, the majority of dentists who discrepancy is the “equalizing difference” and the chose to specialize in orthodontics or surgery were remainder is due to external regulations that proable to recover their specialty training costs as duce noncompeting labor groups. The equalizing well as earn an additional return. differences may be attributed to the internal facUsing central value estimates (that is, median tors of training costs, occupational attractiveness values), we found that orthodontists earned $6.10 and relative risk. more per hour, in present value terms, than did Both orthodontists and surgeons had direct or general dentists. Over a working lifetime, this indirect training costs or both beyond those of a accrued to $271,536 in additional income (present general dental degree program and, therefore, value). When viewed as an interest rate, these some marginal economic returns to specialists earnings translated to 16.62 percent more for could be expected. With regard to occupational orthodontists than for general dentists, as a reattractiveness, the working conditions of a sursult of their additional educational investment. geon generally are considered to be less desirable Surgeons received a greater return on their than those of an orthodontist or a general dentist. educational investment compared with orthodonSurgeons are called to emergencies more fretists, regardless of the economic measure used. quently, keep irregular and grueling hours, and Using central value estimates, we must allow for unproductive office found that oral and maxillofacial surhours while at the hospital. In geons earned $2.30 more per hour addition, because of the higher Surgeons received a than did orthodontists. Over an entire morbidity and mortality risks assogreater return on career, this translates to $134,269, in ciated with many surgical procetheir educational present value terms. When viewed as dures, patients will pay a premium investment compared for the services of a surgeon. As a an interest rate, surgeons earned with orthodontists, 10.21 percent more on their investresult, the surgeon is able to comment than did orthodontists. Again, mand a higher return than an regardless of the this percentage has been adjusted to orthodontist or a general dentist. economic measure reflect the fact that, on average, oral State licensing laws and enrollused. and maxillofacial surgeons worked ment limits in residency programs 198 more hours per year than did restrict competition and produce orthodontists. However, surgeons barriers than hinder people from worked an average of 45.9 fewer hours per year entering the professions. The annual applicant-tothan did general dentists. enrollment ratio is evidence that there is an Educational investment required. Another unfilled demand for specialty education. As a difference between the two specialties is the type result, the labor supply is restricted and specialof educational investment required. Dentists in ists’ wages remain fixed at levels higher than the both specialties were subject to the opportunity supply-meets-demand equilibrium amount. cost of a general practitioner’s income. OrthoCONCLUSION dontic training also required a direct educational cost in the form of tuition and interest payable on The results of this study indicate that, under the tuition and living expenses. Surgical residents, on most likely conditions, dentists receive a positive the other hand, received a stipend large enough return on their additional human capital formato eliminate the direct educational costs and to be tion by specializing in orthodontics or oral and included as income. In addition, surgical training maxillofacial surgery. This positive return—all required a minimum of four years rather than things being equal—should continue to attract the two years required for orthodontic training. dental school graduates to these specialties, proNevertheless, surgeons broke even 11⁄2 years viding the work force with sufficient manpower. earlier than did orthodontists. The amount of average positive returns and the Theory of occupational choice. The discreptemporal position of the break-even points associancy in economic returns and the ranking order of ated with these specialties also may have implicathose returns between general dentists, orthodontions for dental school administrators. The results tists and oral and maxillofacial surgeons are conmay indicate that dental schools have some latisistent with the classic economic theory of occupatude in setting program fees and length of study tional choice. From that theory, a portion of the while still enabling dentists to earn high enough JADA, Vol. 132, December 2001 Copyright ©1998-2001 American Dental Association. All rights reserved.
1683
T R E N D S
returns at reasonable break-even points to attract them to one of these two specialties. ■ Dr. Cordes is in private orthodontic practice, 82 Broad St., Westfield, Mass. 01085, e-mail “[email protected]”. Address reprint requests to Dr. Cordes. Dr. Doherty is the former chairman of the Department of Behavioral Science, University of Connecticut School of Dental Medicine, Farmington, and a retired professor of economics. Mr. Lopez is the executive administrator for the Bluestone Center for Clinical Research, New York University College of Dentistry, New York City. 1. Wilkinson B. Present values of lifetime earnings for different occupations. J Political Economy 1966;74:556-72. 2. Sloan F. The demand for higher education: the case of medical school applicants. J Hum Res 1971;6:466-89. 3. Maurizi A. Economic essays on the dental profession. Iowa City, Iowa: University of Iowa; 1969:14-6. 4. Maurizi A. Rates of return to dentistry and the decision to enter dental school. J Hum Res 1975;10:521-8. 5. Bureau of Economic and Behavioral Research. The dental school applicant pool and the rate of return to dentistry. JADA 1982;105(2): 271-5. 6. Dunlevy H, Niessen L. An economic analysis of the return on investment for dental education. J Dent Educ 1984;48:453-7. 7. Friedman M. Price theory: a provisional text. Chicago: Aldine; 1962:199-225.
1684
8. Stigler G. The theory of price. 3rd ed. New York: Macmillan; 1966: 257-87. 9. Weeks W, Wallace A, Wallace MM, Welch HG. A comparison of the educational costs and incomes of physicians and other professionals. N Engl J Med 1994;330:1280-6. 10. Nas T. Cost-benefit analysis: Theory and application. Thousand Oaks, Calif.: Sage Publications; 1996:117-29. 11. Burstein P, Cromwell J. Relative incomes and rates of return for U.S. physicians. J Health Econ 1985;4(1):63-78. 12. American Dental Association. The 1992/93 annual report on advanced dental education. Chicago: American Dental Association, Department of Educational Surveys; 1993. 13. American Dental Association. The 1993 survey of dental practice: Income from the private practice of dentistry. Chicago: American Dental Association, Bureau of Economic and Behavioral Research; 1993. 14. American Dental Association. The 1993 survey of dental practice: Specialists in private practice. Chicago: American Dental Association, Survey Center; 1995. 15. American Association of Orthodontists. 1995 directory of members. St. Louis: American Association of Orthodontists; 1995. 16. American Association of Oral and Maxillofacial Surgeons. 1996 directory of members. Rosemont, Ill.: American Association of Oral and Maxillofacial Surgeons; 1996. 17. Briggs A, Sculpher M, Buxton M. Uncertainty in the economic evaluation of health care technologies: the role of sensitivity analysis. Health Econ 1994;3(2):95-104. 18. Briggs A, Sculpher M. Sensitivity analysis in economic evaluation: a review of published studies. Health Econ 1995;4:355-71.
JADA, Vol. 132, December 2001 Copyright ©1998-2001 American Dental Association. All rights reserved.
N
D
S
ABSTRACT A
P
methods—the internal rate of return, or IRR, and net present value, or NPV—to determine whether or not positive economic returns are realized from additional training in these specialties. The findings have a twofold impact. First, they are relevant for the recent dental school graduate contemplating entry into one of the specialties, since the results predict the economic feasibility of entering these fields. Second,
A
IO N
I
rospective candidates for the medical professions enter these professions primarily because of their interest in improving patients’ health. Although health care is the main impetus for entrance, students realize that their livelihood will depend in large part on the particular profession chosen. Therefore, candidates for the health professions and their specialties may need to balance the goal of benevolence with the The positive economic realities before investing in financial the advanced professional training that returns is a prerequisite for these fields. The purpose of our study was to brought by determine the economic viability of specialization entering orthodontics or oral and maxindicate that illofacial surgery vs. practicing as a genthe demand for eral dentist. We measured this viability additional by examining the marginal returns from dental additional human capital formation education (that is, the additional revenue gained from an investment of additional should resources in education). To this end, we continue. used two standard capital investment
T
DAVID W. CORDES, D.M.D., M.D.S.; NEVILLE DOHERTY, Ph.D.; RENE LOPEZ, M.P.H.
CON
Assessing the economic return of specializing in orthodontics or oral and maxillofacial surgery
D
J Background. The ✷ ✷ ® authors determined the economic returns from an educational investN ment in the specialties of A U I N G E D U C 1 R orthodontics and oral and TICLE maxillofacial surgery. They also addressed problems found in previous studies. Methods. The marginal return for specializing in orthodontics or oral and maxillofacial surgery was determined using net present value and internal rate of return, or IRR, with the income of the general dentist serving as the common opportunity cost. Extreme scenario, threshold and one-way sensitivity analyses were used to account for variation in the data. Results. The median group of orthodontists broke even 5.9 years after specialty training and had a working lifetime net return of $271,536 above that of general dentists; the IRR for them was 10.36 percent. The median group of oral and maxillofacial surgeons broke even 2.3 years after training and had a working lifetime net return of $587,563 above that of general dentists; the IRR for them was 25.30 percent. Conclusions. Under the most likely conditions, the authors found a positive economic return to dentists in both specialties from their additional dental training. Practice Implications. The positive financial returns brought by specialization indicate that the demand for additional dental education should continue within an environment of increased educational investment costs. T
R
A
T
the findings are relevant for dental educators, since they influence the demand for various levels of dental education and the profession’s supply of manpower. PREVIOUS RESEARCH
Several studies have shown a relationship between expected economic returns and the demand for additional profes-
JADA, Vol. 132, December 2001 Copyright ©1998-2001 American Dental Association. All rights reserved.
1679
T R E N D S
sional training. Wilkinson1 showed that while difgeneral practitioners, and found a higher rate of ferences in ability and tastes, market imperfecreturn for the specialist group. More recently, tions and changes in demand for types of occupaWeeks and colleagues9 determined the return on tions may account for part of the differences an educational investment to various profesbetween enrollments, discounted earnings—or sionals, including general dentists and all dental NPV of returns—are a significant factor in a specialists. They reported that general dentists person’s choice of occupation. In looking at the had an IRR of 20.73 percent, whereas all specialdeterminants of demand for medical education, ists (grouped together) had an IRR of 27.06 perSloan2 demonstrated that career decisions are cent, when adjusted for hours worked. responsive to interoccupational differences in In both studies,3,9 individual specialties cannot expected financial returns. He found, for example, be distinguished, and the earnings of a highthat a $10 increase in physicians’ annual income school graduate were used as the opportunity correlated with a range of four to 13 people being cost. NPV and IRR require a common opportunity attracted to medicine. His study also showed that cost. Although the authors attempted to provide expected long-run returns are only one determione by using only high-school graduates who purnant of demand. sued professions, the investigators may have Short-run costs. Short-run incurred costs selected an educational stage that was too early. are another factor. With each annual dollar Subsequent measurements would not be proporincrease in the direct cost of medical tionately different between profeseducation, six to 14 applicants are sions, but the absolute values of the discouraged from entering the field. Most of the economic measurements would be lower if Likewise, for a dental career, several research done to date college graduates were the comparauthors have shown that students’ ison group. has focused on decision to apply to dental schools is Finally, although the authors returns to general relative to alternative investments attempted to account for additional dentists, with little (such as investment in government dental specialty training time, cerattention given to bonds) and is positively associated tain generalizations limit the appliwith the dental rate of return on cability of the results to any one specialists. human capital investment.3-6 specialty. The median income for all of the specialties was used; direct Possible methodological probeducational costs of tuition for spelems. However, each of these studies cialty residents were excluded; and the analysis may have had methodological difficulties. included only two years of specialty training time, Maurizi3,4 and the American Dental Association’s while, in fact, specialty training can range from Bureau of Economic and Behavioral Research5 two to six years (for example, surgical training). possibly overestimated the rate of return by using We examined the specialties of orthodontics noncomparable groups, while Dunlevy and and oral and maxillofacial surgery separately and Niessen6 did not adjust for the different number of hours worked at different stages of a dental compared them with general dentistry, using the career, or for any possible differences in the income of the general dentist to provide the number of hours worked between professions. opportunity cost. By accounting for differences in These authors also included living expenses as a training time, educational costs and hours component of the direct costs of financing a dental worked, the study addresses the possible probeducation. This is somewhat surprising since lems faced by previous studies and provides a Friedman7 and Stigler8 previously had shown that methodologically sound basis for comparison. living expenses are common to any educational MATERIALS AND METHODS pursuit or occupation and should not be considered part of an analysis of returns on investment. We used NPV and IRR to determine the marginal Most of the economic research done to date in return on an educational investment for dentists the dental sector has focused on returns to genspecializing in orthodontics or oral and maxilloeral dentists, with little attention given to spefacial surgery. The use of these two methods in cialists. In 1969, however, Maurizi3 compared the conjunction with one another follows the recom1961 rates of return for orthodontists and nonmendation of Nas.10 orthodontic specialists as a group with those for NPV. The NPV, in human capital terms, com1680
JADA, Vol. 132, December 2001 Copyright ©1998-2001 American Dental Association. All rights reserved.
T R E N D S
pares in present dollars the future wage difference between two occupations, after deducting the additional training expenses required by one of the occupations. In this study, the NPV was defined as the current value of the future wage difference between a dental school graduate and an age-matched graduate with further training in one of the two specialties, after all educational expenses from the additional training were subtracted. n
NPV = ∑ FR j j = 0 (1 + r) j+1 FRj is the sum of the annual values for the hourly flow of returns to a specialist, j is the number of years from 0 to the nth year and r is the discounted rate of an alternative investment. IRR. The IRR was defined as the discount rate for which the NPV becomes zero (in the equation below, π is the discount rate). n
IRR = ∑ j=0
FR j = 0 (1 + π) j+1
To account for possible differences in work time, we adjusted the number of hours of labor in each area of dentistry.9,11 The resulting measures are based on hourly wages and expressed as hours-adjusted NPV and hours-adjusted IRR. There are two major potential limitations associated with using the IRR alone, which are controlled for by using the NPV along with it. These are that an investment may have more than one IRR if the net stream of returns changes signs (that is, positive and negative returns) more than once during the life of the investment,1 and that because it is a percentage, IRR is not sensitive to magnitude and may be misleading when investments of different sizes are compared. The NPV eliminates these problems by its summative nature and use of the present dollar as its unit of measure. Study protocol. Following the protocol established by Weeks and colleagues,9 we assumed that students entered and completed a four-year dental education program immediately after attending a four-year baccalaureate program. This established the age of a dental school graduate as 26 years. We then assumed that the dental school graduates proceeded directly into specialty training, without an interim period of
advanced general dental training or general practice. Thus, our investment analysis began at the time of dental school graduation, with the cost of becoming a general practitioner regarded as part of the minimum necessary expense common to both general dentists and specialists. We then established a working lifetime from dental school graduation to age 65 years to provide a finite number of years during which we calculated specialists’ marginal returns. We assembled cross-sectional data in regard to length of training, educational costs and stipends associated with U.S. residency programs,12 as well as incomes and hours worked for general dentists, orthodontists and surgeons. Data regarding incomes and hours worked for general dentists, orthodontists and surgeons were obtained from “The 1993 Survey of Dental Practice: Income From the Private Practice of Dentistry”13 and “The 1993 Survey of Dental Practice: Specialists in Private Practice.”14 We assumed that residents worked as many hours during their educational training period as did the specialty-matched practitioners.9 Cost and income data were expressed in 1992 dollars, and the future flow of returns was discounted using the interest rate on longterm government securities (7.52 percent). Despite the vast sample size derived from the ADA’s 1993 “Survey of Dental Practice: Specialists in Private Practice,”14 the age interval data for “under five years since graduation” was insufficient. Therefore, we designed a mail survey to obtain the information for this subsample. We sent the survey to 100 percent of dentists who graduated from residency programs from 1988 through 1991. (Their names were drawn from the 1995 Directory of Members for the American Association of Orthodontists15 and the 1996 Directory of Members for the American Association of Oral and Maxillofacial Surgeons.16) The survey contained some questions that were identical to those on the ADA survey and used the same inclusion criteria to determine the annual number of hours spent in practice, as well as to assess the direct educational costs. To account for variation in the data and to measure the robustness of the results according to the categorization of Briggs and colleagues, we used three types of sensitivity analyses: extreme scenario, threshold and one-way.17,18 RESULTS
This study used NPV and IRR to determine the
JADA, Vol. 132, December 2001 Copyright ©1998-2001 American Dental Association. All rights reserved.
1681
CUMULATIVE NET PRESENT VALUE ($ PER HOUR)
T R E N D S
350 300 250 200 150 100
■ ■
50 0 -50
■
■
■ ■
■ ●● ■ ● ■ ● ■ ● ● ■● ● ■
●●
●
●
■
●
■
●
■
●
■
●
■
●
■
●
■
●
■
●
■
■■
■
■■ ■ ■ ■ ■■ ■ ■ ■ ■ ■ ■ ■
●● ●● ●● ● ● ● ●
● ● ● ● ● ● ●●
●
Orthodontists
■
Oral and Maxillofacial Surgeons
-100 27 29 31 33 35 37 39 41 43 45 4749 51 53 55 57 59 61 63
AGE (YEARS)
Figure. Hours-adjusted net present value of educational investment for orthodontists and oral and maxillofacial surgeons, discounted at 7.52 percent with central values for parameters.
marginal return on an educational investment for dentists specializing in orthodontics or oral and maxillofacial surgery. To provide a methodologically sound comparison between specialists and general dentists, we used general dentists’ income to calculate opportunity cost, and addressed differences in training time, educational costs and hours worked. NPV and IRR for orthodontists. Using general dentists’ income as the opportunity cost resulted in general dentists having an NPV of $0.00, regardless of the discount rate, since the period of investment commenced at the end of dental school. In comparison, the working lifetime cumulative NPV, or NPVc, for orthodontists was $271,536 when unadjusted for hours of labor. In other words, an orthodontist could expect to earn, in present value terms, $271,536 more than a general practitioner. When averaged over a working lifetime, this amount translates to an additional $7,146 per year. In terms of an IRR, the investment had a marginal annual yield of 10.36 percent over the orthodontist’s working lifetime (that is, the cost of investing in specialty training yielded a return of 10.36 percent annually over the working lifetime). When adjusted for hours of labor, the average NPV, or NPVa, was $6.10 per hour. This figure represents the current value of the future additional hourly wages from specializing in orthodontics. The hours-adjusted IRR of 16.62 percent was a 6.26 percent increase over the unadjusted rate. This increase is 1682
attributable to the fact that orthodontists worked an average of 243.9 fewer hours per year and 9,268 fewer hours per working lifetime than did general dentists. NPV and IRR for surgeons. Surgeons experienced an even higher return on their educational investment than did orthodontists. The surgeons’ NPVc was $587,563, which is more than twice the amount for orthodontists. The annual NPVa was $15,462. When expressed as an hourly wage, surgeons earned an additional $8.40 per hour from their investment. In terms of a percentage yield, the unadjusted IRR was 25.30 percent and, when adjusted for hours worked, the IRR increased to 26.83 percent. Although surgeons spent twice the number of years in training and 198 hours more per year in practice than did orthodontists, the break-even point for surgeons occurred 11⁄2 years earlier. (The break-even point is the point at which a specialist begins to earn more than a general dentist after the additional educational costs have been taken into account.) Surgeons broke even at age 32 years, recovering their investment 2.3 years after completing residency training. Orthodontists, on the other hand, broke even at age 33 years, 5.9 years after completing their residency training (Figure). With respect to the rate of increase in the positive flow of returns, up to age 39 years, the surgeons’ investment grew more rapidly than did the orthodontists’ investment. At age 39 years, however, the rate for surgeons slowed, narrowing the gap between the specialties in terms of NPVc. The orthodontists’ rate of increase slowed at age 47 years and never reached the level obtained by the surgeons (Figure). The positive findings and significant results, however, were not robust enough to withstand several of the variations in sensitivity analyses. At the lowest incomes (for example, first quartile), there were no differences in the NPV between specialists and general dentists. DISCUSSION
Using NPV and IRR, we measured the marginal return from an educational investment in orthodontics and oral and maxillofacial surgery. The results indicated that under the most likely conditions (such as a median number of years of specialty training), there was a positive economic return to dentists in both specialties from the human capital formation beyond that of a general dental degree. In other words, under most eco-
JADA, Vol. 132, December 2001 Copyright ©1998-2001 American Dental Association. All rights reserved.
T R E N D S
nomic conditions, the majority of dentists who discrepancy is the “equalizing difference” and the chose to specialize in orthodontics or surgery were remainder is due to external regulations that proable to recover their specialty training costs as duce noncompeting labor groups. The equalizing well as earn an additional return. differences may be attributed to the internal facUsing central value estimates (that is, median tors of training costs, occupational attractiveness values), we found that orthodontists earned $6.10 and relative risk. more per hour, in present value terms, than did Both orthodontists and surgeons had direct or general dentists. Over a working lifetime, this indirect training costs or both beyond those of a accrued to $271,536 in additional income (present general dental degree program and, therefore, value). When viewed as an interest rate, these some marginal economic returns to specialists earnings translated to 16.62 percent more for could be expected. With regard to occupational orthodontists than for general dentists, as a reattractiveness, the working conditions of a sursult of their additional educational investment. geon generally are considered to be less desirable Surgeons received a greater return on their than those of an orthodontist or a general dentist. educational investment compared with orthodonSurgeons are called to emergencies more fretists, regardless of the economic measure used. quently, keep irregular and grueling hours, and Using central value estimates, we must allow for unproductive office found that oral and maxillofacial surhours while at the hospital. In geons earned $2.30 more per hour addition, because of the higher Surgeons received a than did orthodontists. Over an entire morbidity and mortality risks assogreater return on career, this translates to $134,269, in ciated with many surgical procetheir educational present value terms. When viewed as dures, patients will pay a premium investment compared for the services of a surgeon. As a an interest rate, surgeons earned with orthodontists, 10.21 percent more on their investresult, the surgeon is able to comment than did orthodontists. Again, mand a higher return than an regardless of the this percentage has been adjusted to orthodontist or a general dentist. economic measure reflect the fact that, on average, oral State licensing laws and enrollused. and maxillofacial surgeons worked ment limits in residency programs 198 more hours per year than did restrict competition and produce orthodontists. However, surgeons barriers than hinder people from worked an average of 45.9 fewer hours per year entering the professions. The annual applicant-tothan did general dentists. enrollment ratio is evidence that there is an Educational investment required. Another unfilled demand for specialty education. As a difference between the two specialties is the type result, the labor supply is restricted and specialof educational investment required. Dentists in ists’ wages remain fixed at levels higher than the both specialties were subject to the opportunity supply-meets-demand equilibrium amount. cost of a general practitioner’s income. OrthoCONCLUSION dontic training also required a direct educational cost in the form of tuition and interest payable on The results of this study indicate that, under the tuition and living expenses. Surgical residents, on most likely conditions, dentists receive a positive the other hand, received a stipend large enough return on their additional human capital formato eliminate the direct educational costs and to be tion by specializing in orthodontics or oral and included as income. In addition, surgical training maxillofacial surgery. This positive return—all required a minimum of four years rather than things being equal—should continue to attract the two years required for orthodontic training. dental school graduates to these specialties, proNevertheless, surgeons broke even 11⁄2 years viding the work force with sufficient manpower. earlier than did orthodontists. The amount of average positive returns and the Theory of occupational choice. The discreptemporal position of the break-even points associancy in economic returns and the ranking order of ated with these specialties also may have implicathose returns between general dentists, orthodontions for dental school administrators. The results tists and oral and maxillofacial surgeons are conmay indicate that dental schools have some latisistent with the classic economic theory of occupatude in setting program fees and length of study tional choice. From that theory, a portion of the while still enabling dentists to earn high enough JADA, Vol. 132, December 2001 Copyright ©1998-2001 American Dental Association. All rights reserved.
1683
T R E N D S
returns at reasonable break-even points to attract them to one of these two specialties. ■ Dr. Cordes is in private orthodontic practice, 82 Broad St., Westfield, Mass. 01085, e-mail “[email protected]”. Address reprint requests to Dr. Cordes. Dr. Doherty is the former chairman of the Department of Behavioral Science, University of Connecticut School of Dental Medicine, Farmington, and a retired professor of economics. Mr. Lopez is the executive administrator for the Bluestone Center for Clinical Research, New York University College of Dentistry, New York City. 1. Wilkinson B. Present values of lifetime earnings for different occupations. J Political Economy 1966;74:556-72. 2. Sloan F. The demand for higher education: the case of medical school applicants. J Hum Res 1971;6:466-89. 3. Maurizi A. Economic essays on the dental profession. Iowa City, Iowa: University of Iowa; 1969:14-6. 4. Maurizi A. Rates of return to dentistry and the decision to enter dental school. J Hum Res 1975;10:521-8. 5. Bureau of Economic and Behavioral Research. The dental school applicant pool and the rate of return to dentistry. JADA 1982;105(2): 271-5. 6. Dunlevy H, Niessen L. An economic analysis of the return on investment for dental education. J Dent Educ 1984;48:453-7. 7. Friedman M. Price theory: a provisional text. Chicago: Aldine; 1962:199-225.
1684
8. Stigler G. The theory of price. 3rd ed. New York: Macmillan; 1966: 257-87. 9. Weeks W, Wallace A, Wallace MM, Welch HG. A comparison of the educational costs and incomes of physicians and other professionals. N Engl J Med 1994;330:1280-6. 10. Nas T. Cost-benefit analysis: Theory and application. Thousand Oaks, Calif.: Sage Publications; 1996:117-29. 11. Burstein P, Cromwell J. Relative incomes and rates of return for U.S. physicians. J Health Econ 1985;4(1):63-78. 12. American Dental Association. The 1992/93 annual report on advanced dental education. Chicago: American Dental Association, Department of Educational Surveys; 1993. 13. American Dental Association. The 1993 survey of dental practice: Income from the private practice of dentistry. Chicago: American Dental Association, Bureau of Economic and Behavioral Research; 1993. 14. American Dental Association. The 1993 survey of dental practice: Specialists in private practice. Chicago: American Dental Association, Survey Center; 1995. 15. American Association of Orthodontists. 1995 directory of members. St. Louis: American Association of Orthodontists; 1995. 16. American Association of Oral and Maxillofacial Surgeons. 1996 directory of members. Rosemont, Ill.: American Association of Oral and Maxillofacial Surgeons; 1996. 17. Briggs A, Sculpher M, Buxton M. Uncertainty in the economic evaluation of health care technologies: the role of sensitivity analysis. Health Econ 1994;3(2):95-104. 18. Briggs A, Sculpher M. Sensitivity analysis in economic evaluation: a review of published studies. Health Econ 1995;4:355-71.
JADA, Vol. 132, December 2001 Copyright ©1998-2001 American Dental Association. All rights reserved.