Reasons for this concern include the increased cost of dental education as dental schools depend more heavily on tuition and fees as a source of revenue^1,2; lack of support through grants and stipends, as reflected by the tenuous government support for general medical education funding³; and an increase in the average loan for, or indebtedness of, many dental students.^4,5 Given these factors, the need exists to provide a more accurate financial perspective for investing in a dental specialty career as a prosthodontist.

We initiated this study because of the need to determine if entering and completing a three-year residency program in prosthodontics offers an economic advantage over entering general practice immediately on graduation. We used the internal rate of return (IRR), a term borrowed from investments in assets, to help explain why people enter an occupation. The IRR from owning an asset requires knowledge about costs of acquiring the asset (that is, prosthodontic training) and the economic returns from owning the asset during a period.

The IRR is a measure of whether the net returns are large enough to pay for the costs of the investment and yield a profit to the investor. The dentist contemplating a lifetime career in prosthodontics must pay significant costs for the additional education and forgo dental earnings while in training. After completing the training, the new specialist enters practice to realize the economic returns in the form of lifetime earnings from practice. This type of economic decision is particularly complex because the individual must commit to a period of training before he or she can begin to acquire the economic returns from the investment. The IRR is one means of evaluating whether it is worthwhile economically to undertake the residency training before making this commitment.

Researchers have applied the IRR concept to various occupations, including medicine and dentistry.^6–8 The American Dental Association has calculated rates of return for dental education and examined their importance in explaining the number of applicants to dental school.⁹ In 1978, the ADA estimated that the IRR was 25 percent, and current preliminary estimates indicate an IRR of 21 percent (American Dental Association, Dental Economic Advisory Group, Rate of Return to Dental Education, unpublished data, Feb. 27–28, 2003).

In 1994, Weeks and colleagues¹⁰ estimated an hours-adjusted IRR for dental education of 20.7 percent. Cordes and colleagues¹¹ examined the IRR for specialty training for oral and maxillofacial surgeons and orthodontists. Their results included an estimated IRR for orthodontists of 10.4 percent over a working lifetime and 16.6 percent after adjusting for hours of work. In addition, they estimated a working lifetime IRR of 25.3 percent for oral and maxillofacial surgeons and 26.8 percent after adjusting for total hours of work.

	SUBJECTS, MATERIALS AND METHODS

TOP ABSTRACT SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
We designed this study to estimate the IRR for investing in prosthodontic training whereby private practice is the occupational choice once the residency is completed. We based the earnings of private-practice prosthodontists in the United States on results from the 2002 Survey of Prosthodontists conducted by the American College of Prosthodontists (ACP) (unpublished data, 2002). The ACP estimated that 68 percent of active practicing prosthodontists in this country were members of the ACP in 2004 (Edward Cronin, executive director, ACP, written communication, Aug. 10, 2004).

Using a list of practicing prosthodontists provided by the ACP (both members and non-members of ACP), we selected a random sample of 2,500 prosthodontists, including 1,761 (70 percent) who were members of the ACP and 739 (30 percent) who were not members. The sample represented about 77.2 percent of all active prosthodontists in the United States.¹² We included nonmembers of the ACP in sampling because they are part of the total population of private-practice prosthodontists.

An independent firm administered the survey, which included printing and mailing of surveys, processing the receipt of returned questionnaires, mailing follow-up surveys to nonrespondents and transferring survey results to an electronic format. A cover letter was included with the survey that described the purpose of the survey, contained a confidentiality statement with regard to nondisclosure of survey results that would identify any respondent, and provided a contact at the ACP who could verify the legitimacy of the project.

The overall response rate was 50.3 percent (after accounting for practitioners who had died and unclaimed and inapplicable surveys) including a 59.7 percent response from ACP members and a 22.6 percent response from nonmembers. One thousand fifty-one (86 percent) of the 1,222 respondents were members of the ACP compared with 1,761 (70 percent) of 2,500 prosthodontists in the sample.

During 2002, the ACP and ADA¹³ conducted two independent surveys. Each survey was applicable to this study because their respective earnings reports for prosthodontists and general practitioners (GPs) were parallel in timing and each requested information pertaining to 2001 only. The GPs’ earnings are important to this study for two reasons. First, the earnings of the youngest GPs represent income the student must forgo while in prosthodontic training. The forgone earnings are part of the cost of obtaining the prosthodontic training. Second, the lifetime earnings of the GP represent the alternative occupation to practicing as a prosthodontist. The difference in lifetime earnings between prosthodontists and GPs helps us determine whether there is a net economic advantage to the dentist from entering residency training rather than practicing as a GP.

We based our data about prosthodontic residency programs’ tuition/fees and stipends on those collected from the 2001/02 Survey of Advanced Dental Education conducted and published by the ADA.¹⁴ We used the tuition/fees and stipends to estimate the net direct expenses.

	RESULTS

TOP ABSTRACT SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
Calculation of the IRR for prosthodontic training takes several factors into account, including the following:

– future earnings from the practice of prosthodontics;

– future earnings from an alternative occupation (that is, forgone earnings);

– costs of education and training;

– stipends and loans that offset direct training costs.

A majority of the respondents to the 2002 Survey of Prosthodontists reported that they were working full time or part time in private practice. Sixty-eight percent of survey respondents reported that private practice was their primary occupation, 16 percent reported their occupation as being university related and 13 percent reported having prosthodontic occupations in the government services category (military, Veterans Affairs and public health) (the remaining 3 percent reported that they were in other occupations). The remainder of this study focuses on the earnings ability and IRR for training of prosthodontists in private practice in the United States.

Table 1 shows the mean earnings of private-practice prosthodontists. The mean earnings for all full-time and part-time private-practice prosthodontists in 2001 were $213,742. The mean net income of private-practice prosthodontists was estimated to be $227,376 for the 45- to 49-year-old age group and $228,482 for the 50- to 54-year-old age group. The youngest and oldest age groups earned similar mean net incomes ($142,361 for the youngest group and $145,164 for the oldest group). The results in Table 1 indicate that mean earnings grow as prosthodontists gain more experience. Mean earnings reach a maximum before declining as prosthodontists age and choose to work fewer hours or retire.

View larger version (53K): [in this window] [in a new window]   Figure 1. Mean net income of independent-practicing prosthodontists and general practitioners, 2001. No data are provided for general practitioners younger than 35 years because the sample size was too small. Sources: American College of Prosthodontists (unpublished data, 2002) and American Dental Association.13

During the 2001–2002 academic year, 401 people were enrolled in prosthodontic training programs.¹⁴ Prosthodontic enrollment represents 8 percent of enrollment in all advanced dental educational programs in dental and non–dental school institutions. Three hundred thirty-seven (84 percent) of these 401 enrollees received their training in a university or dental school–based program.

Costs of training and stipends. Table 2 summarizes the costs of prosthodontic training in the United States and the stipends received while in training.¹⁴ In dental school–based programs, mean stipends and tuitions almost balanced out across all three years of training. The mean stipend for each year of training was approximately $14,500 for all programs (range, $14,147 to $14,698), while the mean tuition and fees for all programs were $11,065 in the first year, $10,548 in the second year and $8,036 in the final year.

View larger version (51K): [in this window] [in a new window]   Figure 2. Model for estimating the internal rate of return for completing a prosthodontic residency program.

The net earnings of the prosthodontist start on completion of the residency and occur, together with the GP’s earnings, during the 41-year period from ages 31 to 72 years. We labeled the sum of the differences in earnings at each age (that is, prosthodontist’s earnings minus GP’s earnings) as the net monetary benefit, and it represents the premium in earnings of a prosthodontist compared with those of a GP. The premium begins after completion of the residency and continues to the end of the practitioner’s working life (age 72 years), and it is a measure of the difference in the lifetime earnings between prosthodontists and GPs.

The IRR for prosthodontic training attempts to answer the following question: Is the premium earned from undertaking prosthodontic training large enough to repay all costs of training and yield a positive return to the prosthodontist?

Present value of costs. To determine if prosthodontic earnings are sufficient to justify the additional costs of training, we need to use the investment concept of present value. Present value is a means of determining the lump-sum value today of a stream of earnings that will be received during future periods. As shown in Figure 2, the direct expenses of training and the forgone earnings of a GP during a prosthodontic residency occur during a three-year period, while the lifetime earnings of a prosthodontist and a GP occur during a 41-year period. The present value of total residency costs (PVc) is given by the following equation:

The variable c_t is the annual total costs of residency and occurs for each of the three years of residency; r is the discount rate used to determine present value; is a summation symbol; and t is age. All costs are assumed to occur during ages 28, 29 and 30 years and are zero after completion of the residency training.

Present value of the premium. The present value of the premium (PVp) in earnings for a prosthodontist is given by the following equation:

The present value of the premium depends on the number of years a prosthodontist is expected to practice, the earnings as a GP (GY_t) in each period, the earnings as a prosthodontist (PY_t) in each period, the differences in earnings (PY_t – GY_t) and the discount rate r. Table 3 contains the data needed to estimate the present value of costs and the present value of the earnings premium for a prosthodontist practicing full time or part time and having received no financial assistance during the residency (we used a discount rate of 5 percent, which is equivalent to the interest rate on a 10-year riskless government bond in 2001¹⁶).

View this table: [in this window] [in a new window]   TABLE 3 ESTIMATED EARNINGS OF PROSTHODONTISTS AND GPs*, TOTAL COSTS OF RESIDENCY AND EARNINGS PREMIUM FOR PROSTHODONTISTS.

Using the regression analysis, we then estimated mean earnings at each age (31 to 72 years). Direct educational expenses are based on the combined mean tuition and fees shown in Table 2, but they do not include any financial assistance. The forgone earnings or opportunity costs of residency training are equal to a GP’s earnings in the first three years. Table 3 shows the total costs of residency training (that is, forgone earnings plus direct expenses), as well as the differences in earnings between prosthodontists and GPs after completion of residency training.

The present value of the earnings premium ($567,423) exceeds the present value of costs ($335,426) by $231,997 in present-value terms (Table 3). On the basis of our findings, after incurring all educational expenses, the prosthodontist can be expected to earn $5,700 more per year in present value terms than what he or she would be expected to earn as a GP. Our findings also show that prosthodontists can expect to earn in the first three years of practice slightly more in present value terms ($343,369) than the total costs of education ($335,426). In terms of the premium (the difference between the earnings of prosthodontists and GPs), it would take about 14.5 years for the present value of the premium to equal the present value of the educational costs.

IRR. An alternative means of comparing the earnings of prosthodontists with those of GPs is to estimate the IRR. The IRR is the discount rate that equalizes the present value of costs with the present value of the earnings premium, and it is stated as a percentage.¹⁸ We estimated the IRR to be 8.23 percent. This positive percentage means that the earnings premium is large enough to repay all costs of residency training and provide a positive return to the prosthodontist.

	DISCUSSION

TOP ABSTRACT SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
The results shown in Table 3 generally indicate that an investment in prosthodontic residency training (estimated to be $335,426 in present value terms, including direct expenses plus forgone earnings) creates an earnings premium for the prosthodontist of $567,423 (in present value terms), or an IRR of 8.23 percent. Furthermore, as Table 3 shows, various factors influence the IRR. Financial assistance in the form of stipends and loans, which was not included in Table 3, increases the IRR because it reduces the direct expenses of the residency program. Larger differences between the earnings of prosthodontists and GPs would increase the earnings premium, leading to a larger IRR. However, growth in earnings of GPs in the earliest years of practice would lead to a decrease in the IRR because of an increase in opportunity costs (that is, larger forgone earnings). Lengthening the residency training period, other factors being constant, would reduce the IRR because the costs of residency would increase and the number of years in practice would decline.

Table 4 shows various scenarios regarding time in practice, financial assistance and a delay in starting practice. Table 4 also contains estimates of the IRR, the present value of the prosthodontist’s earnings premium and the present value of prosthodontic residency costs. Stipends have been a major source of financial assistance in prosthodontic residency programs, as shown in Table 4. In 2001, these stipends generally offset the tuition costs of residency.¹⁴ Little information is available regarding the effect of educational loans on the rate of return for prosthodontic training.

Educational loans are more difficult to model in terms of estimating the IRR. Little information is available regarding how the loans are repaid, the rates at which they are repaid and forgiveness of loans. The ADA has been actively involved with a program to reduce outstanding loan debt by offering a competitive or reduced interest rate that offsets the impact of outstanding loans.¹⁹ If loans are treated similarly to stipends in that they do not have to be repaid, the estimated IRR increases to 9.71 percent (Table 4). The actual effect of loans probably lies somewhere between the 9.11 percent (estimated with stipends only) and the 9.71 percent (including stipends and loans of $8,300 per person per year).

The age-earnings estimates for prosthodontists and GPs (Table 3) are based on survey responses from both full-time and part-time practitioners. If we limit the IRR analysis to dentists in full-time practice (that is, prosthodontists and GPs working more than 32 hours per week), then we estimate the IRR to be 10.27 percent with no financial assistance (Table 4), 11.40 percent including stipends and 12.18 percent including stipends and loans. The results of the regression analysis of the age-earnings curve for full-time practicing GPs were statistically significant (Prob > F_4,590 = .0017), as were the results of the regression analysis for full-time prosthodontists (Prob > F_4,515 = .0205). Focusing estimates of the IRR on full-time practice apparently results in an increase in the difference in lifetime earnings between prosthodontists and GPs, resulting in a higher rate of return on the investment in prosthodontic training.

One-year delay in beginning practice. Finally, we estimated an IRR using the assumption of a one-year delay in starting full-time practice after completion of the residency program (Table 4) for the prosthodontist who received financial assistance. We used this assumption in the IRR calculations that assumed prosthodontists’ earnings began in the first year after residency. The effect of delaying the start of full-time practice increased the amount of forgone earnings by $114,472 in present value terms (that is, an increase in costs), decreased the first-year earnings premium by $5,341 and reduced the number of productive years by one. Under these conditions, the IRR fell to 7.94 percent; this compares with an IRR of 12.18 percent for full-time practicing prosthodontists who received stipends and loans (Table 4). Even if we ignore the increase in forgone earnings and the decrease in the earnings premium, the IRR still falls to 10.56 percent because of one less year of practice.

Although this is an extreme example, it exemplifies the importance of length of residency, costs of residency and the earnings differences during the early years of practice. Significant increases in the cost of residency training or considerable reductions in the earnings of prosthodontists during the early years of practice will significantly lower the IRR. Increasing the number of years in residency, reducing the number of years in practice at younger ages or both will lower the IRR and likely result in a negative effect on the number of applications to residency programs.

Study assumptions. The estimates of the IRR presented in this report are sensitive to at least three assumptions:

– the cross-sectional earnings of prosthodontists and GPs of different ages can be used to estimate expected lifetime earnings;

– the expected earnings at each age for both prosthodontists and GPs can be determined using statistical regression analysis;

– educational loans are part of the financial assistance received by residents.

We used samples of prosthodontists and GPs in 2001 to estimate lifetime earnings even though practitioners were educated before 2001 (for some practitioners, many years earlier). For example, we assumed that the best estimate of earnings at age 55 years in a lifetime earnings profile is the expected earnings of 55-year-old practitioners in 2001. We used this approach because the preferred data—from a panel of prosthodontists followed from residency to the end of their work lives—were not available.

We also assumed that regression analysis is a reasonable method for estimating the expected earnings at each age (Table 3) when the sample size varies at each age. We did not use the alternative approach, which would be to estimate the mean earnings of a sample of practitioners at each age, because the sample size was not sufficient for every age group. Furthermore, we made a simplifying assumption regarding educational loans (that is, they are treated similarly to stipends in that they do not have to be repaid); further research is needed in this area. Finally, we assumed that all residency costs were incurred during the first three years and economic returns were created during the remaining years of the work life cycle.

The study results also show that estimates of the IRR are very sensitive to the level of educational costs (net of financial assistance), the length of residency and the level of earnings during the early years of practice.

	CONCLUSION

TOP ABSTRACT SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
In this study, we used the IRR to evaluate whether the expected returns from investing in prosthodontic training are worth the direct costs of training plus the forgone earnings as a GP. One indicator of the economic health of prosthodontists is the finding that independent prosthodontists earned a mean of $229,900 in 2001, which is 33 percent higher than the mean net income of independent GPs.

However, comparisons based only on mean earnings do not reflect all of the important elements of an individual’s decision to become a practicing prosthodontist. Other important factors include future earnings of prosthodontists, costs of education, forgone earnings in the next-best alternative occupation and time spent in education. We estimated the IRR for prosthodontic training to range from 8.23 percent for private practitioners with no financial assistance to 12.18 percent for full-time private practitioners with stipends and loans during residency.

The IRR varies depending on such factors as the amount of financial assistance, the time spent in private practice and how soon practice begins after completion of the residency program. Estimates of the IRR for undertaking a prosthodontic residency indicate that lifetime earnings after completion of training are large enough to cover all costs of education and provide a positive return to the prosthodontist. A positive IRR for prosthodontic training signals a likely continuation of the demand for advanced education in prosthodontics.