Physician Labor Supply

Labor supply has been a well-studied topic in the labor economics literature (Killingsworth, 1983; Pencavel, 1986; Killingsworth and Heckman, 1986; Blundell and MaCurdy, 1999). As the key clinical decision-makers, physicians provide an essential input into the production of health care for their patients. Understanding the determinants of physician labor supply has important implications for the production and cost of care and for health care access. Hence, physician labor supply is a topic of considerable interest in health economics as well.

This article examines the factors affecting physician labor supply. Labor supply measures are considered broadly, and may include annual hours worked, numbers of surgeries performed, office visits, and so on. (Strictly speaking, physician labor supply curves must include earnings as a determinant of labor supply. Many studies, however, focus on other aspects affecting labor supply, such as the competitive conditions in the market. These studies typically do not include earnings as a separate factor affecting labor supply. A likely reason is that both measures of competition and earnings would have to be treated as endogenous, prohibitively challenging data estimation issues. However, these studies are included because from a policy perspective, it is important to understand a variety of factors that may affect the physician’s decision to offer services.) However, the related topic of aggregate forecasts of physician manpower is not addressed. The authors begin with a general framework discussing the competing goals of the physician in choosing labor supply and then review studies that have examined a number of key issues affecting the labor supply decision.

This article is organized into seven parts. The Section Introduction discusses the conceptual issues in the physician labor supply decision. The Section Conceptual Issues considers studies of the relationship between physician earnings and labor supply. The effects of competition and physician fee schedules on labor supply are described in the Section Earnings and Labor Supply. This section also reviews labor supply responses as possible evidence of demand inducement. The Section Competition, Fee Schedules, and Labor Supply examines the roles of the so-called ‘target income hypothesis,’ and more recently reference incomes as they pertain to the labor supply decision. Physician labor supply under managed care is discussed in the Section Income Targets, Reference Incomes, and Labor Supply and the effect of malpractice liability on labor supply is examined in the Section Managed Care and Labor Supply. The Section Malpractice Liability and Labor Supply concludes this article.

Conceptual Issues

The physician is assumed to maximize his utility with respect to income and leisure, both of which are ‘goods’. Income includes labor income and nonlabor income. In altruistic variants of this basic framework, patients’ health enters as an argument into the utility function as well. (In altruistic models, the physician plays the role of agent for his patients, with patients’ health entering as an argument into the physician’s utility function. As the physician has competing objectives including the desire to obtain more income and leisure, he may be an imperfect agent.) The physician chooses labor supply so as to maximize this utility function. A complex constellation of factors, including earnings potential, altruistic goals, competitive conditions, incentives provided by insurers, the regulatory environment, and the threat of malpractice litigation may affect this decision.

A long and controversial debate has centered on the notion that income targets affect the provision of physician services. The so-called target income hypothesis asserts that physicians set income targets and, when their actual incomes fall below these targets, will increase the volume of their services to offset, in whole or in part, their perceived income shortfall. In critiquing this hypothesis, McGuire and Pauly (1991) have argued that there is no conceptual basis as to how income targets are set. Moreover, they present a theoretical model which demonstrates that increasing volume of services in response to fee restrictions need not depend on income targets, but may occur in the context of a standard model of profit maximization, provided that income effects from fee reductions are sufficiently strong.

Although the standard model underlying the physician’s labor supply decision is neoclassical, more recent treatments that have considered the role of target or reference incomes on labor supply have adapted models from prospect theory (Kahneman and Tversky, 1979). Borrowing from prospect theory, these models incorporate the notion that physicians set reference incomes and compare these benchmarks to their actual earnings in deciding on whether to adjust their labor supply (Rizzo and Zeckhauser, 2003, 2007). These models also incorporate the notion of loss aversion (Tversky and Kahneman, 1991; Goette et al., 2004), a phenomenon in which individuals strongly prefer avoiding losses to equivalent-sized gains. Loss version posits a kink in the physician’s utility curve for different marginal utilities of income below and above this reference income. When the physician’s actual income is less than the reference income, the marginal utility of income is very steep in that range. The relevance of reference points and loss aversion has been well-established in the experimental psychology literature (Rabin, 1998; Heath et al., 1999; Schmidt and Traub, 2002; Fellner and Maciejovsky, 2007; Rizzo and Zeckhauser, 2003, 2007). Their inclusion in models of physician labor supply provides an empirically validated framework for understanding the relevance of reference of target or reference incomes for physician labor supply.

Earnings And Labor Supply

Perhaps the most salient factor affecting the physician’s decision to supply labor is the economic return that labor will generate. Hence, a number of studies have examined this relationship empirically. Changes in physician earnings may exert both income and substitution effects on the labor supply decision. If the substitution effect dominates, an increase in earnings will lead to an increase in labor supply. At sufficiently high earnings levels, however, the income effect may dominate, in which case one would observe a backward-bending labor supply. Early studies examined the effects of nonlabor income on physician labor supply by focusing on exogenous variations in nonpractice income, typically finding insignificant nonlabor income effects (Sloan, 1974; Hurdle and Pope, 1989).

In terms of labor income, however, most physicians are not paid by wages or salary; hence, their hourly earnings are endogenous. Sloan (1975) examined the relationship between hourly earnings and hours worked employing a two-stage estimation treating earnings as endogenous. The effects of earnings on labor supply varied somewhat according to specification, but were generally modest and in some cases statistically insignificant. Using a nationally representative database of young physicians under the age of 40 years, Rizzo and Blumenthal (1994) estimated a model of physician labor supply, treating physician earnings as endogenous and employing two-stage least-squares estimation. They obtain separate estimates of the income and substitution effects of a change in hourly physician earnings for male physicians, finding a significantly negative income effect with an elasticity of 0.26, and a significantly positive substitution effect elasticity of 0.49. The total effect of wage increase is to raise labor supply, with an elasticity in the range of 0.2–0.3. Small sample size precludes obtaining separate estimates for female physicians. A more recent study from the UK also reports a modest positive association between earnings and physician labor supply, with elasticities in the range of 0.09–0.12 (Ikenwilo and Scott, 2007).

Brown (1994) contrasts the use of aggregate versus physician-level data in estimating the earnings/labor supply relationship. Using aggregated data he finds no effect (Brown and Lapan, 1972; Brown et al., 1974), but a negative relationship using physician-level data, with an elasticity of 0.2. He argues that physician-level data are preferable for estimating labor supply. Bradford and Martin (1995) also find evidence of a backward-bending labor supply curve. Thornton and Eakin (1997) estimate a model of a utility-maximizing solo practitioner. They find that an increase in nonlabor income will lead solo practitioners to allocate fewer hours to medical practice activities. In addition, they also report both income effect and substitution effects of labor income changes. Consistent with earlier research, they find that the net effect of physician service fee reductions leads physicians to reduce their labor supply, so the substitution effect dominates the income effect.

Competition, Fee Schedules, And Labor Supply

Studies relating competition and fee schedules to physician labor supply have typically sought to provide evidence of demand-inducing behavior by physicians. If physicians increase their supply of services in response to greater competitive pressures or reductions in fees, then it is taken as indirect evidence of demand-inducing behavior. In fact, even if such relationships exist, it is unclear whether these behaviors reflect demand inducement. One is on firmer ground simply interpreting them as labor supply responses to changing financial incentives.

Competition And Labor Supply

Studies of the effect of competition on labor supply have employed metrics for competition such as per capita physicians in a market area, relating this measure to various types of physician services. An early study by Fuchs (1978) examines the supply of surgeons on the number of surgeries performed, treating physician supply as endogenous, with variables measuring the appeal of a market (e.g., urban, hotel receipts) serving as instruments. The results suggest that increased competitive pressure as measured by physician supply leads to an increase in surgeries, with an elasticity of 0.3. Subsequent studies employing a similar strategy but with more complete control variables produce similar results (Rossiter and Wilensky, 1983, 1984; Cromwell and Mitchell, 1986; Birch, 1988; Grytten et al., 1990; Scott and Shiell, 1997; Baltagi et al., 2005).

A limitation with these studies is the choice of instruments. Employing an instrumental variable approach typical in these studies, Dranove and Wehner (1994) produce the seemingly bizarre result that greater competition among obstetrician/ gynecologists (OBGYNs) increases childbirths. As it is highly unlikely that physicians induce demand for children, this result is taken as evidence that the instrumental variable approach employed in these studies is suspected. In fact, the likely explanation for their result reflects border crossing, for example, that pregnant women travel to locations where there are ample supplies of OBGYNs. Dranove and Wehner (1994) calls into question any causal interpretation between competition and labor supply or demand inducement using the instrumental variable strategy described earlier.

Gruber and Owings (1996) employ an alternative approach to address the endogeneity problem. They study the relationship between physician financial incentives and cesarean-section delivery. Between state and intertemporal variations in fertility rates are used as exogenous measures of competitive pressures facing OBGYNs. They hypothesize that, in response to declining fertility in the US, OBGYNs will substitute vaginal childbirths for the more lucrative and physician-intensive cesarean deliveries. Using nationally representative data from the period of 1970–82, they find that a 10% drop in fertility leads to a 0.6% increase in cesarean sections.

Fang and Rizzo (2009) argue that the relationship between competition and physician labor supply depends on the nature of third-party reimbursement. Using data from the Community Tracking Study Physician Survey 2000–01, they find that physician volume increases with more competition under fee-for-service reimbursement, but decreases with greater competition under managed care.

Fee Schedules And Labor Supply

A number of studies have examined physician labor supply responses to fee restrictions (Lee and Hadley, 1981; Rice, 1983; Mitchell et al., 1989; Hurley et al., 1990; Hurley and Labelle, 1995; Escarce, 1993, Rochaix, 1993; Nguyen, 1996; Nguyen and Derrick, 1997; Yip, 1998; Tai-Seale et al., 1998; Gruber et al., 1999; Mitchell et al., 2000; Kantarevic et al., 2008). In contrast to the competition studies, fee reductions may be considered exogenous to the individual physician. A number of these studies have found evidence of a volume offset effect; that is, physicians respond to real declines in their fees by increasing the volume of their services. However, the volume increase is not sufficient to fully recoup the income losses from the fee cuts.

Thus, Nguyen (1996) studies physician volume responses to reductions in the Medicare fee schedule reduction. The results indicate that physicians will increase their service volume by 3.7% in response to a 10% reduction in the Medicare fee schedule. Nguyen and Derrick (1997) examine the impact of Medicare fee cuts for certain ‘overpriced procedures,’ finding that for physicians who experience the largest fee reductions, a 10% decline in price lead to a 4% increase in volume. In a comprehensive assessment of volume offset effects for multiple specialties and payers, Tai-Seale et al. (1998) find that the point estimates for volume responses to fee restrictions varied across specialties, but these responses were statistically insignificant in most cases. Yip (1998) studies the effects of reductions in fee schedules on the volume of coronary artery bypass grafting (CABG) procedure using a longitudinal panel of physicians in New York and Washington States. For physicians in both Medicare and private markets, she finds that physicians whose incomes are cut by reduced fee schedules exhibit a large volume response, recouping 70% of their lost income. Gruber et al. (1999) investigate the effect of Medicaid fee differentials on cesarean delivery. They find that cesarean delivery for Medicaid patients is significantly less likely than for privately insured patients, reflecting that the fee differentials between cesarean and vaginal deliveries are smaller under the Medicaid program than the private insurance. Mitchell et al. (2000) analyze physician labor supply responses to Medicare fee reductions during the period of 1991–94 using pooled cross-sectional time series data. They focus on ophthalmologists and orthopedic surgeons performing cataract extractions and major joint repair/replacement procedures, respectively. In contrast to most previous research, they find that fee reductions are associated with fewer surgeries.

Income Targets, Reference Incomes, And Labor Supply

The target income hypothesis asserts that physicians set income targets and will attempt to reach or get closer to this target by increasing their services in response to increased competitive pressures or cutbacks in their fees. Most studies of this issue have been indirect, relating measures of competition and fee cuts to the volume of physician services. Such studies have already been addressed in Section Earnings and Labor Supply.

Few studies have related direct measures of income targets to physician labor supply. Rizzo and Zeckhauser (2003) use a unique panel data of physicians under the age of 40 years that includes a physician-specific measure of target or reference income (in particular, physicians were asked: ‘‘Considering your career stage, what do you consider to be an adequate income after expenses but before taxes from your professional activities?’’ The response to this question was taken as the physician’s reference income (Rizzo and Zeckhauser, 2003, 2007)). They find that incomes increase substantially in response to higher reference incomes for physicians who are below their reference incomes, but not for those who are at or above. However, they also note that physicians appear to raise their incomes, not by increasing labor supply as measured by hours worked, but by performing activities that generate a higher hourly return (e.g., performing more lucrative services). A subsequent study also finds no evidence that hours worked respond to reference incomes (Rizzo and Zeckhauser, 2007).

Managed Care And Labor Supply

Managed care has grown rapidly in the US since 1980 and is the dominant form of health insurance (Robinson, 1999). Intended to control the rapid growth in health care costs in the US, managed care may exert strong effects on physician practice patterns, including labor supply, because the physician’s decision-making process is likely to respond to the financial incentives and restrictions created by managed care.

Hirth and Chernew (1999) discuss two fundamentally different labor market regimes for physicians: fee-for-services and managed care, noting that physicians practicing in managed care environments are less likely to enhance their income by providing more services compared with those mainly reimbursed on the basis of fee-for-service. Libby and Thurston (2001) examine the impact of managed care contracting on physician labor supply by extending the standard labor supply model to incorporate managed care incentives. They find that managed care contracting generally reduces the number of hours that physicians practice, but the net effects become small and insignificant after accounting for the endogeneity of physician managed care contracting behavior.

Malpractice Liability And Labor Supply

Physician labor supply may also be affected by other factors such as the threat of malpractice liability. Thornton (1997) shows a significant income effect of a change in malpractice premiums on physician labor supply. In particular, higher malpractice premiums lead primary care physicians to increase their practice hours, possibly to recoup some of the income losses associated with these premiums. An alternative interpretation is that physicians regard malpractice premiums as a tort signal and attempt to work more hours to reduce the possibility of malpractice liability. Either effect leads physicians to increase labor supply. Thornton (1999) compares the magnitudes of the income effect and tort signal effect in response to malpractice premiums. He finds that the income effect dominates, with the tort effect being much smaller.

In contrast, Helland and Showalter (2009) analyze the effect of state-level malpractice reforms during the period of 1983–88 on physician behavior. They find that an increase of 1% in the probability of incurring a malpractice suit will reduce the weekly hours worked by 0.29%. The magnitude of this elasticity increases to 1.224 for physicians aged 55 years or older. Kessler and McClellan (1996) analyze malpractice reforms designed to reduce the threat if physician liability to examine whether physicians practice defensive medicine; for example, increasing the provision of services to ward off the threat of malpractice suits. They find that malpractice reforms are associated with a significant reduction (5–9%) in physician expenditures for Medicare patients after the malpractice reforms without substantial changes in mortality or medical complications.

Conclusion

As key players in medical decision-making, understanding factors affecting the physician’s decision to supply care will remain an important topic in health economics research. Much effort has been devoted to this issue already and with fruitful results. Most studies suggest that greater earnings potential and fee restrictions both lead to increased labor supply, though the response is typically fairly small and inelastic. Greater competition also appears to increase labor supply, although econometric challenges associated with much of this literature suggest that these results should be viewed with more caution. Perhaps not surprisingly, managed care has had a restrictive role on physician labor supply. Less certain are the effects of target or reference incomes and the role of medical malpractice liability.

Although the notion that reference points are used in decision-making has considerable empirical support (Rizzo and Blumenthal, 1996; Rabin, 1998; Heath et al., 1999; Schmidt and Traub, 2002; Fellner and Maciejovsky, 2007; Rizzo and Zeckhauser, 2003, 2007), McGuire and Pauly (1991) rightly assert that there remains no theory as to how and why physician income targets are set and why they should affect the physician labor supply decision. Understanding these issues is an important direction for further research.

Increasingly, patients are playing a more proactive role in the care they receive from physicians (Fang et al., 2008). This consumerist orientation may have implications for physician labor supply as well. Fang and Rizzo (2009) introduce the notion of ‘physician demand enablement’ as physician labor supply responses to patient-initiated requests for services. The effects of ‘consumerist’ patients on physician’s willingness to supply care also warrant further study.

References:

1. Baltagi, B. H., Bratberg, E. and Holmas, T. H. (2005). A panel data study of physicians’ labor supply: The case of Norway. Health Economics 14(10), 1035–1045.
2. Birch, S. (1988). The identification of supplier-inducement in a fixed price system of health care provision: The case of dentistry in the United Kingdom. Journal of Health Economics 7(2), 129–150.
3. Blundell, R. and MaCurdy, T. (1999). Labor supply: A review of alternative approaches. In Ashenfelter, O. C. and Card, D. (eds.) Handbook of labor economics, vol. 3, pp. 1559–1695. North-Holland: Elsevier.
4. Bradford, D. and Martin, R. E. (1995). Supplier-induced demand and quality competition: An empirical investigation. Eastern Economic Journal 21(4), 491–503.
5. Brown, D. M. (1994). The rising price of physicians’ services: A correction and extension on supply. Review of Economics and Statistics 76(2), 389–393.
6. Brown, D. M., Feldstein, M. and Lapan, H. E. (1974). The rising price of physicians’ services: A clarification. Review of Economics and Statistics 56(3), 396–398.
7. Brown, D. M. and Lapan, H. E. (1972). The rising price of physicians’ services: A comment. Review of Economics and Statistics 54(1), 101–105.
8. Cromwell, J. and Mitchell, J. B. (1986). Physician-induced demand for surgery. Journal of Health Economics 5(4), 293–313.
9. Dranove, D. and Wehner, P. (1994). Physician-induced demand for childbirths. Journal of Health Economics 13(1), 61–73.
10. Escarce, J. J. (1993). Effects of lower surgical fees on the use of physician services under Medicare. Journal of American Medical Association 269(19), 2513–2518.
11. Fang, H., Miller, N. H., Rizzo, J. A., Zeckhauser, R. J. (2008). Demanding customers: Consumerist patients and quality of care. NBER Working Paper W14350. Cambridge, MA: National Bureau of Economic Research, Inc.
12. Fang, H. and Rizzo, J. A. (2009). Competition and physician-enabled demand: The role of managed care. Journal of Economic Behavior and Organization 72(1), 463–474.
13. Fellner, G. and Maciejovsky, B. (2007). Risk attitude and market behavior: Evidence from experimental asset markets. Journal of Economic Psychology 28(3), 338–350.
14. Fuchs, V. R. (1978). The supply of surgeons and the demand for operations. Journal of Human Resources 13(supplement), 35–56.
15. Goette, L., Huffman, D. and Fehr, E. (2004). Loss aversion and labor supply. Journal of the European Economic Association 2(2–3), 216–228.
16. Gruber, J., Kim, J. and Mayzlin, D. (1999). Physician fees and procedure intensity: The case of cesarean delivery. Journal of Health Economics 18(4), 473–490.
17. Gruber, J. and Owings, M. (1996). Physician financial incentives and cesarean section delivery. Rand Journal of Economics 27(1), 99–123.
18. Grytten, J., Holst, D. and Laake, P. (1990). Supplier inducement: Its effect on dental services in Norway. Journal of Health Economics 9(4), 483–491.
19. Heath, C., Larrick, R. P. and Wu, G. (1999). Goals as reference points. Cognitive Psychology 38(1), 79–109.
20. Helland, E. and Showalter, M. H. (2009). The impact of liability on the physician labor market. Journal of Law and Economics 52(4), 635–663.
21. Hirth, R. A. and Chernew, M. E. (1999). The physician labor market in a managed care-dominated environment. Economic inquiry 37(2), 282–294.
22. Hurdle, S. and Pope, G. C. (1989). Physician productivity: Trends and determinants. Inquiry 26(1), 100–115.
23. Hurley, J. and Labelle, R. (1995). Relative fees and the utilization of physicians’ services in Canada. Health Economics 4(6), 419–438.
24. Hurley, J., Labelle, R. and Rice, T. (1990). The relationship between physician fees and the utilization of medical services, in Ontario. In Scheffler, R. and Rossiter, L. (eds.) Advanced in health economics and health services research, pp. 49–78. Greenwich, CT: JAI Press.
25. Ikenwilo, D. and Scott, A. (2007). The effects of pay and job satisfaction on the labour supply of hospital consultants. Health Economics 16(12), 1303–1318.
26. Kahneman, D. and Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica 47(2), 263–291.
27. Kantarevic, J., Kralj, B. and Weinkauf, D. (2008). Income effects and physician labour supply: Evidence from the threshold system in Ontario. Canadian Journal of Economics 41(4), 1262–1284.
28. Kessler, D. and McClellan, M. (1996). Do doctors practice defensive medicine? Quarterly Journal of Economics 111(2), 353–390.
29. Killingsworth, M. R. and Heckman, J. J. (1986). Female labor supply: A survey. In Ashenfelter, O. C. and Layard, R. (eds.) Handbook of labor economics, vol. 1, pp. 103–204. North-Holland: Elsevier.
30. Killingsworth, M. R. (1983). Labor supply. Cambridge: Cambridge University Press.
31. Lee, R. H. and Hadley, J. (1981). Physicians’ fees and public medical care programs. Health Service Research 16(2), 185–203.
32. Libby, A. M. and Thurston, N. K. (2001). Effects of managed care contracting on physician labor supply. International Journal of Health Care Finance and Economics 1(2), 139–157.
33. McGuire, T. G. and Pauly, M. V. (1991). Physician response to fee changes with multiple payers. Journal of Health Economics 10(4), 385–410.
34. Mitchell, J. B., Wedig, G. and Cromwell, J. (1989). The Medicare physician fee freeze: What really happened? Health Affairs 8(1), 21–33.
35. Mitchell, J. M., Hadley, J. and Gaskin, D. J. (2000). Physicians’ responses to Medicare fee schedule reductions. Medical Care 38(10), 1029–1039.
36. Nguyen, N. X. (1996). Physician volume response to price controls. Health Policy 35(2), 189–204.
37. Nguyen, N. X. and Derrick, F. W. (1997). Physician behavioral response to a Medicare price reduction. Health Services Research 32(3), 283–298.
38. Pencavel, J. (1986). Labor supply of men: A survey. In Ashenfelter, O. C. and Layard, R. (eds.) Handbook of labor economics, vol. 1, pp. 3–102. North-Holland: Elsevier.
39. Rabin, M. (1998). Psychology and economics. Journal of Economic Literature 36(1), 11–46.
40. Rice, T. H. (1983). The impact of changing Medicare reimbursement rates on physician-induced demand. Medical Care 21(8), 803–815.
41. Rizzo, J. A. and Blumenthal, D. (1994). Physician labor supply: Do income effects matter? Journal of Health Economics 13(4), 433–453.
42. Rizzo, J. A. and Blumenthal, D. (1996). Is the target income hypothesis an economic heresy? Medical Care Research and Review 53(3), 243–266.
43. Rizzo, J. A. and Zeckhauser, R. J. (2003). Reference incomes, loss aversion, and physician behavior. Review of Economics and Statistics 85(4), 909–922.
44. Rizzo, J. A. and Zeckhauser, R. J. (2007). Pushing incomes to reference points: Why do male doctors earn more? Journal of Economic Behavior and Organization 63(3), 514–536.
45. Robinson, J. C. (1999). The future of managed care organization. Health Affairs 18(2), 7–24.
46. Rochaix, L. (1993). Financial incentives for physicians: The Quebec experience. Health Economics 2(2), 163–176.
47. Rossiter, L. F. and Wilensky, G. R. (1983). A reexamination of the use of physician services: The role of physician-initiated demand. Inquiry 20(2), 162–172.
48. Rossiter, L. F. and Wilensky, G. R. (1984). Identification of physician-induced demand. Journal of Human Resources 19(2), 231–244.
49. Schmidt, U. and Traub, S. (2002). An experimental test of loss aversion. Journal of Risk and Uncertainty 25(3), 233–249.
50. Scott, A. and Shiell, A. (1997). Analyzing the effect of competition on general practitioners’ behaviour using a multilevel modeling framework. Health economics 6(6), 577–588.
51. Sloan, F. A. (1974). A microanalysis of physicians’ hours of work decisions. In Perlman, M. (ed.) The economics of health and medical care, pp. 302–325. New York, NY: John Wiley and Sons.
52. Sloan, F. A. (1975). Physician supply behavior in the short run. Industrial and Labor Relations Reviews 28(4), 549–569.
53. Tai-Seale, M., Rice, T. H. and Stearns, S. C. (1998). Volume responses to Medicare payment reductions with multiple payers: A test of the McGuire–Pauly model. Health Economics 7(3), 199–219.
54. Thornton, J. (1997). Are malpractice insurance premiums a tort signal that influence physician hours worked? Economics Letters 55(3), 403–407.
55. Thornton, J. (1999). The impact of medical malpractice insurance cost on physician behaviour: The role of income and tort signal effects. Applied Economics 31(7), 779–794.
56. Thornton, J. and Eakin, K. (1997). The utility-maximizing self-employed physician. Journal of Human Resources 32(1), 98–128.
57. Tversky, A. and Kahneman, D. (1991). Loss aversion in riskless choice: A reference-dependent model. Quarterly Journal of Economics 106(4), 1039–1061.
58. Yip, W. C. (1998). Physician response to Medicare fee reductions: Changes in the volume of coronary artery bypass graft (CABG) surgeries in the Medicare and private sectors. Journal of Health Economics 17(6), 675–699.