DEBATES ON MEDICAL MALpractice are recurrent themes. Tor reformers emphasize the danger that liability crises pose to cost and availability for medical services, while tort defenders stress the importance of liability to medical care quality. Physicians are said to be more likely to move to other areas, avoid high-risk patients, or leave the practice altogether when there is uncertainty about their coverage. However, even in such crises, the malpractice climate is still a major factor in determining whether or not physicians can enter the medical profession and choose which specialties and where they practice. We also examined how long-term trends of physician supply change from one state to another. 

The American Medical Association's Physician MasterFile provided data on the number and practice of physicians in each state from 1985 to 2001. The number of physicians in a given state at any point in time was modeled as a function of state fixed effects, state fixed effects, and time-variable state characteristics. We also considered the presence or absence of certain malpractice reforms. 

The impact of malpractice climate on the physician workforce is one of many factors. However, it was relatively small in our study. The overall supply as well as the geographic distributions and specialty of doctors can be altered at various junctures, including initial career choice, retraining, relocation, retirement, and retirement. Rizzo and Blumenthal24 have found that an increase of 1% in physician wages would result in a 0.2% to 0.3% rise in the hours worked. The 3.3% increase in physician supply due to direct malpractice reform is approximately equivalent to the 11% increase in labor supply (11=3.3/0.3 span> ). 

Our research reveals the mechanisms that malpractice liability can reduce physician supply growth. The relationship between malpractice climates and physician supply is affected by the characteristics of physicians such as their practice structure, specialty, payer mix, stage in career, and other factors. Our study found that the direct reforms had a greater effect on physicians who work in nongroup settings. This was due to the increased number of doctors moving into nonreform states. This is due to the lower ability of smaller practices, which are less able to spread liability insurance costs among many doctors, cushion premium volatility through high patient volumes, or share risk alongside hospitals and other health care institutions. 

Malpractice insurance prices are determined by location and specialty, rather than individual doctor quality or loss experience.25-26 This would explain why there is a greater supply effect in the specialties that pay the highest malpractice premiums. Point estimates showed that reforms had a larger-than-average impact on the supply of physicians in three of five specialties with the highest malpractice premiums. Although statistically not distinguishable from the average effect, the effect of reforms on the supply of obstetrics, gynecology, and general surgeon practitioners were smaller than the average. 

Malpractice reform has a wider impact on the organization of physician services than just increasing supply. The differential response of the supply of nongroup doctors to reform appears to be due to nongroup physicians becoming group physicians in states that have not implemented reform. In other words, liability pressure is one contributing factor to the growing corporatism in US medicine. Indirect reforms also increased the supply growth for more experienced doctors but decreased that of less experienced ones, according to our research. This would be because more experienced doctors valued indirect reforms higher than their less experienced counterparts. The decline in earnings and partnership opportunities that are associated with more experienced physicians in states that have indirect reforms discourage the entry of new graduates. These effects require further investigation. 

Endogeneity bias could have caused us to either underestimate or overstate the impact of reforms. Endogeneity bias could lead to us understating the impact of reforms if there is a decrease in physician supply. Endogeneity bias, on the other hand, would cause us to underestimate the impact of reforms if both increases in physician supply or adoption of reforms are caused by unobserved factors such as population preferences, for medical services and litigation. Fourth, we only estimated the effect of law reforms upon physician supply. Our results are stronger but less detailed because we didn't make any assumptions about the impact of malpractice pressure on physician supply. 

We cannot rule out the possibility that an increase in physician supply observed in states that adopted reforms during our study period is simply due to states with more growth potential, as those states had fewer doctors at baseline. We did not control for variations in supply levels at baseline, but we did adjust for changes in supply growth rates at baseline. Endogeneity bias is the third. 

Direct reforms have greater positive effects on physician supply in high- and low-managed care states. Physicians may feel more disutilized by managed care and malpractice pressure when they are combined, but we can't determine which aspect. Because managed care enrollment has increased over the past decade, the effects of malpractice reform in high managed states might be better than those in low managed states. 

Our study found that direct reforms had a stronger effect on physician mobility between states and retirees. This supports the argument that direct reforms' supply effects will continue, at least in part, even if all states implement reforms. If tort reform, such as California's Medical Injury Compensation Reform Act, has persistent supply effects then our study will not distinguish between reform- and non-reform states. However, future supply effects may be more accurate if the first wave of reforms has seen greater supply gains. 

We only estimated the effect of law reforms upon physician supply. Our results are stronger but less detailed because we didn't make any assumptions about the impact of malpractice pressure on insurance premiums, awards, and frequency, or the number of awards. Fifth, we only considered the effect of law reforms on the number and not the hours worked by physicians. If the number of doctors decreases the hours per physician, our estimates may underestimate the total impact of reforms.