Charles Silver


In 2003, the Texas legislature adopted a comprehensive package of tort reforms designed to greatly reduce the frequency of medical malpractice lawsuits, the size of malpractice payments, and physicians’ insurance premiums. Supporters of the reforms contended that if patients’ rights were slashed, access to health care would greatly increase.

Soon after the amendments passed, tort reform groups began trumpeting their success, claiming that doctors are moving to Texas in droves. The Texas Medical Association asserts that “[f]rom May 2003 through July 2005,… more than 3,000 new doctors established practice in Texas.” Dr. Howard Marcus, Chairman of the Texas Alliance For Patient Access—a coalition of doctors, hospitals, nursing homes, health care providers and medical liability insurers that might be more aptly named the “Texas Alliance for Access to Patients”—contends that “[t]he physician growth rate is staggering” and predicts “a record 4,000 new Texas physicians will be licensed” in 2006.

These assertions should be met with skepticism. The annual change in the supply of Texas physicians reflects two factors: the entry of new doctors and the departure of old ones. Because the assertions quoted above ignore physician departures, which occur in large numbers every year, the picture they convey is incomplete. To determine whether the 2003 tort reforms actually caused physician supply to increase, one must examine the total physician population and the net rate at which it changed. When one does, one sees that the 2003 reforms have not improved access to care. In 2004 and 2005, Texas’ doctor population grew more slowly than it did from 1990 to 2002.

Information on the number of Direct Patient Care Physicians practicing in Texas can easily be obtained from the Texas Dept. of State Health Services. Figure 1 shows the total size of this population from 1990 to 2005. The population increased steadily across the period, as can readily be seen. Equally apparent, the rate of increase from 2003 to 2005 is not especially sharp.


Figure 2 shows the annual percentage increase in physician supply and the 2-year percentage increase for the same period. Again, it is obvious that the increases in 2004 and 2005 were nothing to shout about. The population of Texas physicians actually experienced its most dramatic growth from 1997 to 2000, years during which Texas was supposedly building toward or experiencing a liability crisis of epic proportions.


In fact, growth rates for the post-reform years were below the Texas norm, as Table 1 makes clear. From 1990-2002, the number of physicians practicing in the state grew at an average rate of 3.21% per year and 6.71% every two years. In 2004 and 2005, supply grew more slowly, averaging only 1.98% per year and 4.74% over two years. Taking account of exits as well as entries, it is clear that Dr. Marcus’ assertion of “staggering” growth in the post-reform era is false.


It is true that applications for licenses to practice medicine in Texas hit an all-time high in 2006, as Dr. Marcus contends. The Spring 2006 Bulletin of the Texas Medical Board (TMB) appears to be the source for his prediction of 4,000 new licensees. Yet, the number of licenses granted fell considerably short of 4,000, as the Fall 2006 TMB Bulletin reveals. It even fell short of the number of licenses granted in FY2005. In other words, in FY2006 the number of new licenses issued actually declined. The cause appears to be a shortage of trained personnel at the TMB, and a resulting backlog of applications. Evidently, the Texas legislature can’t make up its mind whether it wants more physicians or not. It slashed patients’ legal rights, but then denied them better access to care by under-funding the TMB.


It seems likely that tort reform has increased or eventually will increase access to physicians in rural areas. In an unpublished study, David Matsa, a health economist at Northwestern University, finds such an effect nationwide. The increase occurs in rural counties because in these areas demand for health care is unusually sensitive to price. The effect should be small, however, because malpractice premiums constitute a small and declining fraction of physicians’ real operating costs.
By contrast, that impact of the 2003 reforms on Texas patients’ rights has been enormous, judging from early reports. The 2005 Annual Report of the Medical Liability Benefit Plan for the University of Texas System describes the recent history of medical malpractice lawsuits against all U.T. health care institutions. From 2002 to 2005, new lawsuits fell 55%, from 60 to 27. New claims fell 36%, from 163 to 105. Open claims and lawsuits fell 28%, from 354 to 256. All three drops would be even bigger if one used 2003 as the base year. The recent dramatic increase in the profitability of malpractice insurers and the related declines in doctors’ liability premiums were purchased at the expense of injured patients.

In sum, the 2003 Texas reforms transferred a lot of wealth from malpractice victims, their families, employers, and health insurers to physicians and their liability insurers. This was the main object of the 2003 reforms, and the reforms achieved it. But the reforms have not increased physician supply, which grew at a sub-par rate in the post-reform years. The facts show yet again how wealthy and concentrated interest groups use the political process to advantage, at the expense of groups whose members are anonymous and dispersed.

Charles Silver: Author Bio | Other Posts
Posted at 9:44 AM, Nov 02, 2006 in
Permalink | Email to Friend


Professor Silver, I find it curious that in all of your other reports on the Texas malpractice database, you measure data per capita and criticize those who do not, but when it comes to raw numbers of doctors, you ignore population growth.

You mention a backlog: what's the lag time in approval of medical applications? Isn't it curious that the applications went way up from 2003 to 2004 and 2005 to 2006 (and were stable from 2004 to 2005), but approvals went up from 2004 to 2005 while going down from 2003 to 2004 and 2005 to 2006? If there's a one-year lag (not terribly inconsistent with my experience with state bar application approvals), that explains the discrepancy, plus one can expect the number of doctors to go up substantially in 2007, which changes the outlook of your data considerably.

In sum, the 2003 Texas reforms transferred a lot of wealth from malpractice victims, their families, employers, and health insurers to physicians and their liability insurers.

You left out three parties in the wealth transfer there: attorneys and uninjured patients and their employers. The median Texan came out ahead.

Posted by: Ted | November 2, 2006 2:01 PM

A "text only" version of my reply to Ted Frank follows. A version with charts will be posted next week.

Ted Frank finds it “curious” that I focused on the number of physicians practicing in Texas rather than the number of physicians per capita. Really? The Texas Medical Association (TMA) and the Texas Alliance for Patient Access (TAPA) used the number of physicians when asserting that the 2003 tort reforms caused a “staggering” increase in physician supply. I used the same measure when saying they are wrong. Meeting their claims head on is not even remotely curious. It’s the right thing to do. Notice that Frank did not accuse me of misrepresenting their positions or of being mistaken.

Frank did ask about physicians per capita, however, and this might be a better measure of access to health care than TMA or TAPA employed. Having worked with this measure, I know that substituting it for the raw number of doctors won’t affect the results. However, to satisfy Frank (and any other skeptics), I produced the charts below, which display the number of practicing physicians per 1000 Texas residents and the 1- and 2-year percentage changes in this ratio over time. Again the growth of supply is steady, and the increase from 2003-2005 is not impressive. Also again, the largest increases occurred from 1997 to 2000—the years that earned Texas its spot on the AMA’s list of states suppose enduring malpractice crises that threatened patients’ access to care. Finally, and again as before, the increases for 2004 and 2005 are below the 1- and 2-year averages for 1990-2002. Even using physicians per capita, the conclusion is clear: access to care did not grow especially fast in the immediate post-reform years.

[charts here]

Turning to the subject of licensing, Frank eyeballs the time series (which is very short), thinks he spots a pattern, and contends that the number of physicians will jump sharply in 2007. A proper response is, So what? TMA and TAPA claimed that Texas’ physician population surged in 2004 and 2005. Dr. Marcus predicted 4000 new licensees in 2006. These assertions are false and always will be. Nothing that happens in 2007 can change that.

Moreover, it is impossible to say how many new licenses 2007 will bring. Instead of theorizing about this from the armchair, as Frank did, I spent an hour on the phone with a friend at the Texas Medical Board (TMB) before submitting my original post. He did not predict a massive up-tick in 2007. He said TMB cannot process applications more quickly without more FTEs, which he hopes the legislature will fund. He also said the increase in applications in 2006 caused the number of new licenses granted to fall because disgruntled applicants flooded TMB employees with phone calls. My aim in pointing out the bottleneck at the TMB was partly to build a fire under the tort reformers who sold Texans the 2003 reforms. Those who promised Texans more doctors should convince the legislature to appropriate the money needed to clear the backlog. I know of no reason to expect it to clear on its own.

Finally, Frank takes a swipe at attorneys and asserts that “[t]he median Texan came out ahead” on the 2003 tort reforms. As for attorneys, I agree (if this is Frank’s point) that the loading costs on malpractice payments are shamefully high. Doctors, hospitals and their liability carriers could reduce them considerably by paying valid claims without forcing patients to litigate. Maybe in the future, they will. A movement led by COPIC and gaining steam elsewhere gets money to injured patients quickly and avoids litigation. As for the median Texan’s wellbeing, Frank offers no evidence for his assertion and I know of none that would prove it.

Posted by: Charles Silver | November 3, 2006 10:17 AM

Thank you, Professor Silver, for providing additional data.

You acknowledge that, under reform, doctors come out ahead. (This is appreciated: there are numerous dishonest/disingenuous opponents of reform, including several who write on this website, who argue nonsensically that malpractice insurance rates are unrelated to the expense of providing malpractice insurance. We are to believe, under these arguments, that the doctors who run nonprofit mutual insurance companies are conspiring to charge themselves too much money.) Under normal laws of supply and demand, ceteris paribus, the increased financial incentives will increase the supply of doctors.

Of course, ceteris isn't paribus: when one goes to measure this, there are confounding variables; one can't simply plot a single variable with fifteen data points, and claim that this demonstrates anything. (It's perhaps too much to expect a lobbyist group's talking points to be this nuanced, but, you and I, as academics, surely aspire to more than simply creating competing talking points.) For example, during the crisis period, as the number of doctors went up in Texas, it went up in the rest of the United States as well. In fact, the number of doctors went up in the rest of the United States faster than it went up in Texas, even as the population of Texas went up faster than it did in the rest of the United States. One thus needs to control for these confounding variables.

Numerous studies have shown, that when other variables are controlled for, noneconomic damages caps do increase the supply of doctors. (Klick and Stratmann (2005); Kessler, Sage, and Becker (2005); Matsa (2005); Dranove and Gron (2005); Dubay, Kaestner, and
Waidmann (2001); Mello et al (2005); Klick and Stratmann (2007 forthcoming)). The effect is especially pronounced when one observes "high risk" professions such as ob-gyn, emergency medicine, and neurology; doctors in low-risk fields such as dermatology are obviously going to care less about the malpractice regime they practice in, and including doctors relatively unaffected by a liability regime in the mix of data only dampens the real effects. And a single time-series for a single state, as you would surely admit, does not show otherwise.

What your additional data shows, however, is not that liability reform did not create additional willingness for doctors to practice in Texas, but that confounding factors, such as excessive regulation, are preventing the market from reaching its equilibrium point as quickly as it would otherwise. That's not an argument against reform, that's an argument for a better regulatory regime irrelevant to the liability process. In other words, the Texas doctors' groups you cite are guilty of a non sequitur and failing to address another problem in the medical regulatory regime, but the thrust of their comments regarding the effects of liability reform are both correct and consistent with your own acknowledgement that doctors' lots are improved by reform.

I didn't note this earlier, but it's rather misleading to claim that the public choice scenario behind malpractice reform is "wealthy and concentrated interest groups use the political process to advantage, at the expense of groups whose members are anonymous and dispersed." You're surely well aware (from, if nothing else, being a personal beneficiary from pronouncing on the ethics of its lead practitioners) that the organized trial bar is about as wealthy and concentrated an interest group as there is, and that interest group has spent millions calling in chips to block reforms that a majority of the U.S. House of Representatives and Senate supported, and had considerable success blocking reforms in Texas through obstruction in one or more of the three branches of government before the referendum process permitted the popular will--the mass of anonymous and dispersed voters--to finally have its way over the powerful wealthy and concentrated interest group.

You ask me to back up my assertion that the median Texan comes out ahead. There are many reasons, but the easy one is that, empirically, when given the choice of paying for services with or without the additional cost of expected noneconomic damages, consumers overwhelmingly prefer (ex ante) not to pay those costs. New Jersey auto insurance is a marvelous example: there, when they buy auto insurance, residents are given the choice of whether they want full access to the liability system, or whether they are willing to waive their rights to noneconomic damages in certain accidents in exchange for a lower insurance rate. Over ninety percent prefer lower costs ex ante to the opportunity for pain and suffering damages ex post. The revealed preferences of the larger group of "anonymous and dispersed" consumers are that they don't think the trial-lawyer-created game of randomly awarded noneconomic damages (where lawyers and administration costs receive the majority of the money spent) is worth the candle. This alone, without more, shows that there has been an increase in consumer surplus because their preferences have been honored.

There is more evidence, of course, but it's haggling over econometric studies over the health benefits of caps, and I've already gone off topic enough.



Posted by: Ted | November 5, 2006 7:54 AM