Stanford University

News Service



CONTACT: Stanford University News Service (650) 723-2558 COMMENT: Anne Royalty, Economics Department (415) 723-2298,
e-mail; also Thomas Buchmueller, UC-Irvine
(714) 824-5247, e-mail

Price incentives work in health care "managed competition"

STANFORD -- When employers appeared to be footing the entire bill, employees rarely bothered to comparison shop on price for their health care plans.

But in a study of Stanford University employees, economist Anne Royalty has found that a larger number of workers make price-conscious choices when the employer requires them to pay out of pocket the difference between the premium of the lowest-cost plan and the one they choose.

In the study of "managed competition" that she conducted with former postdoctoral fellow Neil Solomon, Royalty also found that older workers, employees with longer job tenure and those who insure a family member with a chronic medical condition pay less attention to price than fellow employees, which could lead them to become concentrated over time in the most expensive plans.

The study, along with another conducted by researchers at the University of
California-Irvine, offers the clearest evidence yet that restructuring health insurance to encourage price competition can help bring down the costs of health care, said Alain Enthoven, the Stanford economist who developed the idea of using managed competition to slow the spiraling growth in U.S. health care costs. Such real-world evidence was not available during the 1993-94 health insurance debate in Congress, he said.

Past studies have indicated that when employees don't pay more for choosing more expensive plans, they do not pay much attention to price differences among health plans. Insurers compete by emphasizing attributes of their plans other than price - akin to an auto salesman stressing the superiority of a car's chrome trim or brakes, but not bothering to post the sticker price. When employees do not pay the price of health insurance individually, it results in what Stanford economist Victor Fuchs calls the "restaurant check problem - you go out with a crowd to a nice restaurant and suddenly everybody develops a taste for lobster and baked alaska."

Prior studies also have found that few people switch health plans from one year to another - about 3 percent of employees in a 1984-85 study in the Boston area. People may be less willing to change brands of health insurance than they are automobiles or potato chips, Royalty said, because there are indirect costs of switching. For example, a person may feel hesitant to switch doctors because the new physician doesn't know his or her medical history. A main purpose of Royalty's study was to see how these so-called switching or "transition" costs affected the sensitivity of various groups of workers to price changes.

Stanford study

The Stanford study looked at how more than 5,000 employees of Stanford University responded to health insurance options offered in November 1993 and 1994 for the years 1994 and 1995. In 1993, the university changed its medical benefit offerings for 1994 in a way that conforms, with minor exceptions, to Enthoven's model. That model calls for an employer to offer employees a choice of insurance plans that:

  • Provide extensive information on insurance coverage provisions so that consumers can compare them.
  • Require employees to pay for their health insurance costs at the margin - that is, pay the part of the premium that is above the cost of the cheapest plan offered.
  • Minimize the competition among insurers on plan provisions other than price.

For example, Stanford would not allow one insurance provider to offer maternity care coverage benefits and another plan not to offer them. All plans are required to charge a patient co-payment, currently $10 for each visit to the doctor. Employees pay out of their own paychecks the difference between the total premium of the plan that they choose and the cheapest alternative.

(These provisions apply to the bulk of the university's faculty and staff but not to union employees. Because of a union contract, the university offers more traditional health insurance payment options to union employees that basically allow an employee without dependents to get health insurance with no monthly premiums, regardless of the plan chosen, but requires a worker with dependents to pay more to insure them than non-union employees pay.)

In order to quantify the sensitivity of employees to the price of four specific health plan choices, Royalty and Solomon calculated what economists call "price elasticities" - how the probability of someone making a purchase correlates with changes in the price of an item. In this case, however, they did not use the full price - the total insurance premium - but only the employees' out-of-pocket cost. Depending upon the plan, the researchers found a 2.8 to 7.9 percent decrease in the probability that an employee would choose a given plan as the
out-of-pocket cost increases by 10 percent.

The majority of previous studies of consumers' sensitivity to the price of health insurance found only a 1 to 2 percent decrease in the probability with a 10 percent increase in price, Royalty said. Those studies were not conducted in "managed competition" settings, she said, so the Stanford study suggests that managed competition substantially increases people's sensitivity to price.

Royalty and Solomon also calculated the price elasticities for the full premium, which is of interest to the insurance companies, because they collect the full premium from the employer and employee combined. Looked at from the insurer's perspective, consumer reaction to a percent change in total price was even greater than from the employee's perspective. The price elasticities are sufficient to encourage insurers to compete with each other on price in order to maintain or improve their share of the market, Royalty said. The study supports the idea, she said, that increased competition led to the decreases in premiums for all Stanford health plans between 1994 and 1995 and for two of the four plans between 1995 and 1996.

One plan decreased its total premium by 13.6 percent, another by 9.7 percent and two by 5 percent for the first year of managed competition. In the second year, one plan decreased its rate by 7.5 percent, another by 4 percent, and the two others' rates remained flat.

"The university's savings the first year was about $2 million, and half of that went back to employees" as a group, Enthoven said. The university increased its share of contributions to health care the following year and gave employees "choice dollars" to spend on other benefits as they wished.

For some employees, however, the out-of-pocket cost of health insurance went up the second year of managed competition, because the university ties its contribution to a given percentage of the cheapest plan. The two plans that kept their prices the same for 1996 as for 1995 actually cost employees more because the cheapest plan had dropped its price.

University of California study

Another study of managed competition, within the University of California's nine-campus, three-national-laboratory system, also found that employees responded to changes in
out-of-pocket prices.

Facing state budget cuts in 1994, the University of California system also decided to switch from a policy of subsidizing more costly health insurance plans to a policy of contributing a constant dollar amount, regardless of the plan that individual employees chose - a major tenet of managed competition. As a result, employee premium contributions increased for roughly one-third of the university system's 100,000 employees.

A study by Professor Thomas Buchmueller and Paul Feldstein of UC-Irvine's Graduate School of Management looked at how the new policy affected employees' willingness to switch health plans. The researchers did not calculate price elasticities per se, but found that "whereas only 5 percent of the employees facing constant premium contributions switched plans, roughly one-quarter of those facing premium contribution increases of less that $10 per month switched to lower cost plans. Higher price increases led to even greater rates of plan switching."

Employee contributions roughly doubled between 1993 and 1994 for California-based UC employees who were enrolled in the high-option indemnity plan, and 42 percent of that plan's membership switched, Buchmueller said.

At Stanford, "we are seeing a gradual movement of approximately 100 employees annually out of the more expensive plans to less costly ones," said Jim Franklin, associate vice president for total compensation.

Until the Stanford and UC experiments were undertaken, Enthoven said, "we had very little good econometric evidence of price elasticity of demand in choice of health plans. In fact, it was very frustrating to me that I couldn't point to any natural experiment where managed competition had been tried and succeeded. That was an important factor in the 1993 and 1994 health care debates because the Congressional Budget Office said 'we can give an estimate of what price controls would save, but we can't estimate what market forces would do,' " he said.

Costs of switching health plans

Critics of managed care have suggested that price competition ultimately might defeat a major purpose of health insurance, which is to spread the financial risks of getting sick over a large number of people. If the sickest people, or people who know they have a high risk of getting sick, are willing to pay more for insurance than others, then price competition among insurers might lead to ever larger gaps in the costs of plans.

This might happen, Royalty explained, because the sickest people would concentrate in some plans, which would then have higher costs because they had sicker people to treat, forcing them to raise their premiums even more.

Royalty and Solomon found some evidence that older and sicker people are somewhat more willing to select more expensive insurance plans and, when prices are changing, less willing to switch plans in response to those changes.

In a survey of 1,377 Stanford employees, about 25 percent said that someone in their family had a chronic illness. This group did not switch plans in response to price as much as did the 75 percent of the survey group who did not have a family member with a chronic illness. Those without chronic illness in their families were, for example, almost three times more likely not to take the most expensive insurance option as those with a family illness.

The survey also indicated that a 30-year-old worker was almost three times more sensitive to a change in price than a 50-year-old worker, Royalty said, a more statistically significant finding for age than for families with chronic illness. "Since older people have, on average, higher medical costs, this too is an indication of some adverse selection in this market," Royalty and Solomon wrote.

An increase in salary also increased the probability of an employee choosing a more expensive plan, but not as greatly as the effects for age and chronic illness, she said.

The results provide some evidence, Royalty said, that "adverse selection" could be a problem for managed competition models, but she added: "I don't think it looks as bad as some people thought it might. This is the first case study to look at differences in people's price sensitivity, and we will need more studies to say anything conclusive about the potential for adverse selection problems."

The employees who are less sensitive to plan price, Royalty said, may be those who place a higher value on the "accumulated knowledge in the doctor-patient relationship. If you switch doctors, there is some medical history that's not going to be picked up by just transferring files. On the other hand, in our setting, a lot of potential plan switchers don't have to switch doctors because many of the clinics and doctors are available under more than one of the health plans."

In Stanford's case, the four primary plan choices were:

  • A prepaid group practice that offers clinics and hospitals all under its own name, Kaiser Permanente, with a $10 co-payment fee for each visit. Its total premiums have been 11 to 15 percent lower than the average total of the other three plans from 1993 to 1995. Therefore, Kaiser has been the base upon which the university calculates its contributions to employee health plans.
  • Two network HMOs, TakeCare and Health Net, which also offer the $10 co-payment structure, but which allow members to choose physicians from a range of well-established local clinics and private practices.
  • A "point-of-service" plan, called Triple Option. The first tier of coverage is similar to HMO coverage, limiting the number of physicians available and charging the $10 co-payment fee for office visits or hospital stays. Specialists are available within this tier only with referral from a primary care doctor. Plan members who want to see a doctor outside the network on their own, however, can do so. After paying a deductible, they are reimbursed for a portion of their costs.



Download this release and its related files.

The release is provided in Adobe Acrobat format. Any images shown in the release are provided at publishing quality. Additional images also may be provided. Complete credit and caption information is included.

© Stanford University. All Rights Reserved. Stanford, CA 94305. (650) 723-2300. Terms of Use | Copyright Complaints