Adverse selection on aisle two

Wading into Wal-Mart's latest memogate over its efforts to control health care costs, Alex Tabarrok opines that the company is suffering from that most dread of insurance-related malladies -- "A.S.":

Wages for low-wage workers have been flat in recent years but health care costs have been increasing.  For a company like Wal-Mart, which pays many of its workers modest wages but does offer a reasonable health insurance plan, this is an invitation to adverse selection.  As the value of the wage component of the Wal-Mart benefit package has declined relative to the value of the health insurance component Wal-Mart has attracted more workers who want the job for the health benefits, i.e. sicker workers.

Though I may have trouble keeping many friends if I constantly subject their health care posts to nit-picky criticisms, I have to say I think the good Dr. T. is a bit off in his analysis, in part led astray by the N.Y. Times article that sparked his post.

Without question, Wal-Mart draws its labor pool from among the least desirable demographics, in a health care cost sense. Thanks in large part to a greater propensity for obesity, the working poor are disproportionately likely to suffer a variety of malladies -- high blood pressure, high cholesterol, Type II diabetes, acid reflux, hypertension, sleep apnea, respiratory problems, liver and kidney problems...you name it. They are also more likely to be smokers, more likely to be victims of violent crime, and have disproportionately more children -- leading to big bills both for obstetrics and for infant and childhood health care.

All of this contributes to what I'd call the "unfortunate selection" the company faces, but this is true of pretty much all large retail operations. The Times quotes Diane Rowland of the Kaiser Commission on Medicaid and the Uninsured saying that the company faces a disadvantage competing against retailers who offer no health benefits at all, but this is a radical overstatement. The number of retail jobs that don't offer SOME coverage is vanishingly tiny, by and large limited to those mom-and-pop operations that can't qualify under the federal ERISA law to be exempt from state-level benefits mandates. For those that are exempt, there is a tremendous advantage to offering compensation by way of benefits rather than wage salary -- both are fully deductible for the employer, but health benefits, unlike salary, are not subject to income and payroll taxes.

But to say that Wal-Mart's demographic problem amounts to truly adverse selection, one must ignore a rather crucial fact about Wal-Mart employees. That is, while most large firms that offer group health coverage typically see take-up rates (the proportion of those eligible for coverage who actually elect to purchase it) of 75% to 90%, at Wal-Mart, the rate is only about 50%. So, if the argument is that some large number of Wal-Mart employees sought their jobs solely to get access to health insurance....then why aren't they buying it?

To answer the question, it's important to keep in mind some other facts about Wal-Mart employees. Namely, that they are disproportionately likely to work two or more jobs (and thus draw coverage from some other employer source); disproportionately more likely to be women and/or "secondary" sources of household income (and draw coverage from their spouse's plan); disproportionately more likely to be minors or students (and draw coverage from their parents); disproportionately more likely to be seniors and/or retirees (and draw coverage from Medicare or a pension package); and, most crucial of all, precisely because of the very low wage rate at Wal-Mart, its employees are far morely likely than those at most firms to be eligible for Medicaid and other public health programs.

This last fact is the one that has done the most to engender public scrutiny of the company (and no small degree of public embarrassment.) According to a study of Wal-Mart's California employees by Ken Jacobs and Arindrajit Dube of the Labor Center at U.C. Berkeley, families of Wal-Mart employees utilized 40% more in public health care (such as that state's MediCal and Health Families programs) than did those of all large retailers. In response to such figures, the state in 2003 passed a "pay or play" health care initiative -- narrowly repealed by way of a Schwarzenegger-backed ballot referendum in 2004 -- that would have, among other things, defined the percentage of premium costs for which large employers would be responsible. Similar measures have since passed in Maryland and New Hampshire, and are gaining ground in a number of other states, helped along by big labor's ability to sell the notion to fiscal conservatives that -- with or without anything resembling Clintoncare -- Wal-Mart is already sloughing off its employee health costs onto the taxpayer.

And so, the company now increasingly finds itself between the rock of rising health care costs, and the hard place of government fiat, which the much-discussed "secret memo" from Wal-Mart's Susan Chambers to the company's board rightly recognizes is driven by some valid criticisms:

Healthcare is our most pressing reputation issue because well-funded, well-organized critics, as well as state government officials, are shining a bright light on Wal-Mart's offering. Moreover, our offering is vulnerable to at least some of their criticisms, especially with regard to the affordability of coverage and associates' reliance on Medicaid.

In total, the memo marks a brilliant strategy of counter-attack that looks to address both problems simultaneously. Moving toward high-deductible plans would lower costs to the company, lower premiums to employees, and likely increase take-up rates of the offered plans (and lower Medicaid take-up rates among employees.) As a sweetener to offset employee dissatisfaction, the high deductible plans would be coupled with HSAs for those who have been enrolled in a company plan for at least one year (giving the HSA benefit right off the bat would likely increase take-ups a little TOO much, and contribute to the rising cost problem.)

The 12-page memo, which I urge any interested parties to read in full, also offers a host of other strategies for addressing the company's labor and health care problems. Cutting cross-subsidies to spouses would decrease the job's attractiveness to older, costlier applicants, while making it easier for part-time associates to gain coverage (moving toward an average of one year for eligibility rather than two) increases its attractiveness to younger applicants. The company also would respond to employee demand for benefits that aid them in obtaining education and home ownership -- both of which not only prove to be more stable than health care costs, but also help attract higher income and higher educated workers, with more attractive health demographics.

But the real question is this: at what point do health care costs, and the relative unattractiveness of the low-income demo, combine to lead Wal-Mart to the seemingly counter-intuitive conclusion that the best way to control its labor costs would be to RAISE salaries?

Comments

[Kevin Brancato]

You wrote:

*****

So, if the argument is that some large number of Wal-Mart employees sought their jobs solely to get access to health insurance....then why aren't they buying it?

*****

This all depends on the pool of alternative jobs WM associates have. I think WM associates are far more likely to work for other retailers than outside the retail industry (but I have no data on that).

http://tinyurl.com/8wajc

The recent internal memo puts Wal-Mart equal to national firms, but considerably higher than other retailers in *eligibility* for health insurance. For most retailers, only 56% of employees are eligible for insurance, while 81% of WM's associates are.

However, the *participation rate* for Wal-Mart employees was slightly lower than other retailers (and way lower than national employers), yielding an *enrollment rate* of 48% of WM associates in WM's health insurance plan (greater than 36% for other retailers, but much lower than the 68% for national employers).

While pretty weak, this does seem to present some evidence that some Wal-Mart employees -- those whose alternatives are other retail -- are in it for the health insurance, as unexpected as that might sound.

[Btw, I couldn't get my TypeKey to work on your site.]

[R.J. Lehmann]

Thanks for your comments, Dr. Brancato. I'm not sure that I'd necessarily jump to the conclusion that Wal-Mart employees with multiple jobs are markedly more likely to work in other retail positions, but I also lack any data on the subject.

I do think it's worth noting that the difference between 56% eligibility of most retailers and the 81% at Wal-Mart is almost entirely explained by the degree to which Wal-Mart makes benefits available to part-timers. Relatively few take them up on the offer, which would be consistent with the make-up of the part-time labor force, consisting largely of students (who tend to be covered by their parents) and moonlighters (who tend to be covered by their primary job.) Speaking anecdotally, it's my understanding that in many communities where Wal-Mart is the only major retail outlet, it's fairly common for teachers and farmers and factory workers and so forth to pull down a few shifts a week at Wal-Mart just to gain access to the employee discount card. I myself worked part-time for three years at a Barnes & Noble for much the same reason.

So my hunch is that, because of the tendency for part-timers to eschew participation, the very high eligibility rate at Wal-Mart (or very low eligibility rates at other retailers, depending on your perspective)skews the enrollment rates for both in ways that are somewhat misleading. For traditional full-time employees, the take-up rates seem fairly comparable among the major retailers, all of whom are generally below the national average because their employees will be more likely to get coverage from spouses, parents or the government.

In any case, the internal memo also cites an employee survey suggesting that only 3% chose Wal-Mart because of the benefits it offers. Perhaps that's not entirely reliable, but I think it's suggestive of the notion that any adverse selection Wal-Mart suffers WRT its nearest competitors is probably pretty small.

*

I'm not sure how the Typekey system works, but I've gotten a number of emails recently suggesting the folks at Typepad are in the process of upgrading the system, so hopefully it's a temporary bug.

[Frank Howland]

A question I had about the memo was the claim that associates with longer tenure are no more productive than associates with short tenure. Yet, according to the memo, employees' wages rise with tenure: "Given the impact of tenure on wages and benefits, the cost of an Associate with 7 years of tenure is almost 55 percent more than the cost of an Associate with 1 year of tenure, yet there is no difference in his or her productivity (Exhibit 2)." (I'm quoting from the 27 page version of the memo available I believe from the New York Times.) Are Wal-Mart's wages not equal to the value of marginal product? Or are Susan Chambers and McKinsey and Co. confused? If Wal-Mart is not setting the wage equal to value of marginal product, why?

[California Health Insurance]

It is unfortunate to hear wages for health care costs have been rising. Health insurance is an important aspect to many lives.

[R.J. Lehmann]

It certainly is.

[Benjamin]

Anyone who wants to tell me that socialized medicine is bad can kiss my exhausted butt. I have been trying to get a bi-pap machine for months and I have insurance!

[Benjamin]

After battling with my insurance company I was finally able to get them to pony up for a biPap machine. After my $1200 deductible plus 20% I a out around $1400 but its better than nothing. Happy holidays.