Comments

Health Affairs encourages readers to engage in discussion via comments on our Web site.

• To RESPOND to a particular article: Click on the link "Submit a response to this article" in the box at the top right-hand corner of the article.
• To READ responses to a particular article: Click on the link "View responses" in the box at the top right-hand corner of the article.

Comments to:

Anna D. Sinaiko, Employers’ Responses To A Play-Or-Pay Mandate: An Analysis Of California’s Health Insurance Act Of 2003, Health Affairs Web Exclusive, October 13, 2004 [Abstract] [PDF] [HTML Version][Supplemental Exhibit] [Reprints & Permissions]

Comments:Submit a response to this article

Comments published:

An economist's analysis lacking common sense

Matthew Holt   ( 14 October 2004 )

A Misconceived Analysis of California's Health Insurance Act

Rick Curtis, Kanika Kapur (RAND), Susan Marquis (RAND), and Ed Neuschler (IHPS)   ( 15 October 2004 )

California's SB2: The Author Responds

Anna D. Sinaiko   ( 26 October 2004 )

Reality Bites: Labor Costs Do Affect Profits

Matthew Holt   ( 22 November 2004 )

An economist's analysis lacking common sense 14 October 2004
Matthew Holt, Consultant and Author The Health Care Blog Send comment to journal: Re: An economist's analysis lacking common sense matthew{at}matthewholt.net Matthew Holt	As a forecaster, I was trained to recognize that there's a big difference between being broadly accurate and precisely wrong. Anna Sinaiko's detailed analysis of Prop 72's likely impact falls into the latter category. Increasing labor costs (which is what SB2 will mean) may result in overall lower wages, but increased labor costs are more likely to result in lower profits. Sinaiko never mentions profits as a share of corporate revenues. Corporate profits as a share of revenues are currently at an all-time high, while wages are at their lowest in real terms for 30 years. Why should the costs of SB2 come out of wages rather than profits, given that one of them has been going up and the other down? Sinaiko correctly points out that the vast majority of jobs affected by SB2 cannot be moved out of California, unless San Franciscans want to drive to Reno to get a cheeseburger. So the total amount of money going into these businesses is likely to stay roughly the same. She almost neglects to mention that the wages of many of those covered by SB2 are at minimum wage or the equivalent (and legally enforced) "living" wage in some cities, so their wages can't be reduced, and the cost of hiring and managing more part-time employees may exceed the increase in labor costs from SB2. The broad analysis tells us that, whatever the libertarians say, laws demanding an increase in the lowest levels of compensation (which is in effect what SB 2 is) have almost no impact on unemployment rates, which are driven by the overall economy. I lived in the San Francisco Bay Area in 1999 when no amount of money could hire people. I was here in 2002 when you couldn't get a job no matter how little money you'd work for. (All the jobs that could be, had already been moved to India). At both times the minimum wages was the same. In the UK the Labour government brought in a minimum wage in 1999. Today the unemployment rate in the UK is the lowest it's been since 1972. Common sense also suggests that the people who oppose Prop 72 think they have something to lose. Who are those people? They are the big fast-food chains and the nonunionized discount stores, who have spent over $8 million against it. If they really believed that they'll be able to push all the costs of Prop 72 into their "labor" segment and have none of it come out of their "profit" segment, why would they bother opposing it?
A Misconceived Analysis of California's Health Insurance Act 15 October 2004
Rick Curtis, President Institute for Health Policy Solutions, Kanika Kapur (RAND), Susan Marquis (RAND), and Ed Neuschler (IHPS) Send comment to journal: Re: A Misconceived Analysis of California's Health Insurance Act rcurtis{at}ihps.org Rick Curtis, et al.	While Anna Sinaiko correctly indicates that there would be long-term labor-market effects from SB2/Prop.72, she makes multiple misinterpretations and errors that invalidate much of her analysis. First, Sinaiko dramatically overstates the potential problem by including in her analysis employers with 20-49 employees, who will not become subject to the mandate unless the legislature enacts a tax credit covering part of their net cost in order to reduce labor market effects in these smaller businesses. Since the tax credit is not available, SB2 will apply only to businesses of 50 or more workers, which makes her profile of affected workers considerably different from reality. We estimate that these smaller businesses constitute 85% of all non-offering businesses (with 20 or more employees) and employ 62% of all workers in non-offering businesses. While wage and employment effects are not limited to non-offering businesses, that is where the biggest effects will be. But we estimate that only 1% of the workers subject to SB2 in businesses of 50 or more employees are in non-offering firms. Sinaiko also greatly overestimates the costs required to comply with SB2. Most newly covered workers--about 96%--work for an employer who already covers other workers. The vast majority of such employers now offer health plans that are more expensive than SB2 requires. Thus, they could offset any cost increases for mandatory new participants or for complying with the required 80% employer contribution rate by reducing benefits. Alternatively, they could maintain the current plan and contribution level as one option for existing workers, and comply by contributing 80% to a new, much lower cost benefit option (which could be made the only option for newly mandated participants). Third, Sinaiko assumes that SB2 is not a mandate on individuals. But a careful reading of the statute reveals no way--other than by changing jobs--that workers can avoid paying their share of the premium. If workers could continue to decline coverage, many would do so, thus obviating most of the expected labor-market effects. Fourth, Sinaiko has misinterpreted the statute when she alleges that SB2 requires that the public program "fee" be set according to the actual risk of each employer group. While risk rating is allowed, it is not required. Further, it would be particularly difficult to implement under a "play-or-pay" construct where all employers are to be prospectively notified of their fee. Finally, Sinaiko is overly glib in stating that "state governments can mandate that employers provide coverage." This is a complex legal issue under the federal Employee Retirement Income Security Act (ERISA). For specific consideration of the ERISA Implications of SB2, we suggest a recent paper by Patricia Butler, which can be found at http://www.chcf.org/topics/healthinsurance/sb2/index.cfm?itemID=21740. The authors are part of a team conducting an analysis of implementation challenges facing SB2, funded by the California HealthCare Foundation.
California's SB2: The Author Responds 26 October 2004
Anna D. Sinaiko, Doctoral Student Harvard University Send comment to journal: Re: California's SB2: The Author Responds sinaiko{at}fas.harvard.edu Anna D. Sinaiko	The responses to my article raise good issues and suggest some limits on the economic analysis in the context of political and other realities, which I describe below. Let me begin with Matthew Holt’s letter, in which he questions: Can we finance health insurance by tapping into (“all-time high”) corporate profits? It has long been recognized in labor and health economics that firms regard the cost to employ a worker as the sum of cash wages and fringe benefits, and offset higher costs of health insurance with lower wages. While in the short term this may be difficult, over the long term, wage reductions are likely to be a consequence of an employer mandate such as SB 2. Rick Curtis and colleagues criticize the inclusion of firms with 20-49 employees in my analysis of SB 2. I included firms of this size because the stated intent of SB 2 is to include these workers under the mandate, and previous studies projecting the number of uninsured workers who will gain coverage under SB 2 included workers at firms of this size.[1] These workers will not be included in the mandate unless the State of California provides a tax credit to their employers in the amount of 20 percent of the required fee, which, given the current political climate in California, seems unlikely in the near future. As Curtis and colleagues correctly point out, if firms with 20-49 workers are exempted from the mandate, a large majority of workers who will become eligible for health insurance would work for employers who already offer health insurance. Employers can thus offset the cost of health insurance premiums for the newly eligible workers by reducing wages across all of their employees, including those who are already insured. In this case, the reduction in wages per employee will be smaller than that estimated in my analysis, and the magnitude of the reduction will vary with the ratio of newly eligible workers to total workers (being smaller when the ratio is smaller). It is also possible, as Curtis and colleagues write, that employers could reduce the net new cost to their business by restructuring their benefit plans and reducing benefits for other workers. In this case, the cost of the mandate would be shifted to those employees who experience a reduction in their benefits. This type of employer response is still a form of cost and would not negate the expense of the law, but employers’ flexibility to respond to the mandate by changing benefits instead of reducing wages is an important feature of SB 2. I also discuss the proportion of newly eligible workers in nonoffering firms who earn at or near the minimum wage. The estimate of the upper bound of this proportion for firms with 50 or more workers is lower than that presented in the paper when firms with 20-49 workers are included. As I point out in Note 29 in my paper, considering only employers with 50 or more workers who do not offer insurance, 32-36 percent of workers eligible for benefits under SB 2 earned within $2 of the minimum wage; however, the sample size for this analysis is small (n = 19 employers). More important than the reduction in the estimate of the upper bound, however, is that limiting the analysis to only those firms with at least 50 workers greatly reduces the number of firms and workers to which this analysis applies. Limiting the analysis of employer response only to those with at least 50 workers will reduce the number of employees who are potentially subject to economic adversity because of SB 2. However, this result underscores the fact that (in the absence of the tax credit discussed above) SB 2 will not affect many of the employers who do not offer health benefits, and therefore it will limit the number of Californians newly eligible for coverage. Curtis and colleagues also suggested that my analyses should have recognized that SB 2 obliges employees in affected firms to pay 20 percent of the required fee (or health insurance premium, in the case that the employer chooses to “play”). I appreciate their attention to this oversight, and I acknowledge that under an employee mandate the only way for employees to opt out of paying for coverage would be to change jobs or become self-employed. At two points in my paper (pages W4-474 and W4-477) I mention the issue of employee take-up of the new insurance benefit without appropriately noting that an employee would have to leave his or her job in order to avoid coverage. This omission was unfortunate and may have led to confusion about the role of employee mandates in my analysis of employer responses. In fact, I conducted the analysis of employer response for the case in which all affected employees would take up the benefit. In other words, the assumption of an employee mandate is already factored into my estimates. Curtis and colleagues’ fourth comment is that risk rating is not “required” but is “allowed” by SB 2. This is true, and I thank them for making this point clear. However, this critique does not change the outcome of my analysis: that if employers with above-average health risk can realize a lower cost in the public pool than in the private market, there is a risk that the public pool will attract groups with higher-than-average health risks, and the cost to participate in the pool will rise. Finally, Curtis and colleagues make an important point regarding the legal issues surrounding the implementation of an employer mandate under ERISA; an analysis of this issue was beyond the scope of my paper. That said, state mandates that employers provide coverage to their workers continue to be debated across the United States, and economic analyses need to be directed at anticipating and evaluating the impact of the mandates under consideration. References [1] Dube, A. 2003. “Impact of SB2 on Health Coverage.” Research Brief prepared at Institute for Labor and Employment, University of California, Berkeley, http:///www.iir.berkeley.edu/research/healthcoverage.pdf (26 October 2004); and Brown, E.R., Y. Hongjian, S.A. Lavarreda, et al. 2003. “SB 2 Will Extend Coverage to 1 Million Uninsured Workers and Dependents,” A Health Policy Fact Sheet (Los Angeles: UCLA Center for Health Policy Research), http://www.healthpolicy.ucla.edu/pubs/publication.asp?pubID=75 (9 September 2004).
Reality Bites: Labor Costs Do Affect Profits 22 November 2004
Matthew Holt, Author The Health Care Blog Send comment to journal: Re: Reality Bites: Labor Costs Do Affect Profits matthew{at}matthewholt.net Matthew Holt	In her reply to my letter, Anna Sinaiko wrote: "It has long been recognized in labor and health economics that firms regard the cost to employ a worker as the sum of cash wages and fringe benefits, and offset higher costs of health insurance with lower wages. While in the short term this may be difficult, over the long term, wage reductions are likely to be a consequence of an employer mandate such as SB 2." According to Sinaiko, economic theory says that overall gross labor costs are necessarily static and that any change in one part of labor costs will be compensated within those labor costs, by either lower wages or lower employment. Those of us who think of economics as a pretty imperfect science would be interested to know how this theory explains the U.K. experience with the introduction of a minimum wage in 1998. In that case, wages/labor costs went up, and unemployment in the affected groups went down. Assuming that the total amount of revenue earned by those employers stayed constant, something else MUST have gone down. Adam Smith said that there are only four elements in the cost of production: land, labor, capital, and "enterprise." Sinaiko effectively said that a change in the cost of one of these is self-regulating and that the others cannot change. That is nonsense, as the "cost" (or share of revenue) of all four elements changes constantly in any market, and if some part of labor costs went up because of a play-or-pay mandate, there is no reason why it has to be another part of labor costs that compensate for that increase. More importantly, Sinaiko never answered my final question. If this all washes out within labor costs and they stay constant, why are corporations (which care only about the "enterprise" or profit part of the equation) so dead-set against employer health benefit mandates? The reason is that they--like labor unions--live in the real world where battles over distribution of revenue happen because there actually IS something at stake. For proof let me quote an article by Reed Abelson in the New York Times, 1 November 2004 (URL = http://tinyurl.com/5w5ry), by Reed Abelson, stating that Wal-Mart offers fewer health benefits than its competitors. One result is that many of its employees wind up on public assistance of various sorts for their health care needs. Wal-Mart is more profitable than its nearest competitor, Costco, partly because of that, and Wall Street notices: "Wal-Mart says that 23 percent of its employees are not eligible for coverage, but that it covers 58 percent of those who are. "That compares with an insured rate of 96 percent of eligible full-time or part-time employees of Costco Wholesale, the discount retailer that is Wal-Mart's closest competitor nationwide. Costco employees--most of whom are not represented by a union--become eligible for health insurance after three months working full time, or six months part time. "At Wal-Mart, which has no union employees, many who work full time must wait six months to become eligible. Part-time workers are not eligible for at least two years. Because of turnover, some employees never work long enough to become eligible. "If there is any place where Wal-Mart's labor costs find support, it is Wall Street, where Costco has taken a drubbing from analysts who say that its labor costs are too high. Costco's pretax profit margin is only 2.7 percent of revenue, less than half Wal-Mart's margin of 5.5 percent." Unless health and labor economic theory doesn't cover Bentonville, Arkansas, and Wall Street (or for that matter, Medicaid payments made in Sacramento or Athens, Georgia), it's clear that different levels of health benefits paid by employers do impact their respective profit margins. Prop 72 was barely defeated by a coalition of fast-food corporations, discount retailers, and (with a late contribution for more negative TV ads) Wal-Mart. All of these groups hid behind propaganda about the terrible impact it would have on "small businesses," which were not even affected by the mandate (it was limited to those with 200+ employees). To suggest that large, low-wage employers were so aggressive in their opposition because they were somehow worried about the composition of overall static labor costs (or even the fate of small businesses!), as opposed to the fact that Prop 72 would have increased their labor costs at the expense of their profits, is refuted by the reality of who spent what in the run-up to the vote on November 2.