The Perils of Regulatory Policymaking by Opinion Poll:
Consumer Reports on "health care reform"
6 Feb 2008 in Regulatory Economics, Regulatory Policy
Consumer Reports has sponsored a poll on health care that yielded results characterized as "revealing a significant disconnect." But there is no disconnect at all if elementary economic analysis is applied.
Phil Crosby at PNN Online is confused by this alleged "disconnect":
Americans overwhelmingly support reforms (82%) that would ensure that all uninsured Americans have access to quality, affordable health care, and a large majority agree on six prescriptions for change, but when it comes to how those changes should be adopted, there is less consensus.
Each of the "reforms" respondents were asked about involved getting more health care:Among the 1,200 American adults who responded to the CR telephone poll, more than 80 percent said that a reformed system should guarantee the following: coverage for all uninsured children; protection against financial ruin due to a major illness or accident; the ability to obtain coverage regardless of a pre-existing condition; coverage that continues even when people are laid off, changing jobs, or starting their own business; premiums, deductibles, and out-of-pocket expenses that are affordable relative to family income; and, the ability of people to keep their current health insurance if they choose.
Given that each "reform" on this list is arguably a good thing, what's surprising is that about 20% of respondents declined to support them. Why might they be opposed? One reason is that the costs of achieving these "reforms" was not included in the question, and even if it had the trade-off would have been only hypothetical. Survey respondents consistently state a greater willingness to bear costs than they actually do in practice. Some survey respondents probably take costs into account when they answer opinion polls For those respondents who don't, it's impossible to interpret what they mean.
Crosby summarizes respondents' discordant views on health care policy changes that have been proposed:
CR asked consumers for their opinions on four approaches to reform, including those proposed by the leading [presidential] candidates. Respondents could choose multiple options, so percentages add up to more than 100:
One reason for discord is uncertainty about whether any of these proposed policy changes would actually achieve any of the benefits their advocates claim. Another reason is that the costs of each proposal are uncertain, as is the crucial question of who would have to pay for it. - The most popular proposal -- a mixed public/private system that would require all uninsured Americans to buy health insurance -- drew support from half of the respondents. Massachusetts is in the midst of implementing that type of system and versions of it have been proposed by the leading Democratic presidential candidates.
- Thirty-six percent of respondents support public insurance, similar to the Canadian health-care system.
- One-third of respondents favored a mix of employer-sponsored plans, private health-insurance plans, Medicare, Medicaid, and other public programs, which is the arrangement we have today.
- Only 26 percent supported the idea of giving tax incentives for individuals to purchase insurance and relying on market pressures and competition among insurance companies to hold prices down. Versions of this idea have been proposed by several leading Republican candidates.
Is there a meaningful difference in preference between the first ("Democratic") option and the last ("Republican") option? Maybe. But it depends a lot on how the questions were asked. Because Consumer Reports (and its parent, Consumers Union) has a well-known preference for the "Democratic" option over the "Republican" option, we'd like to take a look at the survey to see if these preferences infected the survey design. Unfortunately, CR did not publish either the survey or the results -- just its own summaries and editorial opinions about them. This is not the way high-quality survey research is performed.
Consumer Reports also says that respondents blamed insurers and pharmaceutical companies for rising health care costs. This amounts to asking survey respondents for their opinions about what the facts ought to be. The actual sources of rising cost can be ascertained from objective data, and CR acknowledges the discrepancy between opinion and fact:
Drug companies, insurers, politicians, lawyers, and the bad habits of Americans all figure into high and rising health-care costs. But the biggest contributors to high costs—doctors and hospitals—get off easier among consumers, our survey found.
Surveys can be useful tools for ascertaining opinion, but they are rarely good for finding out facts. Opinions can be fickle, and how questions are framed -- and even the order in which they are asked -- can easily affect the results.
Consumer Report's sample consisted of 1,200 people. Were they representative? We don't know, because CR also doesn't reveal its survey methodology.
