The Mad Political Scientist
Hocking loogies from the ivory tower
The Middle Class as the Prime Mover in Public Opinion
Nov 20th
(Or, Good News for Idealists!)
My sister is one of a handful of regular readers on this site. I think my brother is the other. (I have kind of small hands.) When I talk about political science with her, she usually walks away with two things: a better understanding of a concept and a sense of creeping, inescapable doom. The former is a point of pride; unlike some of my colleagues, I actively enjoy talking about politics with the intelligent laity. It’s like teaching, but people ask interesting questions and challenge me. Frankly, I get a better understanding of whatever I’m trying to explain along the way, which is also pretty helpful. The latter point (the doom thing) has to do with my worldview. I can come off as a pretty strong believer in the findings/views of Philip Converse. In short, the average person doesn’t know a lot about politics, nor does he/she particularly care. The people with the most information, for better or worse, are also typically the ones with the most money. If you’re feeling depressed at this point, you’re not alone. Sometimes I wonder why my sister bothers to call me at all.
There is, however, hope. Part of this is a paper that I’m working up, so I’ll save the hopefully-groundbreaking stuff, but here’s what the existing literature supports. I’m making a few connections that aren’t in the original works, so you don’t want to use this as your Cliff’s Notes for public opinion class, but here goes. We know that public policy is at least reasonably responsive to shifts in public opinion. There are anomalies. Policy reacts to preferences of the rich better (more quickly and accurately) than the preferences of the public at large. The correlation is only really strong when issues are salient (when people are thinking and talking about them). To sum up, the responsiveness of our representative democracy is not perfect, but it’s there.
From the previous paragraph, you should take two things. First, policy tends to move at least loosely with public opinion. Second, public opinion moves. You may have read those two points and wondered why I felt the need to reiterate. Well, the simple fact is that although public opinion moves, the opinions of high-information (usually but not always high-income, high education, white, male, and all the other things Glen Beck and Cynthia McKinney don’t have in common) individuals are relatively stable. There are a number of processes at work here, but they tend to follow along one vein. I’m a liberal who occasionally switches from NPR to Fox News when on an afternoon drive. When I hear Beck, Limbaugh, Hannity, etc., making an assertion, I almost immediately have a retort. Some of them are pretty straightforward (“No, Mr. Beck, I don’t want to sneak into your house and kill your grandmother; I just want mine to be able to afford to live to see my graduation.”). Others are more complex (“That amendment is unlikely to get through conference, Mr. Hannity.”). However, my responses are always at the ready, even when I’m not trying that hard. Even though I’d like to think otherwise, it’s pretty unlikely that anyone on Fox News is going to persuade me to abandon my views. I’ve already thought my way through most issues, considered most of the points, and weighed the pros and cons. Similarly, my conservative friends don’t get as into MSNBC, presumably for the same reasons. (My friends are typically pretty high-information folk. It’s one of the benefits of spending your days in the ivory tower.)
OK, so if public opinion moves, and my opinion isn’t moving it, what is? If it’s not the elites, it logically has to be everyone else (or a subset of everyone else, but I’m keeping my ball of academic survey research awesomeness to myself for the moment. Also note that I’m using a very broad notion of “elite” here. Basically, if you care enough about politics to seek out this blog, you’re probably not the one moving public opinion, either. Also, I really should figure out how to do footnotes in Wordpress.) That may seem a little scary. Those of you who don’t spend your days in the ivory tower may think that the average American doesn’t know a lot about politics, and as I’ve mentioned earlier, you’re largely correct, at the individual level. However, the average American has some information, enough to be right on a yes/no question more than 50% of the time, especially if a “don’t know” option is available. If a large number of people is asked to give a “correct” answer, and are successful 51% of the time, the aggregate success rate asymptotically approaches 100% as the group size increases. The probability of a correct answer for a large group of uninformed individuals can easily eclipse the probability of a correct answer from a lone “expert.” In other words, Will Smith and Tommy Lee Jones got it wrong. People are smart; a person is stupid.
So what does this all mean? Well, in a nation of over 300 million, public opinion in the aggregate is going to be pretty intelligent, and can be used to inform good policy. The trick is that “magic” number. Individuals have to be able to get the right answer more than 50% of the time. That means good information is important, and quality trumps quantity. Even if information is not plentiful, it should be factually accurate. Those who over-simplify complex issues aren’t doing nearly as much harm as those who label liberals as socialists and conservatives as fascists. It also means that we should talk with intelligent people with whom we disagree. That’s a pretty tall order. I know a couple of staunch conservatives in my department, and try to talk politics with them on occasion. I’ve come to understand the views with which I disagree much more fully. I still think they’re wrong, but I understand their rationales, and am able to remind myself that not every conservative is a blustering windbag rage-a-holic. Similarly, I hope my conservative friends are able to remind themselves that not every liberal is an ineffectual elitist lodged in an ivory tower, incapable of ordering coffee with a word instead of a paragraph. Maybe it would help if we quit getting together at Starbucks….
The Public Option as a Gender Issue
Oct 22nd
When I was 24 and single, I decided it was time for a new car. The Hyundai that had served me so well since graduation just wasn’t cutting it anymore. Like every 24-year-old, I really wanted to go ahead and get a head start on my mid life crisis, and test drove a convertible, a Mustang Cobra SVT. It was and still probably is the best car I’ve ever driven, and the only one that really qualified as “high performance”. Since it was used, it was even fairly affordable (by car guy standards). I called my insurance company to check on a rate. I had been with them since I was 20, always paid on time, etc., and figured I would get a reasonable rate, having given them lots of money while puttering around town in my painfully practical economy car. You probably know what’s coming next. Despite my spotless driving record and long history, my insurance payment would be roughly the same as my car payment. As it turns out, the problem had to do with the shape of my genitals. Apparently, since I had not yet committed to one person with differently-shaped genitals, I was predestined to use any Mustang in my possession to play chicken with Lamborghinis. Some number cruncher somewhere had crunched some numbers (it’s what they do; it’s right there in the job title), and figured out that putting a single 24 year old behind the wheel of a 400hp convertible muscle car was probably not the best way to keep collecting premiums without paying for any repairs. I mumbled and groaned for about ten minutes. I complained that I was being discriminated against, even though as a white protestant male who grew up in a nuclear family and had both parents and grandparents around to help with the expenses of college, I knew everyone else in the world had it much worse. Then, I got over it. I bought a lower-powered, more practical ragtop, which I still drive and enjoy today. I paid a much less absurd amount in car insurance. The end.
Interesting story, huh? Well, believe it or not, there’s a politically relevant element in this tale of unconscionable social injustice. (I’ll stop complaining about this eventually, but that car was really, really awesome.) My main problem in acquiring affordable car insurance was something entirely beyond my control; I was (and still am, last I checked) male. Through no fault of my own, I’m in a group more likely to drive dangerously, especially at age 24 and behind the wheel of a car capable of race track performance. (Again, awesome car, but stay with me; I’m getting to the political stuff.) As I said, my (admittedly trivial) problem came into being when an actuary somewhere determined that covering the costs of young men in muscle cars would be expensive, and the insurance company would have to charge accordingly to stay in business. A more relevant problem, though, comes about when that same actuary looks at the health care costs of men and women.
I think we all basically know how insurance works, but let’s take a moment to examine it from a mathematical perspective. Let’s assume that, statistically, I will be in a wreck at some point in the next twenty years. Let’s also assume that the average cost of the wreck will be $15,000. Maybe my car will be totaled. Maybe I’ll have some medical expenses. Maybe I’ll just happen to scratch the fender of a very expensive car. There is, however, a problem. I don’t have $15,000 just sitting around, and if I did, I wouldn’t want to keep it laying around waiting on this car accident. Well, there’s good news. I’m not the only person who has this proposition to deal with. There are lots of other guys out there who will also probably get into a wreck in the next 20 years, costing themselves about $15k. Some enterprising guy (or gal) looks at this situation and sees the chance to make a buck. Some of our potential car-crashers would probably prefer to pay a smaller amount every month than pay a very large one-time cost. Since the cost of one $15k wreck per 20 years works out to $62.50, he offers a proposition. For $70 per month, he will agree to cover the costs of an accident for anyone willing to take him up on the offer. He will make, on average, $7.50 per month off each contract, essentially a surcharge for insulating his/her customers against risk. The original $62.50 from each customer is put into an account, from which the costs of accidents are paid. If we assume that all of the probabilities I’ve laid out above are correct, then after twenty years, when everyone has had an accident, he will still come out slightly ahead, and will have provided all of the drivers with some measure of security.
Of course, I’m leaving out a lot of stuff to give a very basic view of the actuarial sciences. The truth is that not everyone has the same probability of being in a crash. Some people are very good drivers, and some people only drive while texting about how much their martini glasses don’t fit in the cup holders. That second group is going to have to be in a second pool, and because of their higher risk, they will have to pay more if our hypothetical insurer is still going to get his $7.50 for tolerating their risk. Basically, if the expenses are going to be higher, then an insurer will have to charge more if he/she is still going to make a profit and stay in business.
OK, now that we have that out of the way, let’s return to health care. Consider for a moment the fact that health care costs for women far eclipse those for men, particularly below about age 50 or so. Seriously, before you go on to the next sentence, consider that point for a moment. Done considering? OK, good. If you really did stop and consider this point, then you know what’s coming next. If you lied to me, please stop that. Here’s the logical conclusion, though. It costs more to insure women, and not a lot can be done about it. My insurance problem had a pretty simple solution: buy a less powerful car. I did that, and aside from a little bit of ennui when I see someone lay a patch of rubber down the main drag, nothing bad happened. Health insurance, however, is far different. I stopped being a single guy in a high-powered car. It’s far more difficult to stop being a woman. (And your insurance probably wouldn’t cover it anyway.)
So what do we do? One solution is simply to require that insurance companies not be able to include sex when figuring out potential health care costs to set their rates. The argument for this is that it would be more fair, or failing that, at least justifiable on egalitarian grounds, since women make less than men on average. There are, however, three arguments against this, and I’ve only seen one of them get serious airplay. The well-known argument is that it’s a bit silly to require that my health insurance premium include the possibility that I will get pregnant, suffer from ovarian cancer, etc. The odds are REALLY long, and I would basically be subsidizing insurance for others who do need this coverage, regardless of whether or not I can afford to do so. Trust me when I say that political scientists are not particularly well paid, and mad ones even less so, making this situation a bit uncomfortable.
The second argument, one that I wish got more airplay, is a statistical one. Removing gender from a list of variables doesn’t really remove it. I know that sounds like an odd thing to say, so let me explain. I have it on good authority that there are a great many ways to skin a cat, but there are far more ways to measure a concept. In fact, early in the design phase of a project, it’s not uncommon to spend hours of office time kicking back and forth different ways of measuring even fairly straightforward ideas. Certain bits of medical history (child birth, for example) are exclusive to women, giving a way to measure sex without technically including it. Even if these details are also removed from the equation, there are plenty of other measures that, while not exclusive to women, still predict gender very well. Women are more likely to have a regular history of checkups than men (we hate the doctor’s office), see a variety of doctors, etc. If I am not allowed to put sex into an equation predicting health costs, I can easily compensate for that by plugging in various bits of medical history instead.
Is this in the spirit of the proposed law? Of course not. Is it evil, underhanded, and despicable? That’s debatable. It’s not fair that women require a greater amount (measured in dollars) of medical care than men. However, one could also argue that it’s equally unfair (and more actively so) to inflate costs for a low-risk group to subsidize a high-risk group’s coverage. Of course, all of this is moot if insurance companies don’t find good predictors of sex in their data sets. This, however, is all but a certainty. Insurance companies put a great deal of time and money into creating accurate data within the confines of state and federal regulations. It’s the lifeblood of their business. Using other data to get at forbidden predictors is nothing new. Insurance companies can and will try to do this; at best, regulatory agencies are relegated to playing whack-a-mole, determining which considerations are tantamount to discrimination and which are acceptable predictors of loss.
What is the implication of all this? It’s harder for women to get and keep insurance, and said insurance is and likely will always be more expensive. Some have even argued that being female is tantamount to having a preexisting condition. (If you just want some Senatorial incivility, click here instead.)That’s certainly some loaded language, and therefore not terribly productive. I would argue that this is better understood as a situation in which an unregulated market will fail to produce the easily accessible public good that is most often cited as a justification for its existence. Please don’t misunderstand; markets work a great deal of the time. When a market produces a repeated failure, however, it’s time to consider other options.
If a public option is sufficiently well-populated and does enough business with sectors of the industry that cater to women’s health concerns, it will be able to get the best possible deals on those sorts of health care. This is exactly how private insurance companies function. By getting deals that are ostensibly bulk discounts on hospitalizations, X-rays, and the like, they are able to turn a profit while still keeping premiums relatively low (remember our earlier $15,000 car crash example and think about the costs of health care; relatively is certainly a key word here).
What we have here is a case where there is no perfect solution. As I’ve said before, health care policy is a policy of compromise. We do not solve every problem, but rather decide which problems we are willing to tolerate in order address others more fully. However, the difficulties in creating a “sexless” insurance system effectively take the option off the table in the private market. All but the most dogmatic supporters of the free market would agree that in cases where the market does not create a feasible alternative (public goods like roads and national defense, for example), the government can intervene. My political leanings aside, I have to wonder why this part of the debate isn’t on the front burner. It seems like a good way to mobilize half the population in favor of a controversial policy.