Listen to the podcast:
Far-reaching reforms of U.S. health care and flood-insurance law continue to stir controversy among voters, policy makers and other industry stakeholders. Against this backdrop, Knowledge@Wharton recently talked with two Wharton professors who are recognized experts on the complexities of the insurance industry: Howard C. Kunreuther and Mark V. Pauly, co-authors of a 2013 book titled, Insurance and Behavioral Economics: Improving Decisions in the Most Misunderstood Industry.
An edited version of the conversation follows.
Knowledge@Wharton: In the subtitle of Insurance and Behavioral Economics, you jointly refer to the insurance industry as ‘the most misunderstood industry.’ Broadly speaking, how and why is insurance so widely misunderstood, at least in the United States?
Howard Kunreuther: I think that people view insurance as an investment rather than as a protective activity. That is one of the basic misunderstandings. As a result, if they haven’t collected on their policy, they will cancel it because they feel that, in some sense, it was not a good deal — rather than [acknowledging] that the best return on a policy is no return at all. That is one aspect. The other feature is that on the very low-probability events, which is where insurance is most valuable, people say [a low probability event] is not going to happen to me. I don’t have to worry about it. I won’t protect myself.
Knowledge@Wharton: Mark, would you care to comment on these issues?
Mark Pauly: Insurance is different. When I go out and buy a clothes dryer, I put down my money. In a few days, the clothes dryer shows up at my house, and it is what I ordered. That’s that. But with insurance, you put up your money in the first instance, and if you are lucky, you get a cheesy piece of plastic that indicates you have insurance coverage. But what you are going to get is determined later by whatever happens. Once a bad event occurs, it is easy for people to forget the details of what they agreed to when they put down their money, and to think that they should be entitled to things that may not have been covered in the contract.
That’s not the only example where people sign long-term contracts. But it is a good example, and it is actually worse than most long-term contracts, because the whole intrinsic idea is that uncertainty and unpredictability will intervene, which makes it easy for people to be confused about whether they made a smart decision in the first instance.
The nature of insurance, of course, is that in a sense, everybody who buys insurance makes a dumb decision. Either you paid your premium, and then you didn’t suffer a loss — we claim that the best return from insurance is no return at all — or you did pay your insurance and you suffered a loss, and you wish that you hadn’t, because [the insurance] didn’t cover everything. Either way, people tend to be dissatisfied.
Insurers also sometimes misunderstand their business, particularly when dealing with catastrophic events. Regulators of insurers often misunderstand insurance as well because, like the rest of us, they want it to do everything good in the world. But sometimes, that’s not always possible.
Knowledge@Wharton: Who is the audience for this book, and if you could choose one major impact it could have on this audience, what would that be? Howard?
Kunreuther: I think the major audience is people who have an interest in what insurance can do — the general consumer. We tried to write this book in such a way that it is understandable. We tried to keep out the technical details in the text and write a series of stories and anecdotes that at least lets people know some of the challenges they face. It obviously should be of interest to insurers [as well]. As Mark indicated, insurers make similar decisions to what consumers do. They often don’t insure events because of the fact that they feel somehow the risk is too high. But, on the other hand, they will not take into account a risk when they feel it won’t happen. Terrorism is a good example that we can talk about.
Knowledge@Wharton: What are your views about that, Mark?
Pauly: We imagine, and I think this is true, that there is a fair number of people who, like us, are fascinated by insurance or who are very interested in it. We think, on the one hand, that this book will help explain to a typical intelligent consumer why insurance works the way it does, which is often hard for people to understand. We think it may help them understand public policy toward insurance, and what would be better or worse public policy; [and that the book w] make them better citizens, as well as better consumers, with regard to this crucially important dimension of public policy. That is the audience we imagined. As the world gets more and more uncertain and insurance plays a larger and larger role, it is much more important for people to understand the strengths and limits of insurance.
Knowledge@Wharton: Howard, how do you think these misunderstandings have played out specifically with the various stakeholders — customers, insurance companies, governments, etc. — with respect to the Affordable Care Act aka “Obamacare” and the Flood Insurance Reform Act of 2012?
Kunreuther: Let me talk about the Flood Insurance Reform Act, and then Mark will talk about the Affordable Care Act. Those [laws] form the centerpiece of our book…. I think the Flood Insurance Reform Act is a really interesting example of where Congress made a decision to require insurance premiums to reflect risk on second homes in hazard-prone areas as well as those homes that are subject to repetitive flooding.
Congress decided to do that because flood insurance is a national program; it is a federal program. The program was in debt because of events like Hurricane Katrina. There was a feeling by a number of Congressional people on both sides of the aisle that it was important to try to deal with insurance the way it should be dealt with, which is to have risk-based premiums. They also provided for a study that is going to take place beginning in January — it has been somewhat delayed — on affordability.
What has really occurred since Hurricane Sandy, which was in October 2012? The Flood Insurance Reform Act was passed in July 2012. If the reverse had happened and Sandy had occurred in July, it is highly doubtful that this piece of legislation would have been passed because it would have opened up a concern with respect to people having to pay much higher premiums than they had in the past. That’s what basically would happen. FEMA — the Federal Emergency Management Agency — is drawing new maps. In many cases, this requires premiums to be much higher, and there has been a great concern on the part of the public [who ask,] “Why am I paying a higher premium?” Some of that concern is understood. These people had a lower premium, and now it is being raised because of the fact that the maps are being corrected.
But there is also a very, very strong backlash to this piece of legislation because of the fact that there are low income people in this area who say [they can’t afford the high premiums]…. These people do not [realize] that the premiums [they were paying] were highly subsidized beforehand.
“People view insurance as an investment rather than as a protective activity…. As a result, if they haven’t collected on their policy, they will cancel it because they feel that, in some sense, it was not a good deal.” –Howard Kunreuther
Pauly: For health insurance, it is a little different. Actually, one of the points of our book is that people aren’t always confused about insurance. In fact, many insurance markets work pretty well. Most people have collision coverage on their car and fire insurance on their house, and they do sensible things. About 84% of the population has health insurance. Compared with, say, flood insurance — where the proportion was much lower — it is a less severe problem in terms of proportions. But in terms of the harm done to those people who don’t have health insurance, it is pretty large.
There were two reasons why people didn’t buy health insurance. The obvious one is they said to themselves, “I can’t afford it. Either I’m a poor person or I am a low-middle-income person, and given what premiums are, I just can’t cut that kind of money out of my budget to pay for health insurance.” Obamacare is going to directly deal with that population, and I think rather well.
But there is another set of people who did not buy health insurance. The typical poster child here is the “young immortal,” the person who says, “I am a young 30-year-old. I make $50,000 a year. What affordability means is not a technical term. Three thousand dollars for a premium is not outrageous. But I just don’t think it will happen to me. I haven’t seen a doctor in years. That might be a reason why it will happen to me, but I don’t think about it that way. And I don’t like being forced to buy insurance.” That is actually part of the law; not only being forced to buy insurance but also being forced to pay premiums that are much higher than what I would ever expect to collect.”
So there is good news and bad news about health care reform. [On the one hand], it does direct subsidies to the people who need them the most – the low-income and the lower middle-income people. But it has caused consternation in the part of the population that is relatively low-risk and doesn’t see that [health care] is their most pressing need.
Knowledge@Wharton: Mark, are misunderstandings about insurance and ignorance about the application of behavioral economics especially common in the U.S. because of the unique characteristics of our insurance market? Or is it because many American consumers dislike government-mandated programs?
Pauly: Health insurance is still voluntary – or will be until January 1. Then there will be a mandate, but the penalty is only a slap on the wrist. In most other countries, consumers can’t make the mistake of not buying health insurance because they are not allowed to. So in that sense, it is different. On the other hand, what we emphasize in the book is that the really hard cases are these low probability–high consequence events – [such as] the occasional flood that is disastrous. [As for] individuals without health insurance, young people don’t get sick very often, but when they do, it can be a disaster. Around the world, people have a hard time understanding why they should have insurance against those kinds of events, or how they should look at it when they do buy it. That’s because, almost by definition, most of the time you will have wasted your premiums.
This kind of misunderstanding is pretty global. However, because we allow insurance to be more voluntary in the U.S., we may hear about it a little more in terms of consumer mistakes. Other countries don’t let people make mistakes – although they may also cause them to do things that really aren’t sensible for them to do.
Knowledge@Wharton: Do young people get sick more often than people experience floods?
Kunreuther: I would think so, but you never know.
Pauly: It depends on where you live.
Knowledge@Wharton: Flood insurance is such a unique animal because of that very low-probability but high-risk characteristic. Why would you ever think consumers are going to change their approach to this kind of protection?
Kunreuther: I think that’s a very important point with respect to the voluntary purchase of insurance. As we note in the book, and as Mark indicated, a lot of people do not purchase flood insurance because they say, “It is not going to happen to me.” And they really don’t think about this low probability event. So from that vantage point, I think you have a major problem with respect to how that insurance has to be packaged.
There are two things I think that are normally done to deal with that. One is that insurance is required as a condition for a mortgage. In the case of flood insurance, that actually happened after [the National Flood Insurance Act], which was passed in 1968, revealed that so few people actually purchased it. They changed it in 1972 to require that any person who was living in a flood-prone area where flood insurance was available — and who [also] had a federally insured mortgage — had to buy [flood insurance]. So the requirements are a very important part of this. The other part that really makes this very challenging is what I referred to earlier, regarding the issue of affordability. It becomes extremely hard for people to somehow say, “I really have to have this insurance” when it is so expensive. In addition, the law requiring flood insurance was never very well enforced and has not been well enforced today.
So there are an awful lot of people who still don’t have flood insurance even if they are required to have it. And many people who were required to get it after a disaster – to get some type of disaster relief — canceled their policy a few years later even though they were required to have it. So we have a real challenge here in terms of how to deal with this.
Knowledge@Wharton: Mark, with respect to health insurance reform, do you think the framers of the ACA really looked into the challenges of health care reform through the lens of behavioral economics? As an economist who specializes in this, you are used to thinking of it that way. But are other people who make policies sometimes just not getting it?
Pauly: Policy makers in general have a hard time dealing with insurance. Because the simplest thing – if you observe something that looks like a mistake – is just to pass a law to make people do it. But as Howard just pointed out in the case of the regulation on flood insurance, it then often becomes very hard to enforce the regulation or very hard to get people to abide by it. I think the same thing is going to happen with the mandate to buy health insurance because it is relatively modest and because people will claim that they cannot afford it. That is why they are not doing it despite what the law says. It makes it hard to deal with insurance just by wielding a stick. So behavioral economics basically says, let’s look at a much broader set of restructurings that can make insurance attractive as well as something your mother told you that you should do, but you just have not gotten around to doing, which is how most of us look at it.
Kunreuther: I think that the regulatory community is an important component in the sense that it is very concerned with these equity issues. And often they will say, “We can’t charge a premium that reflects risk because it would really hurt a certain group.” This has been done with auto insurance, in fact, and other insurance besides the catastrophic risk. It certainly has been done in the case of both health insurance and flood insurance. This creates misunderstanding on the part of individuals because they think their premium is low because their risk is low. They don’t realize that they are being necessarily subsidized. That is particularly true in the case of floods, which is one of the reasons why we have this problem today.
Our response is that we have two very basic principles. One, which we already talked about, is that premiums [should] reflect risk. The second principle deals with equity and affordability. It basically says that if you are going to deal with equity and affordability, don’t do it through an insurance premium. Do it through some other means, like a voucher. Let people know what the risk is and then, if you feel you have to support them or subsidize them, do it with a voucher — like a food stamp is done for food. But this has to be restricted to only buying insurance; [it should not be] a voucher that they could go [and use] to purchase anything [else].
Knowledge@Wharton: Regarding the level of distrust between consumers and insurers, one of the ideas that came up in the book is for insurers to offer multi-year policies so that they diversify their risk over years, and consumers get a sense of stable prices. Do you believe these sorts of nuts-and-bolts suggestions would make a difference? Are there other ideas like this? Could this actually happen?
“Once a bad event occurs, it is easy for people to forget the details of what they agreed to when they put down their money….” –Mark Pauly
Kunreuther: That is our third principle. We really believe in multi-year policies for just the reasons you stated — partly because people would cancel their policies if they had that option and often they do just that. In the case of the flood program, we recommend that it be tied to the property, not to the individual, so that if the property is sold it could be dealt with in an appropriate way by the next homeowner. [He or she] would continue it. As for health insurance, there are actually examples of long-term policies.
Pauly: One of the issues is that, as you could imagine, people are concerned about losing their health insurance if they contract a chronic condition. While little known, it actually was true that in the individual insurance market, if you bought a policy when you were healthy, like you should, and you paid your premiums on time, there were contracts – even before it became the [ACA] rule — in which the insurer was not allowed to single you out for premium increases if you got a chronic condition. This is a big country, and there are sleazy companies in it, and some of them were doing that. But that kind of provision – called ‘guaranteed renewability at class average rates’ – did provide people multi-period protection. So there are ways of doing that, other than the method which is now in the [ACA] law, which is much more of a brute force method. It is saying, “You have to have insurance every year. So you can never not have multi-period protection.” But, again, the enforceability of that, to my mind, is very much in doubt.
Knowledge@Wharton: Mark, the debate about health care reform has been highly polarized between those people who think it is absolutely going down the wrong street, and devout loyalists who say this is the greatest thing and it can be tweaked to be fixed. If you were the czar of the health care industry, what would you have designed? Can we fix this or should it just be thrown out and redone, given all of these challenges that even you outline?
Pauly: Part of the problem is that the political process does not like to admit mistakes. Because of the way that the legislation was passed at the last minute, they should have written provisions into it saying that “We are going to change this as we go along and learn things” because, as it turned out, that’s actually what they ended up doing. But just as in medicine, where doctors have to pretend that they know what they are doing, apparently politicians have to do the same. Economists do not have to do that.
If it was up to me, I would have focused on what I already have [suggested] is the most important benefit of the law, [which is] making sure that you get decent insurance to low income people.
A lot of the other regulatory paraphernalia has been added on to it, and in particular the regulation of insurance premiums requiring them to be high for low risks and young people. That could have been put on the back burner for a while until you could see whether you were able to succeed in covering the bulk of the low-income uninsured. And then we could kind of mop up the stragglers. There are a few high-income people who do not have health insurance, more than you would think – like millions. But we can pick up those Evel Knievels of health insurance later without necessarily adding all this consternation that was added because of the ponderous regulatory apparatus that was constructed.
Knowledge@Wharton: Howard, perhaps you would like to talk about what changes you might make in the Flood Insurance Reform Act?
Kunreuther: I think the Flood Insurance Reform Act – as it was passed in July – was a major step forward in terms of putting on the table that premiums had to reflect risk. It was the first time – certainly in the property area – that it has ever been done by Congress. So that was a very positive step.
What I think we would have done a bit differently is that we would have tried – I would have suggested — that vouchers be instituted initially for everyone whose premiums might have gone up. Low-income people certainly would need a voucher. There was a provision of having the rates increase gradually over a five-year period. But it really discouraged people from taking steps to mitigate their homes and adapt their homes initially, because the premiums would be highly subsidized for a period of time.
I would have liked to see linkage of the insurance premium with the idea that if someone actually decided they really wanted a house to avoid the next flood, they recognized that by doing that they would get a very substantial premium discount — which is what they would get. This would have been a way to encourage that. If the premiums are only going up gradually for the next five years, you find that people would not get the benefits that they deserved for mitigating because the premiums would be too low. So give them a voucher rather than necessarily ramp it up.
“There is good news and bad news about health care reform. [On the one hand], it does direct subsidies to the people who need them the most…. But it has caused consternation in the part of the population that is relatively low-risk.” –Mark Pauly
The challenge we face today – and we should put this on the table right now – is the possibility that this law is going to be modified significantly because of all the concerns that people have with their premiums. There is new legislation introduced in Congress, as we speak, to actually change that, including one [proposal] by Representative Maxine Waters who was part of the Biggert-Waters Act. [Waters] feels that this Act may not be an appropriate one because her constituents are now complaining about the fact that their premiums are too high. So we have to deal with this affordability issue as quickly as we can.
Knowledge@Wharton: Mark, you referred earlier to the fact that insurance is going to be playing an increasingly large role in the world. Is that because of the global marketplace we have, or is it because negative events like viruses or terrorism are so much more widespread these days?
Pauly: As usual, it is a number of things. One is the global marketplace. We now potentially can suffer a loss for something that happens in Bangladesh, whereas when I only collected stamps, it did not matter [what happened] there. The other factor – and this is not particularly true for health care but true in general – is that the financial consequences of losses are now much bigger than they used to be. Those things that were kind of de minimis, you did not have to insure because they did not matter that much. Now they can take a bigger bite out of your capital. And that is going to make a difference.
The third factor, which is obvious, is that this is a much more uncertain world, whether we are talking about a future climate or whether we are talking about politics or future terrorism threats – all of which mean that it seems much riskier now than it used to be. As people’s real incomes rise – and they are still rising, contrary to some extent to economic theory – they seem to want to be protected and not just take their lumps, particularly as the lumps get bigger relative to their income. All of those are reasons why we think insurance’s role – not only in the national economy but also in the global economy – can do nothing but increase in importance. So it is really important to get it right.
Knowledge@Wharton: What are your views about that, Howard?
Kunreuther: From the vantage point of what we have seen, we really are in a new era of catastrophes. We are really facing larger losses with respect, certainly, to natural hazards. Climate change is exacerbating that problem. In the health area, as Mark pointed out, there are now major problems, and the notion of pandemics is certainly something. We face a challenge in this industry because insurers are increasingly saying that, “We are not sure we can insure this event.” For example, terrorism really highlights that point. It also highlights some of the behavioral aspects we were talking about earlier.
Prior to 9/11, the insurance industry did not exclude terrorism from coverage. They did not pay attention to terrorism. They effectively did not charge anything for that coverage on commercial policies and homeowner policies. We were a little surprised to find that out. We only found that out after 9/11; we were not aware of it beforehand. There was the 1993 aborted World Trade Center bombing that did some damage, and [the] Oklahoma City [bombing in 1995]. So there were events [involving terrorism]. They should have paid attention. But the point that really struck us was that after 9/11, [insurers] felt that this was an insurable risk. As a result, Congress was forced to pass – and maybe appropriately so for large losses – the Terrorism Risk Insurance Act. So what I think we are seeing now is a whole set of questions that are being asked: What is the role of government? What is the role of the private sector? And I think the private sector is saying more and more, “We are not sure we can really put insurance on the table in the way we have before. And if you are not going to let us charge premiums that reflect risk, we are going to be even more concerned about doing that.” That is the challenge we face in this country today, let alone all of the other problems that Mark alluded to in the global environment.
Pauly: One other aspect of health reform that could have been done better is to deal with high-risk people, not by raising the premiums of low-risk people and trying to persuade them to make that transfer, but instead, presumably by reasoning that we care about high-risk people because we all care about helping out high-risk people. It should be funded by general federal tax revenues, not by charging an extra premium on insurance purchased by just a small fraction of the population who are, of course, going to have to care for me in my old age anyway.
Knowledge@Wharton: A great deal of attention has been paid to the functioning of the healthcare.gov website. Are there any lessons from behavioral economics that should have been applied to the design or functioning of that site, but were not properly applied? For example, did the website assume too much knowledge about risk among consumers or too much knowledge about the complexities involved in insurance? Or might it have been better to have given people less information right away and then roll out the website more slowly?
Pauly: I think the main problem was that it was designed to be all encompassing and perfect because the politicians believed that they could do this better than individual insurers. And, as it turned out, they couldn’t. They couldn’t even get the software to work. So it never got around to whether consumers would make mistakes or not. They couldn’t even log on to make a mistake or make a wise choice. That was part and parcel the problem.
A simpler version would have been to use the website to allow consumers to see what their options are and then do what had been previously done, which was to find another website for your favorite insurer and go to that insurer and sign up directly there. But there were some good reasons why the government wanted it to be a seamless experience. They just were incapable of pulling it off. I think, in large part, it is because they underestimated the complexity of the most misunderstood industry, and how complicated it actually is to get all these pieces to work together.
Knowledge@Wharton: Howard, how can we link the principles of risk-based premiums and equity and affordability, so that Congress will preserve the current piece of legislation which you said is being threatened?
Kunreuther: At the Wharton Risk Center, we have been paying a great deal of attention trying to make sure that the Flood Insurance Reform Act of 2012 is preserved. And we can see a big concern on the part of people – of Congressional legislators – wanting to change that. What we would like to try to do is to bring these two principles – risk-based premiums and means-tested vouchers – into confluence so that we could not only save the government money, but also make the homeowners and people who are living in these areas feel that they can afford these premiums.
The idea is a relatively simple one. We would say that if it turns out a person is going to get a voucher because his or her premiums are extraordinarily high – either because new maps have been drawn or simply because the subsidized rates are taken away from them – we would say that you can have [that] voucher. We’ll also give you a loan to help you make your house safer – to elevate your home – but that, as a condition of the voucher, you have to mitigate. You have to elevate [your property]. It is a carrot and stick approach that we are pushing forward…. Our feeling basically is – and we have data to support this – that the government will actually be paying much lower vouchers because their insurance premiums would go way down. Homeowners would also find that they are better off because they are in a position where they have a much safer house.