How Behavioral Economics Could Solve America’s Health Care Woes
Dr. Kevin Volpp, a Wharton health care management professor and director of the Center for Health Incentives and Behavioral Economics, part of the Leonard Davis Institute of Health Economics at the University of Pennsylvania, believes that, despite a House of Representatives vote to repeal and replace it, America isn’t done dealing with the Affordable Care Act. He and Dartmouth College economics professor Jonathan Skinner wrote an article in the Journal of the American Medical Association that focuses on the application of behavioral economics to the complex problem of creating a health insurance system that works for all citizens. They propose four general principles that should be part of any effort at health care reform. Volpp appeared recently on the Knowledge@Wharton show on Wharton Business Radio on SiriusXM channel 111 to discuss the article and why health care continues to be such a vexing issue for the United States.
An edited transcript of the conversation follows.
Knowledge@Wharton: What’s your view on the Affordable Care Act?
Kevin Volpp: The Affordable Care Act has had some noteworthy successes, and it leaves some areas with room for improvement. In terms of the notable successes, I think the biggest is it’s successfully covered more than 20 million people. That has a lot of important implications in terms of reducing inequality, reducing the risk of bankruptcy among people in those families. It’s also been widely credited with reducing the deficit based on how it was funded. I think the other key thing is that it’s made health insurance coverage more comprehensive. There are some mandates that are part of the Affordable Care Act — for example, removing pre-existing conditions as a reason people can be denied coverage or required to pay more, and allowing people to have their kid still part of their coverage until age 26. In many people’s minds, those are important provisions in their own right.
Knowledge@Wharton: Why did you and your co-author, Jonathan Skinner, look at this from the viewpoint of behavioral economics?
Volpp: I think a big part of the Affordable Care Act, in terms of increasing coverage, was to think about the underlying incentives of why people are not buying coverage. For many people, it frankly just comes down to weighing the expected costs and benefits in the short term, and concluding that the short-term benefits to people who weren’t buying coverage were less than what their coverage would have cost.
The way the Affordable Care Act tried to deal with that was by subsidizing the cost of coverage by providing people who are low income with fairly generous subsidies. That’s a carrot-type incentive. In addition, they included a stick-type incentive in the form of an individual mandate, whereby people were required to buy insurance. If they didn’t buy insurance, they’d have to pay a financial penalty. One thing we could critique is that this mandate probably wasn’t strong enough. It started out at about $200. Eventually it became about $700. That’s much less than the cost of the cheapest plan. An individual could quite rationally conclude, “I’m willing to pay a $700 penalty. I’m not willing to pay $4,000 or more for my coverage.”
The other issue was that there’s a time delay in terms of people paying this penalty. Often, it would not kick in until you paid your taxes roughly a year and a half later. That isn’t so desirable from a behavioral standpoint because people very much focus on immediate consequences of their actions.
It’s an interesting issue from a philosophical standpoint. The individual mandate was originally an idea hatched by Republican think tanks. It really appealed to a lot of people from the sense of a personal responsibility here that could apply to each citizen, whereby if we’re going to have an individual insurance market function, we have to have everybody in the pool. The underlying challenge is what economists call an adverse selection problem, whereby individuals know more about their health than the insurer does, and people who are above-average risk will be more likely to buy coverage for a given price than people who are below-average risk. Over time, that leads to premiums climbing faster and faster because the lower-risk people differentially decide not to buy coverage. In essence, that’s what we’ve been seeing as the Achilles’ heel of the health insurance exchanges under the Affordable Care Act, that over time in certain markets lower-risk people are choosing not to buy insurance at higher and higher rates, which drives up the average premium more than might be seen as sustainable.
Knowledge@Wharton: How do you think that problem needs to be approached so you can have more healthy people involved in the plans without enforcing a fine?
Volpp: I think there are a number of ways this could be approached. If I were starting from scratch, I would consider having a stronger individual mandate and making that similar in price to the lowest-price plan. That way, everybody would have to do it. The issue becomes one of affordability and whether the carrot incentive is a subsidy or a tax credit. Whatever we might label it, those would have to be large enough so that lower-income families could actually afford the plan.
“One of the problems the Affordable Care Act has not solved, and which a lot of people would love to see solved, is the complexity of the health insurance system.”
I think there are other ways we could try to approach this that would increase the likelihood that people who have coverage will continue coverage, even if the individual mandate goes away. One of the principles we talk about in the article is the behavioral economic principle around inertia, and this underlies a lot of the power of default. In essence, there’s a status-quo bias that a lot of people experience. When they have something, it’s much harder to give it up than it is to acquire something fresh. If you have people opt out of something, typically enrollment rates are in the 90% range. If people have to opt into something, enrollment rates are often in the 10% range. The way you might think about this in context is we could have a form of automatic re-enrollment in a plan, where you could peg the plan to the price of the lowest subsidy or tax credit that’s being offered. You could basically say, “Here, you automatically have this plan. It’s free and you can opt out, but then you’re going to lose your coverage and the subsidy.” But I think very few people thinking rationally would do that. Why would you turn down a free health insurance plan?
On the other hand, if you think about this in an opt-in framework, which is the way it has been structured until now, this information is all out there but you have to find it. You have to make the effort to go sign up, you have to go through that process, you have to sift through all the information, and you can see why that naturally would lead to a lower enrollment rate.
Knowledge@Wharton: It’s a safe assumption that most Americans want it to be as simplified a process as possible. They don’t want to have to spend several days working through a website to find the right health coverage for them. They would like it put it on a silver platter for them.
Volpp: I think that’s exactly right. One of the problems the Affordable Care Act has not solved, and which a lot of people would love to see solved, is the complexity of the health insurance system. It’s tremendously challenging for people to navigate. Many people don’t really understand their health insurance benefits very well. They get a guide to benefits that might be 70 pages long, has a lot of detail. That’s something which I think should be an important priority going forward.
We’ve done some work with Humana on trying to think through that. We actually designed a new health plan with them that’s called Humana Simplicity. It’s simple in a couple of important ways. One is that people don’t understand co-insurance deductibles or maximum out-of-pocket limits very well. They have a lot of difficulty calculating the expected costs of their care. We designed a plan that’s copayments only and has six categories of copayments. Everything is distilled down into those categories. Instead of having a 70-page guide to benefits, everything fits on a page and a half.
Knowledge@Wharton: Can we ever get to a simplified system like that across the entire body of health care?
Volpp: It’s a really good question. If it were easy, then obviously, this would have happened by now. I’m seeing more and more efforts to use design-type thinking and really move towards thinking about how to put the patient at the center of things. For example, there is an interesting initiative in some places that are being built as new facilities to eliminate waiting rooms. But in most facilities, a lot of the space being built is waiting rooms because there’s a lot of time waiting. If you can eliminate those waits, that’s much better for the patient. It requires a whole different model in terms of flow to accommodate that.
“It’s really quite striking that before the Affordable Care Act came into play, we had about 45 million people in the U.S. who didn’t have coverage.”
Knowledge@Wharton: One of the behavioral economic principles that you talk about in the article is instant gratification. People want that immediate feeling of having something fantastic right at their fingertips.
Volpp: Instant gratification is really one of the deepest human tendencies. It probably dates from evolutionary days, when people were running away from saber-tooth tigers. But it is important as you design these programs to think about how people’s desire for immediate gratification influences their choices. In the context of the Affordable Care Act, this has two important implications. One is that from an economic standpoint it makes sense for people to just have coverage for catastrophic events. That’s the lowest-priced coverage. It protects you against financial disasters. But that doesn’t do so well in terms of immediate gratification because people see themselves paying these premiums and feeling like they’re not getting anything for it. I think if you were an economist designing a plan, you’d say, “Let’s put everyone in a low-cost catastrophic-only coverage plan.” That’s the most affordable way to do this, and I think that makes sense from the standpoint of insuring the population against catastrophic risk. But at the same time, we have to recognize that won’t quite satisfy a lot of people because even though they’re getting a lot in terms of financial security, that’s not going to be tangible.
I think the other issue that’s relevant, in terms of thinking about the new proposals that are being deliberated now, is the idea of having a continuous coverage requirement in lieu of the mandate. The idea is that if people don’t maintain continuous coverage, they would face a financial penalty of 30% of the premium when they sign up for coverage later. I think the problem with that is it fails the immediate gratification test. If we’re having trouble getting young, healthy people to sign up now because of the penalty that exists, a penalty that would kick in at some future point is very unlikely to work because that’s not going to be salient to people today.
Knowledge@Wharton: The subsidies are important to be able to have this type of program going forward. But a lot of people will say, “This is an unbelievable cost that the American public is having to pay for.” How do we get past that issue?
Volpp: That’s a very deep question that relates to what kind of country do we want to be in terms of how we care for those who are less able to care for themselves? When we compare ourselves with various Western European countries, Japan, Australia, New Zealand, Israel, they’ve all figured out how to do this. In some sense, the only way to do it in terms of lower-income people is to have cross-subsidization from higher-income people who can support that. It’s really quite striking that before the Affordable Care Act came into play, we had about 45 million people in the U.S. who didn’t have coverage. We still have 20 million to 25 million. I think it does take its toll in various ways to have a lot of uninsured people in the U.S. But there’s no question that if we are going to provide coverage to more people who can’t afford it on their own, it will cost money.
Knowledge@Wharton: You talk about the fact that other countries have figured this out. Is the fact that our system is so expansive and has so many tentacles the main reasons why we are not able to be able to implement something like what you see in those countries?
Volpp: I think the single biggest reason is that health care in the U.S. is significantly more expensive than anywhere else. It’s really quite striking. If you look at a graph of how spending in the U.S. has changed relative to life expectancy, and you compare us with the 26 countries we’re commonly compared with, we all start at the same place. On this graph, let’s imagine you have life expectancy on the Y axis, spending on the X axis. The U.S. basically goes up and then curves to the right, and we sort of stand by ourselves in terms of how much we spend. But life expectancy has actually improved a lot more in the other countries. So, we’re in this strange situation where we rank first in spending by a wide margin and rank 26th in life expectancy. You don’t need to be a business strategist to conclude that that’s not a very good return on investment in terms of the money that we spend on health services. If we spent 30% less per capita, then it would be much easier to provide coverage to everybody.
“If we spent 30% less per capita, then it would be much easier to provide coverage to everybody.”
Knowledge@Wharton: With so much of this in flux right now, this becomes a watch and wait for you and others in this realm, correct?
Volpp: Yes, there’s a lot of concern that the proverbial baby will be thrown out with the bathwater. There are a lot of good things in the Affordable Care Act. I think the biggest remaining challenge is this issue of higher-risk people buying insurance at higher rates, lower-risk people buying no insurance, and that leading to premiums increasing more in some markets than is sustainable. That is a problem that could be solved in isolation, but unraveling the whole system creates a lot of uncertainty for all the insurers. And that in itself becomes a problem because a lot of them may then pull out of these markets, and then we don’t have a market.
Ideally, there would be a bipartisan effort that would say, let’s build on what’s been done. The Affordable Care Act is actually a market-based solution. It’s very far away from what Democrats have proposed in the past. It’s not at all like a single-payer kind of system. It’s saying, “Let’s create these marketplaces, the individual insurance market. Let’s try to figure out how to make that work better. We have to provide subsidies to people that make the insurance coverage affordable.” I think that’s common to how people across the political aisle would think about this. Where there are differences now are in things like how big are these subsidies or tax credits, and how strongly do we feel we can help people to participate? Is there an individual mandate? Is there not an individual mandate? If we don’t have the individual mandate, what are the alternatives? As I mentioned, I don’t think this continuous coverage requirement is going to work. I think it fails the behavioral test in terms of people making decisions based on immediate costs and benefits, as opposed to those they would only face at some point down the road.
Knowledge@Wharton: Going back to the inertia principle that you talked about before, with the baby boomer generation getting older and needing advanced health care, it becomes even more important to have the younger generation to level the field.
“Subsidies alone probably will not work in terms of getting sufficient people to enroll in these marketplaces.”
Volpp: I think that’s fair. Health care has always been an 80-20 proposition. So, 20% of people end up being responsible for probably about 80% of the costs. That is one of the underlying challenges here, particularly as you think about trying to make these individual marketplaces work. If you have a few very high-cost people, that obviously raises the average cost for everybody. One approach that Jonathan Skinner and I suggest in the article is really thinking a bit differently about the high-risk pools. Various groups have talked about the idea of pulling out the higher-risk people. If you pull out the higher-risk people, then this problem of the markets unraveling and the premiums progressively going up becomes much less of an issue because the higher-risk people are then no longer a part of that average cost. If we pull out the higher-risk people, you might actually see premiums go down for lower-risk people. Then where do we put the higher-risk people? One challenge is creating a separate pool for them. There’s a long history of those kind of pools not working very well. They tend to be underfunded. There’s not generally enough political will to support them. A good example of how we’ve done this well is what’s happened with people who are in end-stage renal dialysis, who basically just got folded into the Medicare program many decades ago. It relates to the issue of salience of taxation. It’s very hard to raise money separately for any kind of new, significant programmatic initiative. But if we just folded those high-risk people into Medicare, they’d be absorbed, they’d be a small portion of the Medicare population, and their average cost would be much lower.
Knowledge@Wharton: If you’re taking the high-risk people out, you have a second group with the lower-risk people. Then there isn’t as much of the economic burden on that group compared with keeping high-risk people in there.
Volpp: That’s absolutely right. I think if we pulled out the 20% highest-risk people from the health insurance exchanges and we put them in Medicare, the average cost for the 80% of people who remained would be much lower than it is now. That makes insurance much more affordable. It makes it much more likely that more, lower-risk people would choose to buy insurance. And then you’d have a much better-functioning market.
Knowledge@Wharton: Are we venturing into new ground in terms of how we’re thinking about these challenges?
Volpp: I think a lot of the challenges are well known. Different groups trying to solve those challenges differ based on their beliefs, based on their political philosophy, and that’s where I think the mandate tends to get tripped up. There’s generally an acknowledgement that subsidies alone may not be enough to get people to sign up. There’s a well-documented literature in behavioral economics that carrots are weaker than sticks. People tend to be very loss-averse. Subsidies alone probably will not work in terms of getting sufficient people to enroll in these marketplaces. It’s possible that some combination of these approaches would help. But I think you still have this fundamental problem that in the current setting you have people having to opt in, you have subsidies only, you have a delayed penalty in terms of the continuous coverage requirement, and it just may not be salient enough for that young, healthy person who’s thinking, “I don’t want to pay this premium now for something that might or might not help me down the road.”