What natural disasters, and ostriches, can teach you about risk and decision making
A lack of awareness and preparation is a big part of why hurricanes, floods and other calamities leave such chaos behind. But what if we could change the way we think about risk? That’s the subject of The Ostrich Paradox: Why We Underprepare for Disasters, a new book by Wharton professors Howard Kunreuther and Robert Meyer , who co-direct the Wharton Risk Management and Decision Processes Center . Kunreuther, a professor of operations, information and decisions, and Meyer, a marketing professor, stopped by Knowledge@Wharton to talk about the book and their solution for helping policymakers, governments and individuals to create more prepared communities.
An edited transcript of the conversation follows.
Knowledge@Wharton: I wanted to ask you about the title of the book. I assumed it meant that we should try to avoid ostrich-like behavior, that we shouldn’t stick our heads in the sand when it comes to disasters. But it turns out, that’s not quite the case. Could you tell us more?
Robert Meyer: We were very much interested in writing about why it is that we can’t seem to get a hold on disasters. Even though scientific abilities to forecast hurricanes and so forth have grown considerably over the years, that hasn’t really been matched with any noticeable decrease in the costs, both in terms of lives and monetary losses, from natural disasters or man-made disasters. Often, the reason that’s given whenever that happens, we say, “Well, people are putting their heads in the sand. They’re ostriches.” That’s sort of the old cartoon notion of what an ostrich is. When threat is coming, a lion is coming, they dig their head in the sand and pretend it doesn’t exist.
It turns out that ostriches have been given a bum rap for that over the years, because ostriches are incredibly good at dealing with risk. They have enormous limitations. They can’t fly. But nature has allowed them to overcome that by having enormous ground speed and all sorts of risk-avoidance strategies.
It then occurred to us that therein might be a key to which we can get people to be better at preparing for risk. Rather than being less like ostriches, they actually need to become more like ostriches. To first start off and say, “What are the limitations that we have? What are the psychological limitations that prevent us from being better at preparing for disasters?” And once we understand what those are, think about ways in which we can better adapt to those limitations so we’re more like ostriches rather than less. Hence the paradox.
Howard Kunreuther: We do bury our heads in the sand often. In some sense, we felt that by using the ostrich as an analogy, we might be in a position for people to pay attention. I think the real challenge is for people to pay attention before something happens, rather than afterwards. Ostriches are very good at doing that, and we hope human beings can do a better job than they have up until now.
Knowledge@Wharton: The book starts with a really interesting example from Galveston, Texas. It was two hurricanes, which occurred about 100 years apart. But there was the same reaction both times, even though technology had advanced significantly when the second hurricane hit; there was still this same under-response to the disaster. The book is full of stories like this. Was there one in particular that inspired it?
Meyer: For me, the thing that inspired it is a story that is not in the book. I think it was 2008. I was visiting New Orleans and decided to take a drive along the Mississippi Gulf Coast. I was driving along near the town of Pass Christian, Mississippi, which is an area that had been devastated by Hurricane Katrina, and I was noticing how quickly the place had recovered. There were a lot of nice grass fields. I noticed that in one of the fields was an ATM machine that was just sitting by itself. I looked at that and pulled over. I said, “What’s an ATM machine doing out in the middle of the field?” I kind of walked up to it and began looking around. Then I discovered afterwards that there used to be a shopping center right on the coast that had been completely destroyed by Hurricane Katrina. The only thing that was left was the ATM machine.
It turned out that this shopping center was built on the exact same spot where a condominium complex used to be in 1969. Twenty-three people died when the condominium complex was blown away by a hurricane. I contacted the person who owned the land of the shopping center and asked, “Did you know that this site had this hurricane history?” He said, “Yeah, but it had been so many years since there’d been a storm, we didn’t think it was under much risk.” Then I said, “What do you plan to do with the property?” And he said, “Well, we’re thinking about selling it and hope to build up a condominium complex here again.”
I was sitting there and thinking, “There’s got to be a story there.” Why is it that we don’t learn from experience? Why is it that we’re not better at thinking ahead in terms of what the consequences are of threats like this? What are all the different types of psychological factors that cause us to be so poor at preparing for disasters?
Kunreuther: To follow up on a story in the book that actually did start it off, is the story of Glenda Moore, who was in a position where she had to figure out what decision she was going to make after Hurricane Sandy. I think all of us were very cognizant of the challenges we face after a storm occurs, in the sense that we know there are things we should do and things that we might not do. But we don’t really know them as well as we should. Glenda’s first concern was to be with her husband, who had left [Staten Island] and was in Brooklyn, and she made a set of decisions without carefully thinking them out.
Our concern is that, to a large extent, we react to situations at the moment without necessarily doing an analysis of all of the possible dangers. The communication mechanisms are sometimes good, but we sometimes don’t hear them. [Moore] made some decisions to leave that resulted in her losing her children. There was a whole set of media attention given to that. Our feeling was, it was a way to start the book, to point out the fact that we would like everyone to reflect on what could happen after a disaster and do some steps beforehand to prepare. Also, to listen to the kinds of things that one is hearing and try to get the right kind of information so one doesn’t do the things that could result in tragedy.
Knowledge@Wharton: You outline six biases that may lead to some of this under-preparation for disasters, and you’ve said that perhaps the most dangerous one of these may be optimism. I thought that was interesting because we usually think of that as a good thing.
Meyer: Absolutely. I think it is a good thing to have in life. We’ve been fortunately endowed with an optimistic outlook. The reason we have these things is they serve a good, functional purpose. Most of the time, in most of the decisions we face on a day-to-day basis, they basically serve us well. However, it’s also the case that when we suddenly are in a situation of facing very rare events — decisions we don’t have to make on a day-to-day basis — that’s when you start applying these same heuristics, and all of a sudden all sorts of stuff goes wrong. Optimism is a great case of that. It is good to look on the bright side of life. However, when you’re talking about something that might happen to you that might mean the end of your life, then being optimistic is not necessarily the best thing to do. One of the things we talk about is the ways in which people become erroneously optimistic in a way that is harmful.
Kunreuther: One of the things with optimism is that we try to find ways to defend our optimism. We don’t really want to think about a flood or a hurricane. We’re optimistic because we are living in an area that we say is a great place for us to live. We want to keep that image alive, so we say, “The chances of anything happening are so low that we’re not going to really think about it. It won’t happen to us. It may happen to others. It’s below our threshold level of concern. We’re not going to worry.”
We keep that optimistic bias in the sense of trying to avoid having to think about things that we probably should be thinking about beforehand. Our whole idea in bringing that up, along with the five other biases, is to say that they’re all connected in some sense. We have to try and figure out ways to let people know that these are things we all do. Then we can think about ways to improve the decisions by trying to avoid them in the future.
Knowledge@Wharton: Your method of improving this is called the behavior risk audit. This involves looking at each of these biases, turning them on their heads and figuring out how to create policies that address these biases and maybe get around them. Can you talk about how people could do this?
Kunreuther: Let me illustrate with one, and Bob may pick another one. Myopia is one of the biases that we have. We all have short-term horizons. We want to get immediate returns. If there are things that we can do for the long term, we often find they are very prohibitively expensive. Let’s take an example of having to make our house safer against a flood or hurricane. There’s a lot of cost to doing that. You could elevate your house, but that’s very costly. You could maybe flood-proof it. People will say, “What are the benefits that I’m going to get from that in the next period?” They’ll be very reluctant to put in the money because they say the benefits are going to be very short run. And they’re right. If you’re going to get a short-run benefit like a reduction in your insurance premium, you’ll say, “Well, I’m not going to get enough to pay for that expense.”
We would recommend that instead of thinking about just the long term with respect to this, there are two things that one can do. One is, you might give a person a loan to help them out and spread the cost over time. The other thing to deal with is the “it won’t happen to me.” Instead of saying, “It’s going to be a one in 100 chance of a flood occurring next year,” stretch the time and say, “Think about the fact that there might be a hurricane in the next 30 years, and that likelihood is greater than one in four, or one in five.” Then, people will think about the long term and maybe decide that they can take some action.
Meyer: Another one of the biases we talk about is simplification. The idea there is that most people don’t want to consider lots of different factors when making decisions. One feature of this is a thing called the single-action bias. When you’re faced with a problem, as soon as you take one action to try and solve it, there’s a tendency for your brain to go, “Good. There was the problem. I took some action. Problem solved.”
In a lot of walks of life, that’s kind of an OK thing to do. But think about it in the context of preparing for natural disasters when, in fact, there are large numbers of things that you need to do. You’re trying to build a safer house. In order to make the house structurally sound, there are lots of different things that you have to do. Historically, the way in which agencies like the National Oceanic and Atmospheric Administration and the Federal Emergency Management Agency have encouraged people to prepare for disasters is to give them checklists. The checklists will have 60 items on them. They will say, “Make sure you do this with your dog.” The single-action bias will say that people are going to ignore most of those things. But they are going to see the list. They are going to be aware that there’s a threat. The problem is that they’re going to go down this list randomly and take care of the dog or whatever. And once they’ve done that, that’s the end of the list and they feel sort of prepared for the disaster when they’re really not.
One of the remedies for that is to say, now that we know people simplify and tend to focus on one action, don’t give them a massive checklist. Say, “If you’re going to do one thing, here’s the one thing to do. And once you’ve done that, here’s the second thing to do.” You have to walk people through that. That’s going to be a much more effective way of getting people to prepare.
Knowledge@Wharton: Do you know of examples of agencies or governments or companies that are doing this? Or do they all need to get a copy of the book?
Kunreuther: I’ll give one example and relate it to what I said a few moments ago. FEMA is very concerned with communicating risk. They have actually changed the way they’re presenting information on the flood hazard. They used to talk about a 100-year flood. They’re now telling people, “Think about the next 30 years and what could happen to you. You should consider buying insurance for next year, even though you are thinking about the fact that there’s a low probability for a disaster to occur.” So, they are trying to take that seriously. But they have to do this in such a way that they communicate so people will at least read that, which is a challenge in and of itself.
Knowledge@Wharton: Could a family or an individual use this behavior risk audit to prepare for disasters, or is it more for policymakers or companies?
Meyer: I think it absolutely can be used for individual households. Often, companies are fairly good at thinking through all the details and having well-developed risk avoidance plans. For example, a utility company tends to be very good at this sort of thing. This book is a little more targeted at organizations that aren’t necessarily experts in terms of risk preparation. For any given family, you need to sit down and think through, “How are we as a family are thinking about risk?” These biases apply there.
Howard was giving the example that started out the book about the Glenda Moore family, where a woman hastily put her children into a van and lost the children in a hasty evacuation. We would like to think that this is a book that can help households like that make sure that those events never occur. What it does is say, “Look, when you’re faced with disasters, these are the kinds of mistakes that you’re likely to make.” It’s not because there’s something wrong with you. You’re just human. These are hard-grained, hard-wired biases that we have. Once we understand what those are, you can anticipate the kind of mistakes you’re going to make. That’s the first step in trying to avoid them.
Kunreuther: There are a series of stories that are graphic. People react to graphic stories in a way that they don’t react to just the facts. Our hope is that people will read these stories, recognize that it can happen to them and pay attention to the very last part of the book on ways to improve.
Then, we would like to use one of our biases that we hope makes this book more than just one person reading it, and that’s the herd effect. We hope that people will see this and say, “You know, this is something I’d like to talk to a few other people about. It isn’t just our situation; it may be others.” I hope it becomes a topic of conversation that we have these kinds of biases. We can all think of our own sets of activities that might reflect that. There’s a tendency, with the kinds of events we’re talking about, for people not to want to pay attention. Our hope is that the stories that we tell here will resonate enough so that people will want to actually think about them.
Knowledge@Wharton: It seems like we hear about hurricanes when they’re a day or two away, or maybe in the immediate aftermath. Then we don’t hear about them again until the next one comes along. You also talk about more long-range events, specifically about climate change. How does behavior risk audit apply to something like that, which is more of a long-term, broader risk?
Kunreuther: This is a real challenge. I think we recognize that it’s a challenge we’re facing right now because climate change is not necessarily being discussed in a way that I think we would like to see it discussed. It is a critical problem, and people may not be paying attention right now.
First of all, construct some scenarios that you begin to see what might happen, and think about how you can take steps now to avoid those scenarios. Think about the fact that it isn’t just yourselves, but there’s future generations at stake. What is this going to do to my children and my great-grandchildren? Try to recognize that there’s a tendency to say, “This is not going to happen.” It won’t happen necessarily tomorrow, but there has to be planning. And it isn’t just for the individuals. It’s for communities.
Meyer: As Howard said, I think long-run risk is very difficult. Particularly if, let’s say, you’re living in any coastal area and worried about sea-level rise. You say, “Well, scientists are forecasting that in 60 to 70 years, this area is going to be underwater.” I’m not going to be living there in 60, 70 years, so why should I particularly care about that?”
This is particularly the case for communities that are thinking about doing things like raising taxes and bond issues to put up public works projects to prevent sea-level rise. The people who are voting for them and who are going to have to pay these taxes aren’t going to be the ones who are going to be benefitting from it. People might understand, “Well, maybe we need to take care of it. But maybe we should just do this next year.” It’s easy to put off because nothing’s going to change that much. Of course, what happens is that you keep putting it off until next year, and it never gets done.
Some of the things we were thinking about within the book are ways in which you can get people to take safe actions, or have a culture of safety, in a situation where the uncertainty is on taking risk rather than not taking risk? For example, you have your city budget and one of the line items is, “Should we put multimillion-dollars into a pumping system?” That’s an add-on. Then you have to think through, “Is this a good use of the money now?”
What if it was the case that every year in the budget there’s money set aside for infrastructure improvements? But that can be removed. Now the discussion is, “Do we want to not take a safe action this year or not?” That might be enough to keep some of those items on the budget that might have otherwise been removed. The hope is that through those little sorts of steps, we might get closer to dealing with some of these extremely difficult long-run problems.
Kunreuther: You could couple what Bob is saying with the notion of, “Let’s make sure we don’t have the cost immediately on top of us.” Having a long-term loan so that you spread it over a number of years will make it more attractive. If you can show that there are benefits that are going to occur, even in the short run, maybe property values will go up. Maybe people will want to move into the area, and you can sell your home a little more easily by the fact that one recognizes the community is taking steps now to avoid this problem that might be really serious 20 or 30 years from now. I think those are problems facing coastal cities that really have to be concerned about what is going to happen to the property values when people say, “We’re not really taking the steps.”
I think if we can combine these things with economic incentives, with the default option that Bob mentioned, which is one of our notions that you have to say, “I’m not going to do this,” rather than, “I’m going to do this.” And also, to recognize one of the challenges on the political process. We have a little acronym that we occasionally use: NIMTOFF. Not in My Term of Office. If we can avoid that and recognize that there are benefits of doing this in my term of office, then I think we have a chance of at least getting people to pay attention.
Knowledge@Wharton: What about different kinds of risks, like political risks? Some of the things that I read in the book reminded me of what you heard people in the United Kingdom say around the time of Brexit. “Oh, this won’t happen. It won’t pass.” Or even around the time of the presidential election in the United States. I feel like there was maybe some myopia or optimism at play there. Do you think that the principles in the book can apply to different types of risk other than natural disasters or environmental risk?
Kunreuther: We haven’t thought explicitly about Brexit, to use that as an example. But I think there is a notion here of a feeling that people may not necessarily appreciate all the elements of a particular problem. We could say about Brexit and the election that there was a degree of optimism by people who felt that they really didn’t have to be concerned about voting in a particular way on the basis of feeling that this was a foregone conclusion. In that sense, we come back to the optimism bias.
Meyer: Certainly, you can go through those biases, and they can provide an explanation for why maybe a lot of people were very surprised by the last election. Or alternatively, half the country was not at all surprised, and the other half was extremely surprised. I think the part that was extremely surprised was influenced by things like herd-think. They went around, and the people that they talked to were people that had the same values that they did, and they became of the belief that this is the way the world is. It’s simplification. There’s a tendency to not look broadly or look for evidence that maybe there are people out there who don’t think like you do. When you take all these things together, all of a sudden there’s this groupthink, which in many cases may have led, for example, Hillary Clinton supporters to feel they didn’t need to vote and it’s such a foregone conclusion.
SOURCE: World Economic Forum