Answers to Chapter 10 Review Questions 10.1. Explain why peak end evaluation causes duration neglect. With peak end evaluation an event is remembered solely according to instant utility at particular points during the event the peaks, troughs and end. So, two events where instant utility is the same at these points will be remembered equally even if the duration was very different. This can lead to somewhat perverse possibilities. Some examples illustrate. The textbook discusses an event (someone putting their hand in water) where instant utility at the trough is the same but is different at the end. The longer procedure is remembered as better because the end point is better. This seems perverse given that the longer event replicates the shorter event, and then adds on some more pain. Another possibility is two events that give the same instant utility at the end but differ at the peak. For example, imagine a vacation that lasts for two weeks and is very pleasant. Compare this with a vacation that lasts only one week but had a really pleasant moment after the third day. The shorter vacation may be remembered as better even though the longer vacation is far more likely to have given greater total utility. 10.2. Should a person choose the thing that maximizes decision utility, total utility or remembered utility? There is no answer to this question, only opinions! In the case of short events one can argue that remembered utility is what we should focus on. For example, consider medical procedures that will be very painful but only last an hour. It is probably best that patients go away with the best remembered utility even if the procedure is actually more painful. Or consider concerts that will be pleasurable but only last a few hours. It is probably best that concert goers leave with memories that will last a lifetime. The longer the event the more difficult it is to ignore total utility. For example, consider medical procedures like a hip replacement that will cause pain and inconvenience for months. We might want to choose procedures that minimize pain during this period rather than procedures that will be remembered as better. Or consider someone s choice of job. They will likely be in the job for a few years and so
they might want to choose a job they enjoy while doing it rather than a job that they remember fondly. There are difficult ethical dilemmas here, particularly when we think from a policy perspective. Fortunately, however, decision, total and remembered utility will coincide in most instances. Patients are most likely to remember well the procedure that minimized pain. Workers are mostly likely to remember fondly jobs that they enjoyed while doing them. 10.3. If a person has done something before should we expect that remembered utility equals decision utility? Decision utility determines choice. If we see Ian choosing to go on vacation to Italy then we infer that his decision utility from going to Italy was higher than that from going to Florida, China, or anywhere else he could have gone. If Ian has never been to Italy, Florida or China then his choice is clearly made on beliefs about what he can expect on vacation. His choice might be influenced by a friend s advice, books that he has read, pictures he has seen, reviews of the hotels it plans to stay at etc. Now suppose that Ian is well traveled and has been to Italy, Florida and China. In principle Ian could now make his choice on the basis of remembered utility. So, his choice to go to Italy would suggest his remembered utility from being in Italy is highest. In this case decision utility and remembered utility would be equal. This, however, seems too extreme. Past experience will likely influence Ian s choice but be just one more thing to help inform his decision. The recommendations of friends and so on can still influence his choice. For this reason remembered utility and decision utility can easily differ. 10.4. Why do you think a high national output does not translate to higher levels of life satisfaction? What are the implications of this for how we measure economic performance? We have seen evidence that happiness is relative. Those with higher relative income feel more satisfied but those with lower relative income fell less satisfied. An increase in national output may do nothing to change the relative rankings of income. It makes everyone richer but still leaves those with a lower relative income feeling less satisfied. High national output does not, therefore, necessarily translate into increased life satisfaction.
Indeed, a high national output may increase inequality and lower average levels of satisfaction. For example, an increase in national output that makes five percent of the population a lot better off while leaving 95 percent of the population no better off is unlikely to increase overall life satisfaction. National output can end up in the hands of relatively few while life satisfaction is about the population as a whole. We have also know that money is not all people value. People value good health, a secure job, good family life, free time etc. A lot of these things money cannot buy. In fact, the race to earn as much money as possible may get in the way of achieving these things. A society where everyone works every day from 8am to 8pm would produce a huge national output but not much life satisfaction. Measures of economic performance should take account of life satisfaction. Output, in of itself, is useless without satisfaction. This may lead to new measures of economic performance that take into account measures of life satisfaction. We may also want to give increased weight to existing measures of things like unemployment and inequality. 10.5. Suppose you are taking a course in behavioral economics that involves doing three assignments. Would you rather that the course organizer gives staggered deadlines for handing in the work, or that you merely need to hand in the work before the end of the course? Why? The requirement that you hand in the work before the end of the course seems optimal. It gives you maximum flexibility to organize your schedule as you wish. You can set self imposed deadlines. But, we have seen that some people are poor at sticking to plans because of present bias. In this case the staggered deadlines might sound like a good idea. This, therefore, is a personal choice. If you can stick to plans and avoid present bias then you should maximize flexibility. If you have present bias then go for the staggered deadlines. Just remember that you may be naïve about your present bias!
10.6. How can we take account of adaption when calculating QALYs? Currently QALYs are calculated by surveying people about things they have never experienced or experienced in the past. We know that adaption and projection bias makes this problematic. Those that have never experience something will underestimate adaption. Those that have experienced it will likely underestimate how much they have adapted. How to avoid this? One option is to survey people over time as they experience things. This can give us a more accurate measure of quality of life and also tell us about the rate with which people do adapt. So, what s the problem? In this case we survey a person over time about their personal experience. This takes account of adaption but makes it harder to compare across possible outcomes because the person only experiences one outcome. Ultimately, it is the cross outcome comparison that we need to use to decide which procedures are most effective. A mix of different survey methods are, therefore, probably needed to measure QALYs in a way that does account for adaption. 10.7. Given adaption and projection bias, are patients informed enough to decide what medical treatments are best for them? We have seen lots of reasons why patients are not the best placed to decide on medical treatments. In terms of adaption and projection bias patients will struggle to equate predictions on future health into a realistic prediction on life satisfaction. They will underestimate how much they can adapt to changes. It is also questionable how much choice patients would want to have. We have seen that people often do not like complicated choices and medical treatment involves some very complicated choices. Patient choice is, therefore, something that we may not want to value too highly. Keep in mind, however, that people do value control over what is happening to them. And so patients should feel some control over their treatment.
10.8. Explain why projection bias is different to present bias. How does this show up in people s decision to save for retirement? Projection bias is where a person fails to recognize how they will adapt to a future standard of living. For example, in his 20s David was earning $30,000 a year and thought that he would be happy to retire with a similar standard of living to what he currently has. So he saved little for retirement. Over time Dave s earnings and standard of living increased. At each point in time Dave thinks that he would be happy to retire with a similar standard of living to what he currently has. So he saves little for retirement. Just before retirement he was earning $100,000 a year. Is David happy to go back the kind of standard of living $30,000 a year buys? No. David has failed to take account of adaption. He has failed to take account of the fact he will become habituated to the standard of living he has had. This means he does not save enough for retirement. Worse than that, he saves less than he planned to do. When he was in his 20s he thought when I m earning $50,000 I will save most of it. But, by the time he is earning $50,000 he has got used to a higher standard of living and so actually spends most of it. Present bias is very different to projection bias. Present bias allows David to have an optimal plan for the future. This is a plan that would work. It does take into projection bias and adaption to a higher standard of living. The problem with present bias is that David cannot stick to the plan because he puts things off until tomorrow. He will start saving next year, and then the year after. Before long it is retirement and he has saved little. With projection bias David does not stick to the plan because it is a bad plan. Because it is a bad plan he constantly changes it. With present day bias David has a great plan he just cannot implement it. The end result, however, is the same not enough saving for retirement. Present day bias is, however, potentially easier to fix. David just needs to find a way of pre committing to his plan.
10.9. Should we take account of adaption, or the lack of adaption to unemployment, when calculating the unemployment inflation trade off? What affect would this have? We have seen that people do not appear to adapt to unemployment. This needs to be taken account of when we measure the costs of unemployment, and the costs of longterm unemployment. In terms of the inflation, unemployment trade off it suggests unemployment should get a relatively high weight. Unemployment appears very bad in terms of life satisfaction and this should be recognized. 10.10. Should Europe change policy, and not require firms to include the sales tax when advertising price? The immediate effect of not including the sales tax would be an increase in sales. If the taxes are distortionary then this could be a good thing. The government can raise revenue without distorting the economy as much as previously. Over time, however, we might expect consumers to become used to the omission of taxes and bring back their spending to normal. It is not clear, therefore, how big a long term effect this would have.