- \n
- The COVID epidemic knocked the whole world for a loop. Different nations responded to it differently, as did different regions within the United States. Did we handle it well here in America? Did we handle all<\/em> of it well? How should we go about trying to answer these questions?<\/li>\n\n\n\n
- President Trump has imposed and threatened massive tariffs on many countries. What effects will such tariffs have? What effects should we look for, and how should we measure them?<\/li>\n\n\n\n
- In July 2025, Congress, at the urging of President Trump, passed a massive tax bill that will reduce taxes a lot for some and a little for others. To make up for the lost revenue, the bill contains significant cuts to parts of the budget\u2014most saliently, to Medicaid. How should we assess the impacts of this bill, both short and long term? What effects should we measure, and how should we measure them?<\/li>\n<\/ul>\n\n\n\n
These are all complex issues. Are the tools we typically use to address them up to the task? We think they don\u2019t come close.<\/p>\n\n\n\n
The tools we have are cost-benefit analysis (at the level of public policy) and what is called rational choice theory (RCT), at the level of individual decisions. RCT regards the decisions we make as similar to casino gambles. To decide what bets to make, we need to know how much we can win or lose and how likely we are to win or lose. By multiplying possible gains and losses by their probability, we come up with the expected monetary value of each gamble. According to RCT, this is the right way to decide how to bet at the blackjack table, and it\u2019s the right way to decide where to go to college, what job to take, where to live, whether and when to marry, where to go on vacation, and so on. These real-life decisions are like casino gambles, except that what is at stake is not (just) money, but utility. Applied to public policy, this amounts to aggregating the possible benefits (and costs) of different policies, and multiplying them by the probability that those benefits (and costs) will be realized. The probability assessment is important. We could not have known for certain what our COVID policies would yield, just as we can\u2019t know for certain what tariffs and the tax bill will yield.<\/p>\n\n\n\n
RCT readily acknowledges that real-life decisions are more complex than casino gambles. Such decisions often involve considering multiple dimensions that are not as easy to compare or to quantify as gambles. (How does a high-school senior quantify the quality of the biology department at different colleges, and how does she compare the quality of the biology department to the quality of the theater program or the social life?) Nonetheless, RCT imagines that in order to make decisions we should create big spreadsheets and fill in the cells with quantitative estimates of the importance, quality, and probability of each feature of the options that matters to us. Then we should do the math and choose the winner. This approach is perhaps most explicitly exemplified in the book Thinking in Bets<\/em>, by Annie Duke.1<\/a><\/sup><\/p>\n\n\n\n
Is this how we actually operate when we face decisions, both large and small? Not so much. Researchers in psychology have spent fifty years documenting the many errors and biases people display when they make decisions. There are many popular books elucidating these errors, and two Nobel Prizes have been awarded in recognition of the significance of this research. What virtually all of this research takes for granted is that we know what rational decision making should<\/em> look like.<\/p>\n\n\n\n
RCT is what we might call the normative model of rationality. What research has shown is that RCT fails as a description of how people do<\/em> decide, but it leaves unchallenged that RCT is the model of how people should<\/em> decide. We argue that if people actually followed rational choice theory in making decisions, those decisions would be woefully inadequate\u2014even monstrous\u2014and a society filled with people who decide in this way is not a society we would want to live in.<\/p>\n\n\n\n
It should go without saying that most of us rarely, if ever, create spreadsheets when we face decisions, even when the decisions are important. Perhaps we just go with our gut, or we frame our decisions narrowly, focusing on the most obvious features of a decision and failing to consider the less obvious aspects, which may in fact turn out to be very important. Indeed, this \u201cnarrow framing\u201d lies beneath many of the errors and biases that researchers have discovered over the years. And narrow framing can lead to seriously mistaken decisions.<\/p>\n\n\n\n
We agree with RCT that narrow framing is a problem, but we suggest that RCT is itself guilty of narrow framing, and that it offers no tools for determining the appropriate framing for a decision or policy. RCT essentially frames decisions as having just two components: how good or bad might the outcome of each gamble be, and how likely is it to be that good or bad. Further, RCT frames decisions as if they can be easily quantified, and quantified on a common scale so that different aspects of a decision can be traded off against one another.<\/p>\n\n\n\n
Is this framing reasonable? In choosing where to go to college, how does one quantify the quality of instruction? How does one quantify the appeal of the social life? And how does one compare these two aspects of a decision? Yes, one can quantify the average GPA, the U.S. News & World Report<\/em> ranking, the percentage of premeds who gain admission to medical school, and the average salary of graduates five years after graduation. But is it rational to make the college decision strictly on the basis of what can be easily quantified? Moreover, RCT suggests, or at least implies, that the quantifiable components of individual decisions should also dominate our assessment of various public policies.<\/p>\n\n\n\n
We believe that our task when we make decisions is not to avoid being influenced by frames but to choose the right<\/em> frames\u2014frames that help us to evaluate all that is relevant. And choosing the right frame will depend on the purposes of our evaluation and impending decision. RCT tells us virtually nothing about how to choose the right frame. And if we cannot do that, it is extremely unlikely that RCT will lead us where we need to be.<\/p>\n\n\n\n
A striking example of this framing challenge was described by the journalist Michael Pollan in an article that asks us to consider the \u201ctrue\u201d cost of a pound of beef.2<\/a><\/sup> We know what we pay for it in the market, but is that a broad enough frame for assessing its cost? What about other costs\u2014what economists call externalities\u2014that are not reflected in the market price?<\/p>\n\n\n\n
Beef is fairly inexpensive because the corn crops that feed the cows are subsidized. So we pay for beef with our taxes. Cows eat corn rather than grass because it\u2019s cheaper to feed them corn. But the cow\u2019s digestive system can\u2019t handle corn, so cows must be dosed with antibiotics to keep them healthy long enough to get them to market. The cost to the farmer of the antibiotics is<\/em> reflected in the market price of beef, but we also \u201cpay\u201d for this antibiotic prophylaxis in drug-resistant strains of bacteria that make human illnesses harder and more costly to treat. This cost is not<\/em> reflected in a pound of beef.<\/p>\n\n\n\n
Corn-fed beef is fattier than grass-fed beef, and the kind<\/em> of fat is worse for human health. And growing the corn that feeds the cows depends on heavy applications of fertilizer, which depend on petrochemicals. Thus, if one framed the price of a pound of beef more broadly, to include all these externalities, the cost would have to include some fraction of the cost of bacterial infection and cardiovascular disease. That, in turn, would have to include the costs of treatment, the costs in mortality and morbidity, workdays lost, and quality of life decreases. And it would have to include some fraction of the cost\u2014in money and in lives\u2014of a <\/a>foreign policy that is partly driven by the need for reliable access to petrochemicals.<\/p>\n\n\n\n
Where does this accounting for the price of a pound of beef stop? The narrow frame of the supermarket price puts all these other considerations outside the frame. The broad frame suggested by Pollan turns our grocery shopping into a matter of geopolitics. Each of these frames is useful, but they induce us to think about very different things. Moreover, even using the word \u201ccosts\u201d narrows the frame by suggesting we think about costs and benefits in terms of dollars. Yes, health problems have financial costs, which can at least be estimated, but they also have quality of life costs, community and family stability costs, and psychological costs that cannot be measured in dollars. You can\u2019t do a spreadsheet analysis unless you can quantify the entries in all of the cells, and that quantification must be in the same units of measure so that different effects of producing meat can be compared and traded off against one another.<\/p>\n\n\n\n
Using narrow frames may miss some important externalities, but using very broad frames may create an accounting problem that is simply unmanageable. Deciding on an appropriate accounting frame requires deliberation, and any frame you choose will provoke disagreement\u2014to this extent, it is subjective. But the process of rational discourse may yield frames that are broad enough to capture some externalities but narrow enough to enable your deliberations to come to a reasonable conclusion. Finding this framing \u201csweet spot\u201d is not the sort of job that RCT is equipped for.<\/p>\n\n\n\n
Many issues similar to those raised by Pollan in his discussion of the \u201ctrue\u201d cost of beef arose in connection with the COVID pandemic. What was the right health policy to follow? There were complex issues to consider even when the problem was narrowly framed. How serious was the disease? How, and how easily, did it spread? What were the chances that treatment and a vaccine would be developed, and when? Could we determine whether certain sectors of the population were especially vulnerable and thus tailor health policy with that in mind? As our understanding of COVID grew, it became possible to develop answers to these narrow questions, though none of them was unequivocal.<\/p>\n\n\n\n
As things turned out, the issues went beyond medical concerns. Should we shut down all public gatherings, and, in consequence, shut down schools and businesses? What would be the health effects of a shutdown? What would be the economic effects? What would be the intellectual and psychological effects on our children? Again, these were all relevant questions, and they were impossible to answer with any certainty. In addition, answering these questions would require trade-offs among very different valuable things. Any relaxation of shutdown rules would surely result in more illness and death. How much more, and was it worth it?<\/p>\n\n\n\n
We rehearse here complexities of which you are surely aware. Policies were enacted, or recommended, and they varied from state to state and country to country. Were they the right policies? More than a million Americans died. Could we have done better? Millions of jobs were lost. Could we have done better? Massive government financial aid was provided to both individuals and businesses to help mitigate the consequences of shutdown. Were they adequate, or should the government have done more? Over time, the combination of extra money in people\u2019s pockets, supply-chain breakdown that limited availability of a wide variety of goods, and massive amounts of extra time for consumption by people who weren\u2019t working led to significant inflation. In addition, schoolkids regressed substantially in their educational attainments and also developed a host of psychological challenges. Were these costs to the well-being of children worth it in lives saved?<\/p>\n\n\n\n
As the epidemic subsided, and as some of the long-term effects of the health policies became apparent, it became clear that policy makers had not accounted accurately and adequately for all of the associated costs, short and long term. Given the complexity, novelty, and urgency of the problems they were facing, it would have been a miracle if the policies adopted were judged in hindsight to be the best that could have been.<\/p>\n\n\n\n
Some costs of the policies were clearly (in retrospect) underestimated. Some benefits of the policies were clearly (in retrospect) overestimated. We say this not to criticize the policies that were adopted but to indicate how incredibly difficult the problem was. Closing the frame so that only health effects were considered might have made the problem easier, but that could have led to economic and social consequences from which it might have taken generations to recover. Fully opening the frame, so that all potential economic and social consequences were given due consideration, might have made formulating a policy unacceptably slow, if not downright impossible.<\/p>\n\n\n\n
RCT would have helped policy makers decide, but perhaps at the cost of ignoring major potential consequences. What was needed, and arguably what we got, was good thinking and good judgment. Perhaps we did reasonably well under the circumstances. And we can hope that should a similar crisis arise, policy makers will have learned how to improve their thinking, their judgment, and their policies.<\/p>\n\n\n\n
In another well-known example, lawyer Kenneth Feinberg was charged with the task of getting the families of victims of the 9\/11 terrorist attacks to agree on compensation for their lost loved ones. What is the value of a life? Some would say lives have infinite value, but that view wouldn\u2019t solve Feinberg\u2019s problem. Alternatively, you could calculate the expected career earnings of each person who died and compensate the victims\u2019 families on that basis. The families of victims who worked, say, in finance, would get much more (to compensate for their loved one\u2019s lost earnings) than the families of janitors, food service workers, or firefighters. On the other hand, you could operate on the view that each human life has equal value and offer compensation on that basis. Or you could consider how old victims were, calculate the expected number of years of life lost, and compensate on that basis. Each of these approaches is reasonable, but they lead to very different judgments about compensation. The task of quantification is essentially impossible. But failing to solve this problem in some way would leave victims\u2019 families with nothing. Feinberg\u2019s efforts to do the impossible while being fully aware that it is<\/em> impossible are vividly depicted in the 2021 movie Worth<\/em>.<\/p>\n\n\n\n
Now consider two very current examples. What will be the costs and benefits of President Trump\u2019s policies regarding tariffs, on the one hand, and the recently passed budget bill, on the other? There is partisan disagreement about each of these policies stemming partly from ideology, partly from massive uncertainty, partly from different predictions about the future of the economy, and partly from matters of framing. This last has mostly been ignored.<\/p>\n\n\n\n
In the case of tariffs, estimating the economic costs and benefits would be extremely complex even if the experts knew for certain what size tariffs would be imposed on which countries and how those countries might retaliate. It has been a century since tariffs like these have been imposed, and the world has changed quite a lot in a century. But beyond the economic costs, what will be the psychological costs? We know that massive uncertainty is bad for our psychological health, leading to anxiety disorders and learned helplessness. And what will be the effects on communities whose main source of economic wherewithal may be substantially or even completely eroded as a result of economic changes brought about by the tariffs? How will the inability to earn a living and keep a family together affect the victims of these changes? And how are the economic costs and benefits to be compared with the psychological, social, and cultural costs and benefits? Most assessments of the tariffs have been within economic frames. The psychological and social costs have not been much in the picture.<\/p>\n\n\n\n
In the case of the new budget, again, aside from the economic costs and benefits, what about lost health insurance and its effects on how hospitals manage, how doctors practice, and how the stress on patients losing their coverage will affect them and the communities they live in? What will increases in wealth inequality produced by the tax bill do to the allegiance of many citizens to the American project? What will be the effects on community well-being if many community institutions are forced to close their doors? Again, the narrow framing encouraged by RCT leaves these kinds of concerns out of the picture. Nor does it tell us how to compare the benefits (if they exist), denominated in dollars, to the costs (if they exist), denominated in individual and community well-being.<\/p>\n\n\n\n
RCT is essentially blind to much that should be considered when policies like tariffs and tax laws are being evaluated. It yields a very incomplete picture. This does not stop us, as individuals, from thinking about a broader set of possible effects in order to enrich the picture. But if we become adherents of RCT, will we still know how to do that, or still know that we should<\/em> do that?<\/p>\n\n\n\n
This brings us to perhaps the worst aspect of RCT. It is not only incomplete and inaccurate, but it is dangerous. Why dangerous? If RCT is anointed as the epitome of rationality, then people may stop thinking in broad frames and accept that the output of RCT is all we need. The alternative vision of rationality that we offer in our book Choose Wisely<\/em> is intended to resist the use of narrow frames that leave out much of what is most important.<\/p>\n\n\n\n
If people stop thinking in broad frames, they will lose the ability to consider thoughtfully. Moreover, the broader effects of policies, or decisions, cannot be quantified in the way that strictly economic effects can. And we know from recent research by <\/em>Linda W. Chang, Erika L. Kirgios, Sendhil Mullainathan, and Katherine L. Milkman that dimensions of decisions that can be quantified overshadow dimensions of decisions that can\u2019t, even if these latter dimensions are more important.3<\/a><\/sup> One of the things RCT teaches us is that if you can\u2019t attach a number to an aspect of a decision, that aspect should diminish in importance.<\/p>\n\n\n\n
RCT has its place, but that place is not every place. It is only after a decision or a policy is well framed that RCT can play its role. And RCT tells us nothing about how policies and decisions should be framed.<\/p>\n\n\n\n
\n\n\n- Duke, A. (2018). \u00a0Thinking in Bets. <\/em>New York: Random House.
![\"\u21a9\"](\"https:\/\/s.w.org\/images\/core\/emoji\/17.0.2\/72x72\/21a9.png\")
\ufe0e<\/a><\/li>- Pollan, M. (2002). Power steer. New York Times Magazine, <\/em>March 31, 44-51, 68,
71-72, 76-77.\ufe0e<\/a><\/li>- Chang, L.W., Kirgios, E.L., Mullainathan, S., & Milkman, K.L. (2024). Does counting change what counts? Quantification fixation biases decision-making,\u00a0Proc. Natl. Acad. Sci. <\/em>121 e2400215121,\u00a0https:\/\/doi.org\/10.1073\/pnas.2400215121<\/a>\u00a0(2024).
\ufe0e<\/a><\/li><\/ol>\n\n\n
\n\n\n\nBarry Schwartz<\/strong> is professor emeritus of psychology at Swarthmore College and visiting professor at Haas School of Business, University of California, Berkeley. His books include The Paradox of Choice<\/em>, Why We Work<\/em>, and (as coauthor) Practical Wisdom<\/em>. He lives in Oakland, CA. Richard Schuldenfrei<\/strong> is professor emeritus of philosophy at Swarthmore College. He lives in West Townshend, VT.<\/a><\/p>\n
The post Choosing the Right Frame and Its Effects on Public Policy<\/a> appeared first on Yale University Press<\/a>.<\/p>\r\n
American politics,donald trump,economics,politics,public policy,rational choice theory\r\n
Source link yalebooks.yale.edu<\/a>","protected":false},"excerpt":{"rendered":"Barry Schwartz and Richard Schuldenfrei\u2014 The COVID epidemic knocked the whole world for a loop. Different nations responded to it differently, as did different regions within the United States. Did we handle it well here in America? Did we handle all of it well? How should we go about trying to answer these questions? President Trump has imposed and threatened…","protected":false},"author":603,"featured_media":69,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_analytify_skip_tracking":false,"footnotes":""},"categories":[1],"tags":[682,218,576,200,1214,1215],"class_list":["post-2837","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-general","tag-american-politics","tag-donald-trump","tag-economics","tag-politics","tag-public-policy","tag-rational-choice-theory"],"_links":{"self":[{"href":"https:\/\/wokeantifa.org\/topics\/wp-json\/wp\/v2\/posts\/2837","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/wokeantifa.org\/topics\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/wokeantifa.org\/topics\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/wokeantifa.org\/topics\/wp-json\/wp\/v2\/users\/603"}],"replies":[{"embeddable":true,"href":"https:\/\/wokeantifa.org\/topics\/wp-json\/wp\/v2\/comments?post=2837"}],"version-history":[{"count":0,"href":"https:\/\/wokeantifa.org\/topics\/wp-json\/wp\/v2\/posts\/2837\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/wokeantifa.org\/topics\/wp-json\/wp\/v2\/media\/69"}],"wp:attachment":[{"href":"https:\/\/wokeantifa.org\/topics\/wp-json\/wp\/v2\/media?parent=2837"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/wokeantifa.org\/topics\/wp-json\/wp\/v2\/categories?post=2837"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/wokeantifa.org\/topics\/wp-json\/wp\/v2\/tags?post=2837"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}} - Pollan, M. (2002). Power steer. New York Times Magazine, <\/em>March 31, 44-51, 68,