What Intelligence Tests Miss

by Keith E Stanovich

  Status Quo Bias: The Default Heuristic

  Another tendency of the cognitive miser that robs us of personal autonomy is the overuse of the so-called default heuristic.11 This heuristic operates via a simple rule: If you have been given a default choice, stick with it. That humans have such a heuristic is suggested by two decades of work on status quo biases in decision making. That humans overuse the default heuristic to the point of failing to maximally achieve their goals is also demonstrated in these same two decades of research. People who overuse the default heuristic give up their autonomy by ceding control of their lives to those with the power to set the defaults.

  The default heuristic operates in many real-life contexts of economic and public policy choice. One group of investigators described a survey conducted by Pacific Gas and Electric in the 1980s. Because of various geographic factors (urban-rural, etc.), service reliability varied in the company’s service area. Some of their customers suffered more outages than others. Customers with unreliable service were asked whether they would be willing to pay for more reliable service and, if so, whether they would accept increases of various percentages. Customers with reliable service were asked if they would be willing to accept somewhat less reliable service and receive a discount on their bills of a certain percentage (in fact, the same percentages as the other group, only a decrease instead of an increase). Although there were not income differences between these groups of customers, neither group wanted to change. People overwhelmingly wanted to stay with whatever their status quo was. The service difference between the two groups was large. The unreliable service group suffered 15 outages per year of 4 hours’ average duration and the reliable service group suffered 3 outages per year of 2 hours’ average duration, yet very few customers wanted to switch!

  Hostile and Benign Environments for Heuristics

  Of course, I do not mean to imply that the use of heuristics always leads us astray. As I argued above, they often give us a useful first approximation to the optimal response in a particular situation, and they do so without stressing cognitive capacity. In fact, they are so useful that one group of influential psychologists has been led to extol their advantages even to the extent of minimizing the usefulness of the formal rules of rationality.12 Most psychologists, though, while still acknowledging the usefulness of heuristics, think that this view carries things too far. Here is why.

  The usefulness of the heuristics that the cognitive miser relies upon to lighten the cognitive load is dependent on a benign environment. By a benign environment, I mean an environment that contains useful cues that can be exploited by various heuristics (for example, affect-triggering cues, vivid and salient stimulus components, convenient anchors). Additionally, for an environment to be classified as benign, it also must contain no other individuals who will adjust their behavior to exploit those relying only on heuristics. In contrast, a hostile environment for heuristics is one in which there are no cues that are usable by heuristic processes. Another way that an environment can turn hostile for the cognitive miser is if other agents discern the simple cues that are triggering the miser’s heuristics and the other agents start to arrange the cues for their own advantage (for example, advertisements, or the deliberate design of supermarket floor space to maximize revenue).

  Take as an example one chapter in an edited book extolling the usefulness of the so-called recognition heuristic.13 The chapter is subtitled “How Ignorance Makes Us Smart.” The idea behind such “ignorance-based decision making,” as it is called, is the fact that some items of a subset are unknown can be exploited to aid decisions. In short, the yes/no recognition response can be used as an estimation cue. For example, novice tennis fans correctly predicted the winners of 72 percent of all the men’s matches at the 2003 Wimbledon by using the simple recognition heuristic of: If you recognize one player’s name and not the other’s, predict that the one you recognize will win. This heuristic does just as well as Wimbledon experts’ rankings.

  With ingenious simulations, Gerd Gigerenzer and colleagues have demonstrated how certain information environments can lead to such things as less-is-more effects: where those who know less about an environment can display more inferential accuracy in it. One is certainly convinced after reading material like this that the recognition heuristic is efficacious in some situations. But one immediately begins to worry when one ponders how it relates to a market environment specifically designed to exploit it. If I were to rely solely on the recognition heuristic as I went about my day tomorrow, I could easily be led to:

  1. buy a $3 coffee when in fact a $1.25 one would satisfy me perfectly well

  2. eat in a single snack the number of fat grams I should have in an entire day

  3. pay the highest bank fees

  4. incur credit card debt rather than pay cash

  5. buy a mutual fund with a 6 percent sales charge rather than a no-load fund

  None of these behaviors serves my long-term goals at all. Yet the recognition heuristic triggers these and dozens more that will trip me up as I try to make my way through the maze of modern society. The commercial environment of my city is not a benign environment for a cognitive miser.

  The danger of such miserly tendencies and the necessity of relying on Type 2 processing in the domain of personal finance is suggested by the well-known finding that consumers of financial services overwhelmingly purchase high-cost products that underperform in terms of investment return when compared to the low-cost strategies recommended by true experts (for example, dollar-cost averaging into no-load index mutual funds). The reason is, of course, that the high-cost fee-based products and services are the ones with high immediate recognizability in the marketplace, whereas the low-cost strategies must be sought out in financial and consumer publications. An article in a British publication illustrates the situation by asking “Can 70 per cent of people be wrong?” and answers “yes, it seems.” In the article we learn that, at that time, seven out of ten people in Britain had money in checking accounts earning 0.10 percent with one of the big four banks (Barclays, HSBC, Lloyds TSB, and Royal Bank of Scotland) when interest rates more than 30 times that amount were available from checking accounts recommended in the Best Buy columns of leading consumer publications. The reason millions of people were losing billions of dollars in interest is clear—the “big four” were the most recognizable banks and the cognitive miser defaulted to them. The marketplace of personal finance is not benign. It requires that the investor avoid behaving like a cognitive miser and instead consciously—sometimes disjunctively—think through the alternatives.14

  Just how easy it is to exploit the miser tendency to rely on easily processed stimuli is illustrated in a study by Marwan Sinaceur and colleagues.15 They presented subjects with the following hypothetical situation: “Imagine that you have just finished eating your dinner. You have eaten a packaged food product made with beef that was bought at the supermarket. While listening to the evening news on the television, you find out that eating this packaged food may have exposed you to the human variant of bovine spongiform encephalopathy (BSE).” After reading this, the subjects were asked to respond on a seven-point scale to the following questions: “After hearing this, to what extent would you decrease your consumption of this type of packaged beef?” and “To what extent would you alter your dietary habits to de-emphasize red meats and increase the consumption of other foods?” Not surprisingly, after hearing this hypothetical situation, subjects felt that they would decrease their consumption of beef. However, another group of subjects was even more likely to say they would decrease their consumption of beef when they heard the same story identically except for the very last words. Instead of “human variant of bovine spongiform encephalopathy (BSE)” the second group read “human variant of Mad Cow Disease.” It is clear what is going on here. Our old friend vividness is rearing its head again. Mad Cow Disease conjures creepy imagines of an animal-borne disease in a way that bovine spongiform encephalopathy doe
s not. In short, when we are being cognitive misers, our actions and thoughts are readily influenced by small changes in wording that alter the vividness and affective valence of our reactions. It is a pretty sure bet that Social Security taxes would be less if Social Security was called instead Welfare for the Elderly.

  In short, extreme cognitive misers literally do not have “a mind of their own.” What their mind will process is determined by the most vivid stimulus at hand, the most readily assimilated fact, or the most salient cue available. The cognitive miser is easily exploited by those who control the labeling, who control what is vivid, who control the anchor. We shall see even more dramatic examples of how over-reliance on shallow Type 1 processing threatens our autonomy as independent thinkers in the next chapter when we consider framing effects.

  SEVEN

  Framing and the Cognitive Miser

  Decision makers are generally quite passive and therefore inclined to accept any frame to which they are exposed.

  —Daniel Kahneman, Choices, Values, and Frames, 2000

  Edward McCaffery, a professor of law and economics, and Jonathan Baron, a cognitive psychologist, have collaborated on extensive studies of people’s attitudes toward aspects of the tax system.1 They have found, to put it bluntly, that people’s thinking about taxes is incoherent. I am going to focus on one particular type of incoherence that they have studied, because it illustrates a critical pitfall of the cognitive miser.

  Focus for a moment on how you would set up an idealized tax system in a hypothetical country. Imagine that in this country a family with no children and an income of $35,000 pays $4,000 in tax and that a family with no children and an income of $100,000 pays $26,000 in tax. Imagine that it is proposed in this hypothetical country that there be a $500 tax reduction for having a child for a family with an income of $35,000. Thus, that family’s tax would go from $4,000 to $3,500 when they had one child. The question is, should the reduction for the family with an income of $100,000 be the same? Should their tax go from $26,000 to $25,500 or should they be given more of a reduction because of their higher income?

  Nobel Prize–winning economist Thomas Schelling notes that there are some arguments for the latter (giving the higher-income family a larger tax reduction): “One way to make the case is that the high-income families spend much more on children and the ‘cost’ of raising their children is much more” (1984, p. 19). In short, the high-income family is going to output more money in raising their children, so they deserve more of a reduction. Perhaps you do not find this argument to be convincing. Most people don’t. Most people reject this argument outright and respond instead that the reduction for having a child should be at least the same for low-income households as for those with high income and, if anything, it should probably be higher for the lower-income households.

  This is where economist Schelling steps in and teaches us that we have not thought hard enough about the logic of this situation—that we have not, in particular, thought about alternative ways that it might be framed. He points out that it is arbitrary that we originally framed the issue starting at the rate for a childless household. In thinking about setting up this hypothetical system, we could just as well have started from a different baseline—for example, the baseline of a “typical” family of four (two adults and two children). Of course, as before, children affect the rates, so we would have to figure out the fair tax for families with one child or no children (and for 3, 4, etc.).

  Imagine that in this hypothetical country, a family with two children and an income of $35,000 pays $3,000 in tax and that a family with two children and an income of $100,000 pays $25,000 in tax. What would the rates be here for one child and zero children? We would adjust the rates upward because families without children can afford to pay more. Instead of speaking of a reduction for children, we might call this type of adjustment in the tax schedule the “penalty for the childless.” And here I am giving away a hint at what Schelling is teaching us about framing and tax policy (and what McCaffery and Baron have studied empirically)—every “reduction” (tax credit, or deduction) for a family with a certain characteristic (children, home ownership, farm status, self-employment status, and all the many other characteristics in the tax code) is in effect a penalty for those who do not share the characteristic (because there is a fixed number representing what the sum total of government services has cost, and this must be paid even if the government must borrow).

  So let us imagine in this case that the family with an income of $100,000 and one child has their taxes set at $26,000 and the same family with no children has their taxes set at $27,000. That is, there is a childless penalty of $1000 per child who does not reside in the family. The question is, should the poorer family which makes $35,000 and has no children also pay the same $2000 childless penalty as the richer family—should the poorer family’s taxes go from $3000 to $5000 as the richer family’s taxes went from $25,000 to $27,000?

  Most people have the instinctive feeling that this is not right. Most people feel that the $2000 penalty represents a much more severe hardship for the poorer family and that it should be less than the penalty paid by the richer family with no children. But this judgment does not square with people’s feelings about whether the reduction for children should be the same for rich and poor families. People want the “bonus” for children to be equal for low- and high-income families, but they do not want the “penalty” for lacking children to be the same for high and low income. This is incoherent thinking, because the bonus and the penalty are exactly the same thing—just with different names that direct the focus of attention in different directions. And this is the point of the example, and of this chapter—that cognitive misers allow their attention to be focused by others. Cognitive misers let the structure of the environment determine how they think. The miser accepts whichever way the problem is presented and thinks from there, often never realizing that a different presentation format would have led to a different conclusion.

  In cognitive science, the tendency to give different responses to problems that have surface dissimilarities but that are really formally identical is termed a framing effect. Framing effects are very basic violations of the strictures of rational choice. In the technical literature of decision theory, the stricture that is being violated is called descriptive invariance—the stricture that choices should not change as the result of trivial rewordings of a problem.2 Subjects in framing experiments, when shown differing versions of the same choice situation, overwhelmingly agree that the differences in the problem representations should not affect their choice. If choices flip-flop based on problem characteristics that the subjects themselves view as irrelevant—then the subjects can be said to have no stable, well-ordered preferences at all. If a person’s preference reverses based on inconsequential aspects of how the problem is phrased, the person cannot be described as maximizing expected utility. Thus, such failures of descriptive invariance have quite serious implications for our view of whether or not people are rational.

  Tax policy is a good domain in which to see framing effects operating because reframing can be done so easily, yet the possibility of reframing often goes completely unnoticed. The idea of a “tax deduction” seems to most people such an unequivocally good thing that any policy attached to the term is usually endorsed. It is rarely noticed by anyone other than economists that a tax deduction for citizens having a certain characteristic is the equivalent of a penalty for those not sharing the characteristic. As two economists describe the situation, “by requiring higher tax rates, subsidies cause everything else to be penalized. . . . The point is that these features are enormously popular because they have been enshrined as ‘tax reductions,’ but these exact same features probably wouldn’t stand a chance as stand-alone policies” (Slemrod and Bakija, 1996, pp. 113, 141). This quote draws attention to the fact that no matter what amount of government services (be it national defense, health care, roads, or payments to the elderly) we deem appr
opriate, there is some fixed sum of money that must be raised to pay for it—either now or in the future (the latter if the government takes on debt to pay for the services). Thus, deductions for certain classes of taxpayer necessarily mean that those not qualifying for the deduction will pay more.

  Consider the tax deduction for home mortgage interest. On its surface it seems like a good thing, but it would seem less benign if we were to describe it as “the rent payer’s penalty.” When we recognize that this is an equivalent reframing, we realize the sense in which phrasing the issue as “should there be a deduction allowed for home mortgage interest paid?” biases the question. Rephrasing the question as “Should renters pay more tax in order that home owners can pay less?” is an equivalent framing biased in the other direction. Likewise, a lower tax rate for capital gains sounds less benign when counter-phrased as “the wage earner’s penalty.”

  Framing and Personal Autonomy

  The fact that our opinion of a tax policy can be changed by a mere reframing of the policy clearly illustrates that we lose personal autonomy when we act as cognitive misers. We literally allow whoever chose the framing to “make up our minds” for us.