Explaining Behavior 10: Rational Choice Theory
Incentives Matter
This is part of a series on how to explain human behavior. It is based on a course developed by the late Donald Black at the University of Virginia. Parts 1 through 6 discussed the general logic of scientific theory. From Part 7 onward we cover different strategies or paradigms for developing explanations. The last installment was Part 9: Opportunity Theory.
One way to think about different sociological paradigms is that they tell you what questions you ought to ask.
Someone does something strange. First you might ask: “What on earth was he thinking? What was his intent?” This is phenomenology.
The motivational paradigm would lead you to ask: “What made him want to do that? Did he lose his job? Was it peer pressure? Was it how he was raised?”
The opportunity paradigm would lead you to ask: “How was he able to do that?”
Now let’s consider that someone has multiple opportunities to choose from. Which will he choose? One strategy for answering that question is to look at the costs and benefits of each course of action.
The Logic of Suicide Terrorism
In the wake of the September 11 attacks, political scientist Robert Pape made an argument that was calculated to be interesting. While many people attributed suicide terrorism to irrational fanaticism, it was actually a rationally calculated political strategy. He examined this strategy in an article “The Strategic Logic of Suicide Terrorism,” later expanded into the book Dying to Win.
In these, he points that that most acts of suicide terrorism aren’t isolated attacks by lone crazies — rather, most attackers are dispatched by organizations like Hamas, Hezbollah, the Tamil Tigers, the Kurdistan Workers’ Party, and Al-Qaeda. And most of these attacks occur as part of a larger campaign of terrorism, itself part of a larger political struggle carried out with a variety of violent and nonviolent means.
Militarily, terrorism isn’t something that can conquer and hold territory, nor can it expel an invading force. What it can do is punish an enemy — raising the costs they bear for not giving into your demands. Punishment can be enough to get an enemy to change course, or at least grant concessions.
As a form of punishment, suicide terrorism has advantages. The suicide bomber can sneak past security, mix with civilians, and find the most advantageous time to detonate the bomb so as to maximize carnage. Pape reports that suicide attacks tend to have higher casualties than non-suicide attacks. As many have put it, the suicide bomber is the poor man’s guided missile.
Suicide attacks also have the advantage of being a credible signal of more punishment to come. Attackers who are willing to die for their cause are not easy to deter. People rightly fear them more.
When will a militant group turn to suicide attacks? Pape’s explanation hinges on identifying factors that make it more cost-effective than alternatives.
One factor is the relative strength of the groups. Militants turn to terrorism to the extent they can’t defeat the enemy army in the field, and find it too difficult or costly to punish the enemy through more conventional means. Suicide terrorism is more attractive to the extent your enemy is more powerful.
Another factor, Pape argues, is the nature of the political goal. He claims suicide terrorism tends to arise when groups have nationalist goals, trying to expel an occupier from their perceived homeland. This kind of conflict helps militants rally the surrounding civilian populations to support such extreme actions and generates recruits willing to sacrifice themselves.
Religious differences between occupier and occupied have the same effect. The strategy of suicide terrorism just works much better when cultural factors facilitate a clear “us versus them” distinction and zealous popular support.
The political organization of the enemy also matters. Pape argues that suicide terrorism tends to target democratic countries. This is because democracies are more vulnerable to punishment. There are always people who will vote for peace and safety, and grieving parents who will ask “why are our soldiers over there, anyway?” An authoritarian regime can ride roughshod over civilian objections, but a democratic one must placate its terrorized populace.
A democratic opponent makes suicide terrorism more effective, and also less costly. For an authoritarian regime is freer to bring extrem retaliation against militants and their broader community. Democratic citizenry have a limited appetite for the suffering of their enemies, and every war brings its humanitarian peace movement. Outright massacres are bad optics. Dictators face no such limitations. According to Pape’s theory, this is why one sees Kurdish separatists using suicide bombing in Turkey but not in Saddam Hussein’s Iraq.
Combining these ideas: Suicide terrorism is most likely when a country that is more powerful, has a democratic government, and a different majority religion is perceived as occupying a militant group’s homeland. And we arrived at these propositions based on an analysis of what would make suicide terrorism a cost-effective means of achieving political goals.
The Rational Choice Paradigm
Pape’s theory of suicide terrorism is an example of a rational choice theory. Rational choice theories, as Donald Black defines them, explain behavior as the least costly means to a goal.
Black considers rational choice explanation in social science a subset of the utilitarian style of thought, which is concerned with weighing costs and benefits. Thus we also have utilitarian ethics (associated with philosophers like Jeremy Bentham and John Stuart Mill) in which one judges the rightness or wrongness of behavior with how well it meets certain goals (such as providing the most happiness for the most people).
Black also considers Marxism to be a breed of utilitarian thought, in that it analyzes behavior in terms of the costs and benefits it brings to a social class. Ironically, then, both the pro-capitalist discourse of economist Adam Smith and the anti-capitalist discourse of Karl Marx share an underlying similarity.
According to Black, all rational choice theories make four main assumptions:
They assume actors are rational — that is, that they can and will weigh costs and benefits.
They assume that actors are self-interested — though of course this can extend to an interest in loved-ones, allies, and so forth.
They assume the actor’s goals
They assume the least costly means of meeting those goals.
The first two assumptions are widely recognized, and sometimes jointly described as the assumption of rational egoism. This might also be referred to as the rational actor assumption or rational actor axiom.
Rational choice theorists are usually quite clear about this being an assumption — not ideas they seek to deductively justify or directly test, but a fundamental assumption (or axiom) that is part of a larger theoretical system. According to sociologist Rodney Stark, critics who argue with this assumption are fundamentally misguided: They should be more concerned with whether any particular rational choice theory is making correct predictions.
Black’s notion that rational choice theories assume goals isn’t something I’ve seen discussed elsewhere, but it would seem to be true by definition, in two senses.
Firstly, in Black’s classification of paradigms, if we were explaining goals as such we’d be constructing a motivational theory (or, if the goals are instinctual, a neo-Darwinian theory).
Secondly, costs and benefits can only be assessed relative to a given goal. Drinking hemlock is rational if your goal is to die, irrational if your goal is to live. Rationality is thus a property of means, not of ends. Means can be rational or irrational, but ends are non-rational.
Thus rational choice theories assume goals in the sense that they don’t make them an object for explanation. This does not mean they’re simply made-up: When Pape says terrorist groups have nationalist goals, for instance, he’s reporting what the groups themselves demand.
But it’s worth noting that sometimes rational choice thinkers impute goals to make behavior consistent with the rational actor assumption. For instance, if it appears that most people’s way of arguing is an irrational means of either uncovering the truth or persuading one another, their real goal must be something else: Perhaps status seeking or tribal affiliation.
The final assumption Black lists is is the assumption of what is least costly. This appears to be true in the same sense as the assumption of goals: The relative cost of different means are not themselves explained, but notions about them are used to predict behavior. These notions might be based on prior empirical observations, or they might be imputed so as to make behavior consistent with the rational-actor assumption.
Though less prominent than other approaches, rational choice theory is used in sociology: Major advocates include George Homans, James Coleman, and Rodney Stark.
The paradigm also has a prominent place in political science — in addition to Robert Pape’s work, there’s a cottage industry in rational choice accounts of terrorism.
But rational choice is unusual among social science paradigms in that it almost entirely dominates one field: Economics. Indeed, economists venturing outside of their traditional specialties to explain religion, family, or other aspects of the social world will speak of applying economics or “the economic perspective” to these topics.
Conceptual Theory
Important concepts in the rational choice paradigm include ways of discussing and classifying costs and benefits.
These only exist relative to an actor’s goal, which might be referred to as their interest or preference. These might be material or immaterial — people pursue prestige, positive self-image, emotional attachment, and other things besides material wealth and comfort.
Economists commonly use the term utility to refer the benefit provided by something. The rational actor assumption is sometimes expressed as the idea that people maximize utility.
The benefit of acquiring something will vary depending on how much of that something one already has. For instance, a glass of water provides a much bigger benefit if I’m dying of thirst than I’m already hydrated. The term marginal utility refers to the benefit conferred by some additional increment of a good or behavior.
Costs can vary in the same way. Once you’ve set up your widget factory and produced your first 1,000 widgets, making one more widget add very little to the expense. Thus just as there is marginal utility, there is also marginal cost.
Costs come in many forms, some less obvious than others. For instance, the opportunity cost of a behavior is the cost incurred by not doing something else with your time and resources. All actions have this cost, and so does inaction. Just sitting around doing nothing comes with the cost of what you could have gained if you had done something.
An incentive is the cost or benefit of a given behavior. To make a behavior more beneficial is to incentivize it — assuming the actor is rational, this makes the behavior more likely.
Some incentives are more obvious than others. One of the great stumbling blocks of social policy is that quite often we can be mistaken about what sort of incentives we’re creating. A perverse incentive is one that has unintended consequences — policymakers intended to incentivize one behavior, but accidentally incentivized another. A common example is using bounties for pest control. If you put a bounty on rat tails to incentivize the killing of rats, you might find yourself paying people to breed rats and harvest their tails.
Economics and Politics
As noted, the rational choice paradigm dominates modern economics. The assumptions aren’t always made explicit, but one can see them at work in propositions about supply and demand curves.
When the price of a good rises, the aggregate quantity demanded falls — because potential consumers are incentivized to find a cheaper alternative. But as the price of a good rises, the aggregate supply increases — because potential producers are incentivized to go to greater lengths to produce it. The interaction of these two relationships produces the actual quantities and prices we see in the marketplace.
For anyone unfamiliar with these theories, economists Tyler Cowen and Alex Tabarrok have some good videos: Supply curve, demand curve, equilibrium price.
See also this video on comparative advantage and gains from trade.
The rational choice paradigm dominates the study of buying, selling, and marketplaces for goods. And indeed shopping is an area of life that most of us experience as a weighing of costs. But economists and others often apply similar ideas to other realms of the social world.
Economist Gary Becker, for instance, wrote A Treatise on the Family, which uses an economic framework to explain aspects of marriage, divorce, and childbearing. Why do people have fewer children these days? Absent a farm for them to work on, they’re less of an economic benefit and more of a cost.
One of his more interesting ideas about divorce was that it shouldn’t happen if the sum of both spouse’s utility from staying married is postive: As long as the benefit one spouse gets from leaving is smaller than the benefit the other gets from staying married, there ought to be bargaining — being nicer, doing more housework, or whatever — sufficient to make the less satsified spouse stay around.
Economist Anthony Downs examined the rationality of voters in his Economic Theory of Democracy. For instance, on the fact that voters tend to be extremely ignorant of the issues of the day, he observed that their ignorance was rational: It’s not worth the cost of informing yourself on the issues when your individual vote will have near-zero effect on the election.
In his Myth of the Rational Voter, economist Bryan Caplan goes beyond this notion of rational ignorance and argues that voters have rational irrationality. Absent some cost for holding erroneous beliefs or supporting bad policies, people are free to gravitate toward whatever gives them psychological benefits. So not only is it rational for voters to be ignorant, it is rational for them to have flatly mistaken beliefs and vote for policies that would be detrimental to themselves — after all, their individual vote doesn’t matter much anyway.
The most interesting thing about this argument is that it purports to explain why the same person who carefully compares the price at the grocery store, or displays a shrewd skepticism at a sale pitch, will support policies and candidates based on feelz. And it might by extension say something about why the rational choice paradigm has made more headway studying the economy than in other realms of social life.
But for all that it seems like those who work in this paradigm delight in applying it to behaviors that are widely thought of as governed by irrational forces. Love, marriage, martyrdom — even irrationality is rational! Perhaps, then, it is not surprising that the rational choice approach within sociology has been especially prominent in the study of religion.
The Religious Marketplace
Rational choice theorists conceptualize various parts of social reality as markets, even if they’re not realms where we usually buy things with cash. In his article “Rational Choice Theories of Religion,” sociologist Rodney Stark defines the religious economy as “all the religious activity going on in any society” and notes that “they consist of a market of current and potential customers, a set of firms seeking to serve the market, and the religious ‘product lines’ offered by the various firms.”
These religious firms are offering goods to people — not just the content of the faith, but things like community, charity, entertainment, guidance. Firms vary in what mix of products and services they offer, including in things like their style of worship — it could be a jump-and-shout church or somber hymn church or a quiet meditation temple or whatever.
Stark proposes that the religious economy works much like the regular economy, and this can help us explain why some countries have much higher levels of religious participation.
His propositions are:
“The capacity of a single religious firm to monopolizes a religious economy depends upon the degree to which the state uses coercive force to regulate the religious economy.”
“To the degree that a religious economy is unregulated, it will tend to be very pluralistic.”
“To the degree that a religious economy is pluralistic, firms will specialize.”
“To the degree that a religious economy is competitive and pluralistic, overall levels of religious participation will tend to be high.”
The logic here is that religious firms want worshippers — they are rational actors with the goal of gaining “customers.” In a monopoly, where there’s no competition, gaining customers is easy — there’s little incentive to bust your hump attracting them. Monopolies breed lazy firms.
Potenial customers, for their part, vary in what religious products they prefer — some like mild and laid back, some like fire and brimstone, etc. If they find a product that suits them, they’re more likely to be a customer.
When firms have a lot of competition, they’ll work harder to get customers, and are incentivized to specialize in different customer preferences. When lots of firms are specializing, potential customers are more likely to find a firm that suits them. Therefore, the overall level of religious participation increases.
Meanwhile, in a society with a state-backed church that has something approaching a monopoly, people have a more restricted choice: It is often go to the church or stay home. Many stay home. Thus overall participation is lower.
According to this theory, America is more religious than many European countries because it has a free market in religion, leading to a greater and more varied supply.
As I’ll discuss in a later post, the validity of this theory is up for debate. But it shows the sort of predictions one might make when one starts seeing markets in everything.
Cooperation and Coalitions
Game theory is a branch of rational choice theories dealing with contingent rationality: What is the rational move in a competitive situation where the other player is also going to be making a rational move?
One of its classic thought experiments is the Prisoner’s Dilemma. If two criminals are each offered a lighter sentence for being the first one to rat on the other, is it rational to rat? If you’re one of them, you have to consider that your accomplice is getting the same offer that you are. So even if you’d rather be loyal, you have to worry if you can trust him not to rat. And if he is rational, he’s thinking the same thing about you, which means the safe bet for him is to rat, which means the safe bet for you is to rat even faster….
This sort of thinking underlies sociologist Theodore Caplow’s theory of coalitions in the triad.
“A triad is a social system containing three related members in a persistent situation” writes Caplow. “The most significant property of the triad is its tendency to divide into a coalition of two members against the third”
In his book Two Against One, Caplow presents a general theory that aims to explain, for any given triad, which if any members will form a coalition against the third. The main variable that explains coalitions is the distribution of power: “The appearance of particular coalitions can be predicted with considerable accuracy if the relative power of the three members is known.”
For instance, imagine three smugglers alone on an island, deciding how to divvy up their loot. Two of the smugglers are equal in strength. The third is stronger than each of the others, but weaker than the both of them joined together.
The most likely coalition would be between the two weaker smugglers. Neither of the weaker parties would have a reason to form a coalition with the stronger one, because in such a coalition they’d still wind up being dominated by the strongest. They’d still be less likely to receive an equal share of whatever goods were being divided.
But in a coalition between the two weaker equals, neither has to worry about being dominated by the other — they’re evenly matched. And by joining together, both can avoid being dominated by the strongest. In this way they maximize their share of the loot. The rational and therefore most likely strategy is for the two equally weak parties to club together against the stronger third.
This example shows the basic logic with which Caplow predicts coalitions. It also points to an interesting property of coalitions: Sometimes strength can be a weakness, in that it keeps one from being a preferred coalition partner. The strongest party in the triad might well be the one who winds up getting dominated.
Collective Action
In his book The Logic of Collective Action, economist/political scientist Mancur Olson asks when social groups will pursue some common interest — for example, when workers will form a union to seek higher wages.
This is worth explaining because people don’t just automatically band together to pursue common interests. One reason they don’t is because in addition to the common interest, all members of a group have their own individual interests. As a worker, it is good for me if all the workers get higher wages — but I’ve got an interest in not personally sacrificing to make it so. If I can get higher wages without paying union dues and participating in costly strikes, I will.
This presents a problem for groups pursuing any kind of public good — defined as a good that goes to all members of the group. By its nature, this good isn’t limited to people who worked to achieve it — if the union gets a new minimum wage in the stockyards, the guys who never paid union dues get it just as much as the guys who had their skulls cracked by company police. Likewise, people who don’t pay taxes drive on the same interstate highways as people who lose a massive chunk of their income to the government. And companies can get an industry-wide subsidy because other companies in their industry were funding lobbyists to obtain them.
This is the free-rider problem, and it renders all collective action problematic. In the case of a public good, the rational thing for all actors to sit around and wait for some other actor to provide it. But if they’re all following this logic, none of them will act, and the public good will never get provided.
Thus, Olson said, groups that successfully pursue public goods have certain properties. One is that they also provide additional, non-public goods that can be used as selective incentives. All workers might benefit from the raise, but only those who pay the union dues can get help with workplace grievances or drop off their kids at the union-funded daycare.
Sometimes these selective incentives are immaterial social incentives — the group accords respect to those who help the collective cause, but looks down on those who free-ride. These social incentives surely play an enormous role in group action across human history.
Negative incentives like social disapproval shade into outright coercion. And Olson notes that groups pursuing collective action often employ coercion. Unions have a history using violence against non-union workers, attacking, beating, and even killing “scabs.” And for all the patriotic fervor that people sometimes display, no state has ever funded itself with voluntary donations — they take taxes under threat of severe punishment. You pay for the interstate even if you don’t want to have it.
Olson also proposes that group size has an impact on collective action: Smaller groups are more likely than larger groups to take collective action. This is because as group size increases, the cost of action increases. Not only are basic coordination problems greater, but it becomes costlier to monitor contributions and police free riders. Small groups can do this quite easily, and will find even informal social sanctions more effective.
Also, at least in some circumstances, increasing group size decreases the share of the public good doing to each individual. We can see this most clearly in the power of special interest groups versus the average taxpayer. A consortium of cotton growers lobbies Congress for a few billion in subsidies — on average, each of these companies is getting a few hundred thousand dollars. Meanwhile, the rest of us taxpayers are on average losing a few cents each. Any particular cotton grower has a strong incentive to chip in for the lobbyists and campaign donations that keep the subsidies flowing. Any particular taxpayer has next to no incentive to campaign against their subsidies. This is so even if the cumulative effect to of many such special interests is quite burdensome. Thus concentrated interests have power over diffuse ones.
If we follow Black in calling Marx a utilitarian, he was one who was blind to individual interests. For even if its true that workers have a class interest in fighting a revolution, who wants to die on the front lines? Workers of the world, shuffle awkwardly and look at your feet and wait for someone else to go first!
This is one reason that real revolutions are a nightmare. For the revolutionaries, like the established powers, rely on coercion to compell collective action from a populace who’d rather stay out of the fight. Pity the poor man who just wants to quietly provide for his family, but will be shot by one side for a flag he flew to avoid being shot by the other.
Thanks for reading! If you appreciate Bullfish Hole, please become a subscriber at the link below. You can also leave a tip of any size in the Tip Jar.