The behavioural economist's dilemma: induction versus deduction

This summer I went to a conference for behavioural economists and economic psychologists. The presentations were entertaining.

Did you know that when coffee is given a fair trade label, people say it tastes better, even if it's just regular coffee? And fair trade coffee, without the fair trade label, doesn't taste as good?*

Did you know that an increase in a co-worker's pay will cause a person to reduce her own work effort?**

But somehow the intellectual excitement wasn't there.

Then I met Pete Lunn, an economic psychologist, who articulated the cause for my disquiet.

Economics, he said, is basically a deductive discipline. Psychology, on the other hand, is inductive. 

It's a caricature, a gross over-simplification. But it has an element of truth.

Economists like to begin with a general model, work out its implications, and then go to the data to test the theory. Yes, there are applied economists who spend their lives trying to find out facts about the world. But facts in and of themselves are uninteresting unless they illuminate economists' view of the world, and can be explained in economic language.

Psychologists begin by observing people's behaviour, looking for generalizable patterns. For example, if people are observed acting as if they are averse to losses, psychologists infer that loss aversion exists. Yes, psychologists have conceptual explanatory frameworks, such as prospect theory, to explain such phenomena,but these theories are not the elegant mathematical creations favoured by economists. And where is the status and prestige?  Look at the psychology journal rankings. There isn't a top-ranked journal with "theory" in the title.

It's a dilemma for behavioural economists. Yes, it's possible to go into the lab or into the field, run experiments, and see how people actually behave. To do what economic psychologists do, in other words. But what do you do next?

My co-author and colleague David Long sets out some options in his research on the nature of interdisciplinarity.

One is the multidisciplinary approach: take some psychology, take some economics, and use whatever works to explain the behaviour at hand. It's intellectual parallel play - working side by side, but not really interacting. 

The danger of multidiscplinary is sinking to the lowest common denominator: combining economists' knowledge of experimental methods with psychologists' understanding of economic theory. The upside potential is arriving at better understanding of real-world phenomena than any one discipline could achieve individually.

Yet multidisciplinarity is not a stable long-run academic equilibrium. Parallel play isn't as fun as working with colleagues who share a common intellectual understanding. So multidisciplinary projects tend to morph into other things...

One possibility is what David Long calls "transdisciplinarity" or, less charitably, academic imperialism. Transdisciplinarity takes the paradigmatic approach of one discipline, and applies it to another discipline's subject matter. Economic psychology, for example, applies the methods of psychology to economic choices. Not every paper in the Journal of Economic Psychology simply reports the results of some laboratory experiment, but a lot of them do.

These papers are unsatisfying for someone who thinks like an economist, that is, deductively, because the results aren't integrated within a coherent overarching model of human behaviour. But they're a profitable line of inquiry for psychologists - intellectually, because they allow for a broader application of psychological methods, and also literallly, because there is a strong demand for research that tell firms how to sell more stuff.

What about behavioural economics? Mullainathan and Thaler describe it as "the combination of economics and psychology that studies what happens in markets in which some of the agents display human limitations and complications."  Those limitations are bounded rationality, bounded will-power, and bounded self-interest, or altruism. The behavioural economics research agenda, according to Mullainathan and Thaler, is "identifying the ways in which behavior differs from the standard model," and "showing how this behavior matters in economic contexts."

Behavioural economics differs from economic psychology in that it is typically conducted by economists rather than psychologists, and published in economics journals. But is behavioural economics just the application of psychological methods to economics, or is it something more? Does it really involve economic thinking?

It might be easier to see what I'm getting at by looking at specific example, such as health care. The standard analysis of health care says that, if users do not have to pay the full marginal cost for their health care, they will demand excessive quantities of health care, and a deadweight loss will result:

 
 

In the diagram above, MB is the consumer's marginal benefit from health care services, which is assumed to reflect the actual benefit they receive from consuming health care, MC is the marginal cost to the insurer, society, or whoever pays in the end of providing the health care services. But the actual cost to the consumer, in this example, is zero. The optimal level of health care is Q*, where the marginal benefits of health services are just equal to the marginal costs. But because the consumer doesn't pay that full marginal cost, he or she actually demands excessive amounts of health care - health care for which the costs of provision are greater than the benefits - that is Q^a. The consequence of this reckless overconsumption of health care is a misallocation of resources, shown as the deadweight loss (DWL) in the figure.

That's the standard analysis.

The problem is: people have no idea what the marginal benefits of health care actually are. So I'm sure that some behavioural economist could show that people's demand for health care was highly influenced by things like

• doctor's recommendations ("I'll see you in six months for another check-up") or
• the quantity of health care demanded by other people ("All my friends have had a complete body-scan, I'd better get one too") or 
• awareness/salience ("Comedian David Mitchell has just been diagnosed with a fatal case of hurty elbow - my elbow has been feeling funny lately, I'd better get it checked out.")

O.k., so how does this get incorporated into the diagram above? Suppose that by manipulating framing, salience, or options, behavioural economists can cause people's demand for health care to go from Q^a to some other point like B or B'? 

Economists' policy recommendations - our ideas about which policies enhance economic efficiency and which ones detract from efficiency - are all based on the idea that individuals know what's best for themselves. That we can draw demand functions and use them to infer the marginal benefits to consumers of goods and services. 

But if people's demands are just a product of framing, salience, and the public prominence of hurty-elbow syndrome, how can we use them to infer the marginal benefits to consumers of consuming health care? How can we make statements like 'the marginal benefits of health care are less than the marginal costs'? How can we make any inferences about the efficiency of different modes of health care provision?

The economic approach to policy analysis is deductive - economists begin with assumptions about consumers' and firms' preferences and behaviour, and these assumptions allow us to say things like 'insurance leads to excess consumption of health care services.' (Or 'firms will pass through savings from the elimination of PST on inputs to their consumers.')

Perhaps there is some way of including behavioural considerations into that framework. Perhaps we can say "the consumer's demand is really what is shown by MB, he just chooses to consume B because he is ill informed - but the happy thing is that, in this circumstances, there is no inefficiency resulting from the provision of insurance."

But that doesn't feel right to me - if a consumer's marginal benefit from consumption is something other than what is revealed by his or her demand I could just make up anything as a marginal benefit curve, and economics would become a completely ad hoc exercise.

Dabbling in economic psychology or behavioural economics is a little like taking the red pill -  you go down the rabbit hole, and wake up realizing that the entire world is an illusion.

How can economists evaluate policy if they can't use people's preferences as revealed by their choices? We still need some way of figuring out which policies help people and which policies hurt people.

For example, opt-out policies have much higher take-up rates than opt-in policies. But in picking between and opt-out and opt-in policy, we still need to know in an ideal world, would we want a high take-up rate or a low-take-up rate? Causing people to opt-out rather than opt-into savings plans makes them save more - but if that causes excessive reductions in current consumption that might be a bad thing.

Once you fall down the rabbit hole, you just have to keep on going. If people's choices are not a reliable guide to their well-being, you have to turn to something else. Ask people how happy they are and measure well-being in terms of happiness. Evaluate health care spending by looking at objective measures of health, such as mortality, morbidity, or survival rates. Chuck out the entire elegant theoretical framework of welfare economics.

That idea has me, for one, reaching for the blue pill - after all, people aren't stupid, so standard economic analysis isn't a bad approximation of the real world, is it?

But I can't find one. I can't get all of that behavioural stuff outside of my head.

I want a purple pill - a merging of the red and the blue - that would allow me to merge behavioural insights into a coherent model of economic behaviour.

(Evolutionary economics - and other research programs that explain why humans behave the way they do - might have some promise as a purple pill).

But I don't know if such a thing is even possible.

 

* Christandl, F., Lotz, S., & Fetchenhauer, D. (2011). The Taste for Fairness – How Ethical Labeling of Consumer Goods Shapes People's Taste Experience. Talk. IAREP/SABE/ICABEEP Conference, July, 13th - 15th, 2011, Exeter (UK).

**Bracha, Anat; Gneezy, Uri; Loewenstein, George (2011) "Relative Pay and Labour Supply"

 

Comments

[W. Peden]

Min,

Sure. But I think necessity helps to explain the lack of non-paternalistic charities.

[Tracy W]

W. Peden, my response remains as before: I don't see how I committed the fallacy of division any more than I committed the fallacy of composition, as I only talked about voters as a whole, not about individual voters.

To ask one more question, are you in favour or against publicly-funded healthcare systems?

[W. Peden]

Tracy W,

You said-

"To take your case, voters voted for publicly-funded healthcare, or at least keep voting for parties that keep continuing it, is it not equally reasonable that we put up with the consequences, including that we might be paying higher costs in some places?"

- which is true of voters as a group, but not of all individuals. The electorate chooses governments collectively, but this does not mean that each and every voter voted for the incumbent government.

"To ask one more question, are you in favour or against publicly-funded healthcare systems?"

I'm in favour with about a thousand provisios. I would prefer that the bulk of healthcare was handled through private insurance, with government's role being restricted to cases where individuals cannot cope i.e. catastrophic cases and provision for the poor. Only where government gets involved would I want any string attatched: if someone smokes and they are charged higher premiums by their insurer, that's fine by me. Your money, your choice.

[Tracy W]

W. Peden - for the third time, I was not talking about individual voters, I was talking about voters as a whole. I defy you to find any point in what I wrote where I asserted that what is true of voters acting collectively must be true of each and every individual voter. I didn't spell out in detail that not every single voter would have voted for something because I am aware that I write long comments already, and I assumed that any readers would be familiar enough with how democracy works to fill in the details. When writing, one always assumes a certain level of background knowledge from the readers (what level varies depending on the audience), and part of that is that one assumes that readers know that what is true of a body as a whole is not always true of every single component. For example, if I say "being hit by a fast-moving car or truck tears the human body apart", I assume that my readers know that I don't expect a nuclear explosion from the individual atoms being rendered asunder.

I notice that you yourself said "In a society where the costs of unhealthy eating are not fully born by the individual but by taxpayers (due to free healthcare) isn't the taxation of dietary fat (and tobacco and alcohol and narcotics) a reasonable request from taxpayers? " You didn't bother spelling out all the ways it might not apply to an individual (examples: a government could fund some healthcare expenses by borrowing, and some taxpayers will likely die before the taxes to pay for the borrowing come due, or the government could borrow and then declare a sovereign default), nor did you spell out that any "request from taxpayers" would likely have some individual taxpayers disagreeing (I have a great uncle who goes out of his way to disagree with public opinion, almost regardless of how many contradictions it involves him in, so I'm confident that someone would disagree) so by your logic you also committed the fallacy of division, or composition, or whatever fallacy you've decided it is now.

But I don't think that you committed any such fallacy. You just stated your argument briefly, and I took it on the basis that it was written, rather than trying to drag out some logical fallacy based on what you did not state. It's rather annoying that you have not extended the same courtesy to me. (If you really honestly did not understand the gap, why not ask me what I meant, rather than jumping to telling me that I was committing a fallacy of composition or division even though I had said nothing about individual voters).

Thank you for answering my last question, I am a bit disappointed though, as you are no longer such a good example of someone favouring the welfare state and a reduction of freedom as I had first thought :(, I can easily imagine some full-on liberal dismissing you as not applicable because you don't favour single-payer for everyone. (But that's my disappointment, not yours).

[the_iron_troll]

Wow! Frances, this article is awesome. I wish I was more up-to-date on WCI, so I could have caught this when it was being heavily commented-on.