Sunday, November 23, 2014

Behavioral Political Economy

I was interested to read "Behavioral Political Economy: A Survey" by Jan Schnellenbach and Christian Schubert. (HT marginal revolution's irresistible links.)

Context: I have long been puzzled at the high correlation between behavioral economics and interventionism.

People do dumb things, in somewhat predictable ways. It follows that super-rational aliens or divine guidance could make better choices for people than they often make for themselves. But how does it follow that the bureaucracy of the United States Federal Government can coerce better choices for people than they can make for themselves?

For if psychology teaches us anything, it is that people in groups do even dumber things than people do as individuals -- groupthink, social pressure, politics, and so on -- and that people do even dumber things when they are insulated from competition than when their decisions are subject to ruthless competition.

So on logical grounds, I would have thought that behavioral economists would be libertarians. Where are the behavioral Stigler, Buchanan, Tullock, etc.?  The case for free markets never was that markets are perfect. It has always been that  government meddling is  worse. And behavioral economics -- the application of psychology to economics -- seems like a great tool for understanding why governments do so badly. It might also inform us how they might work better; why some branches of government and some governments work better than others.

This nice paper got my attention, since the paper says that's starting to happen.
...Assuming cognitive biases to be present in the market, but not in politics, behavioral economists often call for government to intervene in a “benevolent” way. Recently, however, political economists have started to apply behavioral economics insights to the study of political processes, thereby re-establishing a unified methodology. This paper surveys the current state of the emerging field of “Behavioral Political Economy”
I came away horribly disappointed. Not with the paper, but with the state of the literature that the authors ably summarize.

I notice a lot of theory rather than fact. Stigler and company were deeply empirical.  That theory seems focused almost entirely on individual perceptual and decision-making biases, rather than how people in groups produce bad decisions.

On "theory," you can see where we're going with
We distinguish between a “weak” and a “strong” variant of BPE. The former merely alters specific auxiliary assumptions on either agents’ cognitive capacities or the content of their utility functions, by arguing, for example, that voters not only care about political outcomes, but also about their “citizen duty” when going to the polls, or that they care about other aspects that seem irrelevant from an orthodox instrumental standpoint, like a candidate’s looks... The “strong” variant of BPE goes beyond this and attempts to actually explain (rather than just postulate) motivational and other psychologically informed extensions to the standard model. For instance, it may try to examine the mental processes causing differences in agents’ susceptibility to certain biases
In particular, the first substantive section is
2. Voter preferences and voter behavior
which deals with the age old question, why do people bother to vote? Some of the answers
compliance with social norms..costs of moral behavior...utility gained by expressing one’s democracy appears to provide voters with procedural benefits by letting them participate in the decision-making process, independent of altering results in their favor...voters may overestimate the  probability of their personal vote being decisive...
This little quote suggests some of the character of the enterprise
...expressive utility is by now probably the most widely accepted element of BPE. Hamlin and Jennings (2011) define the “expressive” aspects of voting behavior as reflecting benefits from the act of voting that neither derive from its instrumental nor from its consumptive value, but from its symbolic or representational aspect: not from the act, but from its meaning”
The review covered not one salient fact, other the fact that people vote at all. They are all apparently highly complex ex-post stories. How would one even tell these theories apart?

I was expecting (hoping?) for things like, "XYZ study the FAA's perplexing inability to write rules allowing commercial use of drones, analyzing meeting schedules, showing that PDQ's theories of small group dynamics account for the pattern of indecision,'' or "ABC study data collection by Federal Agencies and how the agencies use control of the data to influence academics to write articles supportive of the agency's goals."  I was hoping even for some good stories of how bureuacratic decisions, lobbying results, bill writing, or anything political/economic can be understood by psychology -- or anything else. Alas, no.

After just one section, it's back to people are dumb, so omniscient bureaucrats should manipulate them (not the authors, to be clear, but the literature they are surveying)
3 Policy-makers ...
3.1 What should politicians do? ..
...attempts to derive policy implications from an explicitly “behavioral” model of how citizens respond to policies. Bolton and Ockenfels (2012) suggest an approach they call “behavioral economic engineering” that tries to integrate such ideas. Harstad and Selten (2015) also model citizens as responding to policies in a non-optimizing way: They are guided instead by “lower-dimensional” rules of thumb that are much easier to apply. The authors argue, quite convincingly [?], that policymakers should take these observations into account, as many important policy recommendations go awry when inadequately modeling citizens as homines oeconomici.
Yes, but we were supposed to be here to see what recommendations go awry when inadequately modeling bureacrats and politicians as homines oeconomici, not as deus-ex-machina, or pater familias.

Another little quote to alert you to the prose style you're going to have to master if you want to take up this topic.
Other approaches extend the process-oriented perspective. ...a theory of economic policy-making should take into account how politicians and citizens interpret political problems.... social communication constructs the beliefs in which concrete policy-measures are rooted.... A theory of economic policy would then need to be open to explanatory approaches from social psychology and sociology, which help to understand how common beliefs are formed and dispersed.
Again, and again, things we ought to listen to in order to construct new theories. Please could we try to study a single fact?

Finally I seemed to find what I was looking for
4. Bureaucrats, regulators, and lobbyists
While there is a very broad literature on how bureaucrats should efficiently regulate the actions of individuals suffering from choice imperfections, [!] so far there is very little research on the biases that regulators themselves may be subject to (see however Kuenhanss et al. 2015 and Tasic 2011 for a first survey). 
Aha! But the hoped-for research isn't there. For example,
 Guided by re-election concerns, they [politiicans] will choose to regulate those risks that are perceived to be particularly salient by the general public at a given point in time, which may direct regulatory resources away from other, objectively more pressing but less salient risks (Jolls et al. 1998)....
regulators face a trade-off between, on the one hand, maximizing social welfare, and, on the other hand, serving their career concerns by following the politicians’ demands. 
That is not terribly deep to put it mildly.

I was hoping for solutions, empirical evaluation guided by theory of what larger frameworks produce better outcomes. Alas what we get is
The authors suggest some institutional remedies that could alleviate the problems of bias in bureaucracies. An example is “de-biasing” by assigning a team within the bureaucracy to play the role of advocatus diaboli, and thereby make sure to get all arguments on the table. Another proposal is to incentivize bureaucrats by making parts of their rewards dependent on long-term outcomes
Madison and Hamilton on separation of powers this is not. (And a great example of my least favorite verb voice, the regulatory passive. "To incentivize." Who is going to do this "incentivizing" please?)

I had some hope finally for
5. Applying BPE: Two examples 
But no. The two examples are
5.1. Libertarian popularized by Thaler & Sunstein (2008)
Back to Gruberism.
5.2. The Long-Term Effects of the Welfare State 
A suggestion that "social norms" can mollify the disincentives of the Welfare State.

The conclusion is ringing. Once again, bravo to  Schnellenbach and  Schubert:
We have also seen that what makes political behavior a particularly suitable candidate for applying insights from psychology is the fact that it exhibits incentive structures that differ markedly from those prevailing in the marketplace: Behind the veil of insignificance, people are essentially free to pursue any kind of non-standard goals. [Amen!]
... BPE models still often display an asymmetry with regard to their basic assumptions, when, for example, “behavioral” voters are modeled as interacting with perfectly rational policy-makers or lobbyists. There is a danger here to introduce a new dichotomy in behavioral assumptions without much concern for the empirical evidence.

The opportunity for deeply empirical, behavioral public-choice economics, studying how individual and group psychology helps us to understand government failures. And hopefully, to craft instutitonal structures that will lead to better outcomes.

Why not? Perhaps, to indulge in a little behavioral ex-post story telling of my own, a behavioral Stigler would be hated equally by the public choice school, which uses rational-actor economics, and by behavioral economists, who seem, in this wide-ranging review, to remain overwhelmed by dumb-voters-in-need-of-our-enlightened-guidance dirigisme.


  1. My biases towards social scientists lead me to expect technocratic sentiment. If a group of capable investigators can discern fundamental human patterns, then add a value framework and you've got something going. However, that runs into my bias towards academics, whom I believe to secretly think, "It can be done, but only I can do it."

  2. Where is your entrepreneurial spirit? You find disappointment. Have you no Ph.D. students or post docs searching for a topic? I see opportunity for research grants, papers, Ph.Ds and possibly even a Nobel prize in your disappointment. The field is undeveloped! Shallow insights and trivial models are publishable. Low hanging fruit is all around you!

  3. ( )

    Cargo Cult Science - 1974 by Richard P. Feynman - Commencement speech at The California Institute of Technology

    The speech is a must-read. It explains the difference between mimicking the language and process of science compared to the real thing. It explains what "cargo cult" means applied to our present lives. [edited excerpts]

    === ===
    I often talked to the people in the psychology department at Cornell. One of the students wanted to do an experiment. Others had found that under certain circumstances X, rats did something A. If she changed the circumstances to Y, would they still do A?

    I explained that she should first repeat the experiment of the other person. Create condition X to see if she would also get result A. Then change to Y and see if A changed. She would know that the real difference was the thing she thought she had under control.

    She proposed this to her professor. He replied no, you cannot do that, because the experiment has already been done and you would be wasting time. This was in about 1947. It seems to have been the general policy then to not try to repeat psychological experiments, but only to change the conditions and see what happens.

    [Later, Feynman tells about the studies done by Mr. Young which determined how perceptive rats really are when trying to test their behavior]

    Young did an A-number-one experiment from a scientific standpoint. His research made maze-running experiments sensible, because it uncovered the clues that the rat is really using, not what you think it's using. And that tells exactly what conditions you have to use in order to be careful and control everything in an experiment with rat-running.

    I looked into the subsequent history of this research. The next experiment, and the one after that, never referred to Mr. Young. They never used any of his criteria of putting the corridor on sand, or of being very careful. They just went right on running rats in the same old way, and paid no attention to the great discoveries of Mr. Young.

    His papers are not referred to, because he didn't discover anything about rats. In fact, he discovered all the things you have to do to discover something about rats. But not paying attention to experiments like that is a characteristic of cargo cult science.
    === ===

    What is the similarity between our government and some superstitious Pacific islanders during World War II?
    Cargo Cult Economics

  4. "The case for free markets never was that markets are perfect. It has always been that government meddling is worse"

    This is a simple point that I believe Milton Friedman used to make over and over. If individuals are all full of flaws and biases and are prone to making decisions that aren't always completely rational, then shouldn't we expect the lawmakers, bureaucrats, and regulators to have all of those same qualities? Throw in the lack of competition and the openings for corruption and favoritism that inherently go with any government program and I think you have a clear reason to at least be skeptical of any new intervention.

  5. I think the argument of the non-libertarian is more that every market is itself the product of government meddling. The government is meddles, just by saying "the sale of stuff is binding" and sending goons around to lock up anyone who treats sales as non-binding. So the question is "how should the government meddle best".

    It's not clear to me a-priori how government bureaucrats - or anyone indeed - can be qualified to say "sales of financial securities shall be binding, that is the best for society" or "people can create limited liability companies and not suffer personal bankruptcy, that is the best for society" and also be substantially less qualified to say that "payment XYZ should be structured as a fine rather than a tax, that is the best for society".

    In short, your argument demands that someone buys into the particular libertarian notions of natural law and property rights for it to make any sense.

  6. I agree with you 100%! I always thought that behavioral economists stated the obvious most of the time. Yes, people do dumb things and behavioral economists are no exception.

  7. Definitely second the disappointment. Might have been better if the authors accepted that they're just re-discovering political science and tried to engage with recent literature over on the poli-sci end of the aisle. There is some work on bureaucratic preferences, strategies and biases, although it's still a developing area of study.

    Their point about regulators prioritizing salient risks isn't bad. The problem of assuming that regulators' default behavior is 'maximizing the public interest' (political science would indicate that this is completely unknowable but, again, they didn't read much political science) is still very common, unfortunately. There's a good edited volume on regulatory capture that gets at some of these issue (Carpenter and Moss 2013); same problem of assuming that regulators default is 'the public interest', although they're nice enough to (implicitly) admit that they're just defining the public interest as whatever regulators do that's free from external influence and holds up in court.

  8. We need a meta-behavioral analysis exposing the biases that a behavioral economist or researcher may be subject to :)

  9. James Q. Wilson's book "Bureaucracy" covers a lot of what you're looking for.

  10. As I see it, the big problem is that economists don't think a lot about how government actually works. This lets proponents of intervention simply point to any market failure and say "Look, we need intervention!" It also lets proponents of laissez-faire wave their hands and say "Nah, government will just mess things up!"

    I've seen some "public choice" research that sort of takes a swipe at modeling the inner workings of government, but the models are too stylized to really test. I think it would be interesting to gather data on bureaucracies and use it to back out the true incentives of the various players involved.

  11. "But how does it follow that the bureaucracy of the United States Federal Government can coerce better choices for people than they can make for themselves?"

    This isn't quite right.

    If there are regulations governing how a company provides 401(k)'s (and there are), there has to be some rule about how people get signed up. And a question naturally arises: Does the company default to keeping employees out of the 401(k) until and unless they specially ask for the paperwork, or does the company default to enrolling employees until and unless they specially ask not to be enrolled? The default choice makes a huge difference, yet there is no option of *not* picking a default rule. By definition, there will be some default rule.

    So the only relevant question is not "does the government pick a default rule that affects people's choices" (it always will), but "does the government pick a default rule that lets more people sign up for the 401(k), while still giving them the ability to opt out?"

    Another finding from behavioral economics is that the complexity of enrollment forms affects how many people make it all the way through. So with the FAFSA, for example, behavioral economics would suggest creating a form that is intuitive and simple to fill out, rather than a form that is dense, complex, and indecipherable to anyone except an accountant. If someone in the government decides, "As long as we're taking the trouble to provide federal student aid, we might as well make the forms straightforward to fill out," that is not a matter of "coercing" anyone's choices.

    1. There you go again... " behavioral economics would suggest creating .."
      Who, please, is going to do this "creating?" And, those same people have for centuries produced "dense complex and indecipherable" forms. Why? Were they just too dumb to think "hey, some nice simple forms might be a good idea" until Thaler and Sunstein came along to shine the light of reason on the process? With billions at stake, is there any chance at all that a government "nudging" process for 401(k) allocations will not steer those allocations to the big high fee banks whose consultants and lobbyists are crawling all over this regulation?

    2. ??? Which do you think has the higher expense ratio, Vanguard VTI or the average expense ratio of the Federal TSP (this is helped by the fact that the Fed plan is reduced by ppl who forefit their free money but whatever). Also interesting that U of Chicago puts regular staff in a defined benefit plan but faculty in an opt-in 401(k)

    3. 1. It is indeed intuitive that indecipherable forms are hard for people to fill out, but it is still useful to study the magnitude of that effect, and to exert pressure on government bureaucrats to make people's lives easier rather than harder. But in any event, making a form more navigable in no way amounts to "coercing" anyone's choices.

      2. High 401(k) fees is orthogonal to the question of whether people sign up for 401(k)'s in the first place. It's not as if one would say, "It's a good thing that people save zero for retirement, because if they saved anything, someone might have an incentive to charge them fees for it."

  12. The sad fact is that those who tend to act sub-optimally (ie, not in their best long-run interest) should face more incentives, not less. For example, public policy prof Mark Kleiman at UCLA is big on giving criminals quick and certain punishment over later, less frequent, longer punishment. In markets, that would mean poor people can get aid quickly if hungry or homeless, but there's an immediate quid pro quo of some sort, something that signals to them 'change your behavior.'

    The behavioralist top-down solutions to stupid human tendencies tries to avoid inequity, but the consequence to eudaimonia is considerable.

    1. De Toquevlille noticed much lower crime in the US vs. France, and chalked it up to much greater likelihood of being caught, despite much less severe punishments. The "economics of crime" school following Becker has argued the opposite, that you can save a lot of police money by imposing rare very severe sentences. An interesting question.

  13. I kept on looking for the eventual reference to the "rational irrationality" of voters in the excellent study made by Bryan Caplan in "The Myth of the Rational Voter" (2007). If I have to encapsulate it in a couple of sentences I will quote the following: Neither, "democracy works because it does what people want", nor "it fails because it doesn't do what voters want". It rather "fails because it does what people want".

  14. I don't see any economics in these studies. It feels like mostly armchair psychology. What's next year's new trend, the economics of anatomy?

  15. I recall a faculty lunch at the Quad Club about 10 years ago. Someone raised the question of why people vote, given that the likelihood of being pivotal was vanishingly small. Gary Becker gave the person a pitying look before turning to Kevin Murphy, who was eating voraciously. "Murphy," Becker growled, "why are you eating those french fires?" Murphy responded with a muffled "Cuz it tastes good!" "Exactly," Becker said. "And that's why people vote." He went on to draw several parallels between cheering for a sports team and supporting a political party, but I don't recall the details.

  16. Is it not David Levine at Wash U who has a book on "Is Behavioral Economics Doomed?"

  17. Would Americans take $18k out of every pocket to occupy Iraq and Afghanistan?

  18. I would say that even Galton knew that people certainly do /not/ necessarily make worse decisions in groups. In fact, sometimes they're rather clever, which is why we have the term "wisdom of the crowds". Now, if you want to argue about a biased sample...

  19. Mario Rizzo: Glen Whitman and I published an article in the Arizona Law Review in 2009 applying the bias theories to the actions of policymakers. When one does that, it is clear that government intervention to debias is highly problematic. Also in 2009 we published an article in the Brigham Young University Law Review showing that the knowledge requirements for debiasing are enormous.
    Please forgive the self-promotion.

  20. You asked where the Tullocks and Co. of behavioral political economy are to be found...isn't Tullock's research on the fiscal illusion exactly the branch of behavioral political economy founded in empirics that you are discussing?

  21. Would you consider the following as a kind of policy recommendation/design based on behavioral biases of the policy-maker:

    a "benevolent government" has time-inconsistent preferences. when she sees a distress financial institution, she faces the urge to bail it out. because she cannot credibly commit not to bail out ex ante. the financial institutions take advantage of the government's limited commitment by investing in "undesirable" risks. Therefore, the government should consider imposing higher equity requirements instead.

    I believe this idea is commonplace, although it is not marketed as a behavioral model as much. Yet the situation can be modeled as a career-political who has to respond to imminent risks to garner votes, and it is a rational choice. The conflict is that the politician is short-lived but the economy/society is not.

    1. I think the general rules of the game is that "behavioral" means you invoke psychology rather than maximization. You are suggesting maximization subject to bad rules of the game, which is usually characterized as "rational" not "behavioral".

    2. Anonymous,

      "Therefore, the government should consider imposing higher equity requirements instead."

      I presume that you are referring to government requiring a higher level of equity relative to the amount of debt outstanding. How should the government value both the debt and equity of the company?

      Should it rely on the market value only? What prevents the market from undoing any attempt by a company to change its debt / equity mix? Company sells equity to buy back debt, market responds by bidding down equity prices and bidding up debt prices.

  22. Why doesn't Booth take the lead and offer the first sociology/economics PhD program? Clearly the psychology/economics behavioralists are ill equipped to apply decision making theories to people in groups.

  23. GIlles Saint Paul uses these issues to argue against utilitarianism, as inherently susceptible to paternalist capture:
    Utilitarian foundations for limited government are shaky insofar as they assume rational and
    consistent individuals. Recently economists’ assumption of rational actors has come under sustained
    attack. Behavioural economics has suggested that people are plagued by irrational biases and
    inconsistencies. The author elucidates how these developments have led to a post-utilitarianism
    which is held to justify paternalistic interventions by the state via ‘sin taxes’ , direct bans or new
    obligations. Individual responsibility is seriously undermined, as is faith in markets. He concludes that
    supporters of individual freedom need to move away from utilitarian reasoning, reassert core values
    of autonomy and responsibility, and define strict limits on the scope of government intervention.


Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.