fbpx

Thaler’s Bias

As everyone knows, Richard Thaler has won the Nobel Prize in economics.  Thaler’s prize was mainly for behavioral economics, which built upon the work of the earlier nobel laureate Daniel Kanneman.

This work has been controversial.  Many rational choice economists have strongly criticized it and some of those people have been free market advocates.  But free market advocates come in different stripes and certainly the Austrians would criticize the rational choice perspective as well.  For an interesting exchange on these matters, take a look at this symposium on behavioral economics on the Truth on the Market Blog and the response from Richard Thaler himself.

While I find some of the work of the behavioral economists quite interesting, my main concern is with their failure to apply the same methods to government in general and to administrative agencies in particular.  What could possibly justify this failure other than political bias?

I have been writing about this for a long time.  See, e.g. this post from 2012.  Happily, this failure is increasingly receiving attention.

One of the participants in the Truth on the Market Blog argued that the same behavioral economics assumptions that are applied to private markets would change the law applied to administrative agencies.  He gives the example of equitable estoppel:

To demand private firms hold consumers harmless for their irrationality is to ignore that behaviorally-informed administrative law would, or should (in a world of undivided light and perfect intellectual consistency) do so as well.  Are behavioralists prepared to call for robust equitable estoppel against government agencies with superior information that provide information to consumers?  This, of course, is merely a narrow, trivial example; the entire construction of the Administrative Procedure Act presumed rational – or super-rational – technocrats capable of “solving big problems;” to broach the possibility of individual irrationality at the regulatory level is both a new heresy and a very, very old one.

While this example involves “a narrow, trivial example,” that example is similar to the type of private sector issues that behavioral economics has been applied to.

The bigger problem in the government sector is when government officials must evaluate government policies.  The most significant issue here is that of ideological or political bias.  Ilya Somin, who has complaints similar to mine about Thaler, cites to a study of Danish politicians who evaluated information on two schools.  When the schools were listed as school A and school B, the politicians analyzed the information correctly.  When the schools were listed as “private school” and “public school,” the politicians who favored private schools mistakenly drew inferences in favor of the private school. The same bias was exhibited by politicians who favored public schools.

As David Henderson writes in the Wall Street Journal:

Mr. Thaler has yet to apply in a serious way his theory of irrationality to government officials. Their bad decisions are even worse because citizens bear most of the costs. It would be great if Mr. Thaler explored this area more. Someone should nudge him.

Reader Discussion

Law & Liberty welcomes civil and lively discussion of its articles. Abusive comments will not be tolerated. We reserve the right to delete comments - or ban users - without notification or explanation.

on October 18, 2017 at 12:19:55 pm

The beauty of the book Nudge was to note that 1) context influences people's decisions in predictable ways, and 2) we could use this knowledge to influence people's decisions toward outcomes that studies suggest are most beneficial, rather than the opposite. One critique of the book is that the opportunity to devise appropriate, "elegant" nudges are rather sui generous. Rather than articulating a rule, Nudge contains a collection of quirky exceptions. (Then again, that's the nature of differential equations, too.)

I share Rappapport's view that these same insights should also apply to government. But I suspect that their application would also form a collection of quirky exceptions. I'm happy to consider examples, but the mere assertion that, "well, government isn't perfect either!" isn't a policy.

Rappaport links to Josh Wright’s observation that some administrative agencies offer advice—advice letters, no action letters, private letter rulings, etc.—stating how a given bureaucrat would apply the law (over which the bureaucrat has relatively greater knowledge) to a person’s facts (over which the person has relatively greater knowledge). Typically these letters state that they provide no legal protection if, subsequently, the law actually applies differently to a given set of facts. Wright suggests that people may have a predictable, if erroneous, belief that the letters will provide an accurate prediction of how the agency or courts will apply the law. From this set of facts, Wright concludes that behavior economists should advocate a policy of estopping government (including courts?) from adopting any ruling that differs from the content of the advice letter.

Now, I see the merit of adopting such a policy. But I can also see the rather obvious potential for harm. At the outset, such a policy would permit bureaucrats to supplant executive heads—appointed by the President and confirmed by the Senate—as decision-makers. At a minimum, we’d get radically inconsistent rulings, with no mechanism to reconcile them. If we think that bureaucrats are as fallible as everyone else, why would we want to adopt such a policy?

And that’s assuming good faith. Now, assume I write to my buddy in the General Services Administration to say that my mom always told me that my family owned the land underlying the Capital Building. My buddy writes back to say yup, sounds like a valid property claim. Bada-bing, bada-boom—I own the Capital. Now, I'm generally in favor of making easier for people to acquire capital, but honestly: This would be good public policy—why?

We wouldn’t encounter too many of these problems before agencies started banning the issuance of advice letters. Or (effectively the same thing) agencies would conclude that before issuing an advice letter, they need to conduct all the proceedings required to litigate the matter to finality. And this would be good public policy—why?

Maybe I’m missing the point, but in the absence of a practical remedy, Wright’s complaint looks a lot like “well … government isn’t perfect either!”

Look, there are many circumstances in which the rules that apply to government differ from the rules that apply to private parties. Thus, a party may be able to take your land by adverse possession if you fail to notice their obvious adverse claim over the span of many years. But government is immune from such claims, precisely because 1) no government agent has the same self-interested incentives to defend government land, 2) government land holdings are vast and remote, and 3) the opportunity for collusion between government agents and private actors is vast. Likewise, many government agencies cannot waive conflicts of interest when former government agents litigate against government. Likewise, under the Hatch Act, government employees are barred from engaging in certain campaign-related speech/association, on the theory that politicians would be able to coerce government employees into serving as campaign staff and offering kickbacks as a condition of employment. These policies do not reflect a failure to appreciate behavior dynamics on the party of government agents; to the contrary, they reflect an acknowledgement of those dynamics.

[T]he entire construction of the Administrative Procedure Act presumed rational – or super-rational – technocrats capable of “solving big problems;” to broach the possibility of individual irrationality at the regulatory level is both a new heresy and a very, very old one.

Uh ... no. Congress adopted the APA in 1964 precisely to rein in the power of administrative agencies, especially the New Deal agencies. The law requires agencies to provide due process in conducting litigation, and regarding regulation, to engage in a more public form of record development than is required by legislators. If anyone has a better remedy for individual irrationality at the regulatory level than the APA, again, I'd be delighted to hear it.

read full comment
Image of nobody.really
nobody.really
on October 18, 2017 at 14:06:18 pm

Whoops--Congress adopted the APA in 1946, not 1964. (I've got the Civil Rights Act on my mind....)

read full comment
Image of nobody.really
nobody.really
on October 18, 2017 at 16:01:29 pm

"Uh … no. Congress adopted the APA in 19[46] precisely to rein in the power of administrative agencies,.."

Curiously, I agree with you.

THE problem however is that it has been a spectacular failure, in part due to a SCOTUS that somehow viewed itself as an enabler of the FAS.

"1) no government agent has the same self-interested incentives to defend government land, 2) government land holdings are vast and remote, and 3) the opportunity for collusion between government agents and private actors is vast. "

While this may be valid for the specific issue of "adverse possession" (I actually did this, BTW), clearly it is not true for other FAS actions / intents / motivations. Heck, look to the WOTUS requirements wherein private land is "adversely" (not legal sense) possessed and / or diminished by the Government. In fact, my own backyard is liable to enforcement action under WOTUS.
And you assert that no government agent has an incentive or motive to adversely seize land. Nonsense - the motive may be as simple and forthright as the government employee wanting to rise from a GS-8 to a GS-9 and she views her best means of rising as being perceived as a robust enforcer of the WOTUS requirements.

Gubmint workers are neither more nor less *motivated* by personal ambition than privtae sector employees. Well, then again, perhaps they are MORE susceptible to this trait as they know, that with gubmint backing, they can get away with it.

Then again, I refer you to an informative book by Paul Nollette, "Federalism on Trial: State AG's and National Policymaking in Contemporary Americe" wherein the collusive activity between FAS and *interested* agencies result in a corruption of the Judicial system.

Curiously enough, today's headlines bring news that EPA Administrator Scott Pruitt has prohibited the EPA from engaging in "legislation by litigation" - in some instances, Massachussetts v EPA, the EPA actually celebrated their *loss* at SCOTUS by breaking out the champagne. What does that do to your "unbiased", dedicated public servant thesis?

I have no doubt that you are a good Public servant - BUT it is questionable whether the rest of the FAS can be so characterized.

And it is a dang shame because agencies such as the NTSB are consequently grouped in with the likes of EPA, CFPB, etc.

read full comment
Image of gabe
gabe
on October 18, 2017 at 17:15:47 pm

“1) [N]o government agent has the same self-interested incentives to defend government land….

And you assert that no government agent has an incentive or motive to adversely seize [private] land.

Nope. Read it again.

If we think that bureaucrats are as fallible as everyone else, why would we want to adopt such a policy?

And that’s assuming good faith. Now, assume [the contrary]….

[I]n some instances, Massachusetts v EPA, the EPA actually celebrated their *loss* at SCOTUS by breaking out the champagne. What does that do to your “unbiased”, dedicated public servant thesis?

It does nothing to my “unbiased,” dedicated public servant thesis, because I don’t have one. To the contrary, I explicitly invite the reader to assume that bureaucrats are as fallible as anyone else, and even to assume that they may act in bad faith.

But the Massachusetts v EPA case raising an interesting point that I’ve discussed elsewhere: The Constitution gives federal courts jurisdiction only to address “cases and controversies” between directly adverse parties. If Massachusetts and the EPA were actually on the same side, then the federal courts would have lacked jurisdiction to hear the “case.”

While this may be valid for the specific issue of “adverse possession” (I actually did this, BTW), clearly it is not true for other FAS actions / intents / motivations. Heck, look to the WOTUS requirements wherein private land is “adversely” (not legal sense) possessed and / or diminished by the Government. In fact, my own backyard is liable to enforcement action under WOTUS.

Ok, I’m intrigued. What’s the story?

read full comment
Image of nobody.really
nobody.really
on October 18, 2017 at 19:03:22 pm

The APA was passed to cut back on administrative agency power a bit from the New Deal baseline. But the APA still accepted tremendous administrative power (and for that reason was rewritten by the D.C. Circuit in the 1960s and 1970s). Thus, the APA did accept administrative power, even though it cut back on it a bit.

read full comment
Image of Mike Rappaport
Mike Rappaport
on October 19, 2017 at 11:14:31 am

EPA and Massachussetts were only "ostensibly" on opposing sides, both having the same aim and both agreeing to enlist the device of certain NGO's to initiate the suit. Very well stage managed, I would say.

As for my backyard: Briefly:

Agency determination(s) has decreed that any (and they mean ANY) land, under which, and at a depth of 12(?) inches, water may be present for as little as two weeks per year, are wetlands. (Goes back to an Army corp ruling some 25 years ago). WOTUS decrees that any water that *ultimately* flows into waters of the US is susceptible to government control. My backyard, (hey the Northwest is always rainy and wet) meets the first condition. It also drains into my neighbors orchard, which in turn drains under a gravel road which we maintain and ultimately into a lake, which in turn drains out rhough an overflow spillway and ultimately into a local river (One which I, even with my deteriorating joints( can still toss a baseball across).

So, I guess your bureaucrat friends will now be telling me I cannot fertilize my rose bushes!

read full comment
Image of gabe
gabe
on October 19, 2017 at 12:36:13 pm

Oh, and here is one where gubmint agents once again go after private property in defense of a frog that does not AND CANNOT even live on the property:

https://spectator.org/markle-interests-administrative-overreach-and-regulatory-dark-matter/

Of course, even this is not as bad as the EPA going after a fellow for CLEANING UP AN ILLEGAL TIRE DUMP SITE:

http://reason.com/archives/2012/06/20/everybodys-a-potential-criminal-in-the-e

read full comment
Image of gabe
gabe
on October 19, 2017 at 15:54:40 pm

A couple of thought experiments for the sake of discussion:

1. Assume you have two 72 year old men, same socio-economic circumstances, same cultural backgrounds, and both are given a diagnosis of pancreatic cancer. Both are given the same same treatment options. The first says "give me a fishing pole and a case of beer and I'm good with it." The second says "Mortgage the house! fly me to the Mayo clinic,, M.D. Anderson, Zurich. I want every chance to beat this thing no matter what the odds. If I am going to die it will be fighting."

Which man is being irrational? Is there a bureaucrat that can answer this question, and propose a solution to prevent one man or the other from making the "wrong" choice?

Similarly, consider two professionals, each making $231,000 a year. the first spends her income on things that give her pleasure. The second scrimps and saves and invests to use her resources to generate even more income. Is one of them being irrational? Is a person irrational if they think of money as a means to and end rather than an end in itself, or the other way around?

2. President Kennedy gave his Rice University moon speech in September of 1962. Eighty four months later Neil Armstrong and Buzz Aldrin touched down in the Sea of Tranquility. Imagine that tomorrow, President Trump announced. "We are going back to the moon folks, and it will be wonderful, and you are going to be so proud, it's going to be great. Thank you very much." What are the odds that we would (not could) replicate, with modern technology, a feat we performed almost a half century ago, within the same 84 month time-frame?

The premise of the first thought experiment is that what is rational or irrational is not necessarily objective. What is rational to Tom Steyer is not necessarily the same as what is rational to T. Boone Pickens. It may in fact be the case that being free to make decisions that some academic thinks is irrational is an end in itself, independent of the consequences of the decision. Is it irrational to give up liberty for security, or vice versa? Can this be objectively proven and made policy?

The theory of the second thought experiment is that many bureaucracies, like people in general, tend to become risk averse over time, and that this itself can become a bias that affects their behavior, and their effectiveness. The hypothesis is that priorities and motives change over time, and the determination of what is rational or irrational from a bureaucrat's perspective is also somewhat time-dependent. The most sane, rational and disinterested bureaucrat imaginable will eventually make life hell for some ordinary guy who just wants to live simply and treat others the way he wants to be treated.

read full comment
Image of z9z99
z9z99
on October 19, 2017 at 17:31:38 pm

Z:

Good stuff (as always).

Re: Risk averse: don;t know if you would be interested but at American Affairs Journal there is a great piece on "risk aversion" and factors engendering this behavior. It is called "Capitalism without Capitalists:

A great read and it demonstrates that such behavior is NOT limited to the FAS.

read full comment
Image of gabe
gabe
on October 19, 2017 at 18:27:14 pm

1. Assume you have two 72 year old men, same socio-economic circumstances, same cultural backgrounds, and both are given a diagnosis of pancreatic cancer. Both are given the same same treatment options. The first says “give me a fishing pole and a case of beer and I’m good with it.” The second says “Mortgage the house! fly me to the Mayo clinic,, M.D. Anderson, Zurich. I want every chance to beat this thing no matter what the odds. If I am going to die it will be fighting.”

Which man is being irrational? Is there a bureaucrat that can answer this question, and propose a solution to prevent one man or the other from making the “wrong” choice?

In one sense, this is a profoundly difficult question of public policy, and especially health care policy: What should government subsidize? We can’t afford everything—yet there may be no objective standard by which to judge where to draw the line. Wherever we draw the line, people will complain that the choice was arbitrary and results in hardship to somebody—and they’ll be right. But we’ll need to hold the line anyway.

But in another sense—and in the sense of Nudge—this is an easy question: “Libertarian paternalism” doesn’t make decisions for other people, it just sets the default. People are then free (within the limits of their own resources) to alter the default option for themselves. Thus, a government health plan would establish some standard for how much subsidy to provide under these circumstances, and people would be free to accept those subsidies, accept and add to them from their own resources, or reject them entirely. That’s the beauty of the system.

(Of course, as z9z99 hints, many people have observed that medical interventions under these circumstances often do more harm than good, and that the most knowledgeable consumers (physicians) are especially likely to refuse such interventions. Consistent with Nudge, the original Obamacare plan provided funds to educate people about end-of-life care prospects so that they could make more knowledgeable decisions. Opponents of the bill attacked this proposal as creating “death panels,” and so the proposal was eventually dropped from the bill.)

read full comment
Image of nobody.really
nobody.really
on October 19, 2017 at 18:29:18 pm

2. President Kennedy gave his Rice University moon speech in September of 1962. Eighty four months later Neil Armstrong and Buzz Aldrin touched down in the Sea of Tranquility. Imagine that tomorrow, President Trump announced. “We are going back to the moon folks, and it will be wonderful, and you are going to be so proud, it’s going to be great. Thank you very much.”

WOW. Are you Trump’s speech-writer?

What are the odds that we would (not could) replicate, with modern technology, a feat we performed almost a half century ago, within the same 84 month time-frame?

* * *

[M]any bureaucracies, like people in general, tend to become risk averse over time, and … this itself can become a bias that affects their behavior, and their effectiveness. [P]riorities and motives change over time, and the determination of what is rational or irrational from a bureaucrat’s perspective is also somewhat time-dependent. The most sane, rational and disinterested bureaucrat imaginable will eventually make life hell for some ordinary guy who just wants to live simply and treat others the way he wants to be treated.

It’s surely true that priorities change over time. And I expect that it is also true that we become more risk-averse as we have ever less opportunity for marginal benefit, and ever more risk-avoiding alternatives available. When the US perceived itself to be in a life-or-death struggle with the USSR, and when the opportunity to send a man to the moon was deemed highly prestigious, the US was willing to invest a lot of resources in the effort. Today we don’t have the same incentives to reach the moon, and pretty much anything a human might do up there can be done cheaper by a robot. (Damn automation taking American jobs!) Given all the anticipated costs, benefits, and alternatives, I rather doubt people would want to make the investment.

Is it true that people are always unwilling to engage in risky behavior? People seem to engage in risky sex and drug behavior all the time. People have kids outside of marriage all the time. Etc.

More generally, people’s behavior is driven by hope and fear: what we anticipate we might gain, what we anticipate we might lose, and what alternatives are available. On 9/11 when passengers on a jet airliner perceived that they were on a suicide mission, they were willing to storm the cockpit. It was a desperate strategy with low chances of survival—but still a larger chance of survival than any other strategy. Lacking alternatives, they were willing to take bold, risky action. Yet we rarely observe such behavior—because we rarely find ourselves with such circumstances.

At this point, let me assert that I’m not well acquainted with “some ordinary guy who just wants to live simply….” Let this man’s kid get a life-threatening condition, or live through a natural disaster that destroys his water supply—or, as C.S. Lewis observed, even experience a few minutes of genuine toothache—and then see how much this man values simplicity. I rather expect this man will develop a sudden interest in measures, perhaps even non-simple measures, to eliminate the threat.

So what? Well, much of contemporary government exists to manage these kinds of threats. And when we experience the threats, we become informed about government’s role and grateful for the assistance (or angry with the screw-ups). But when government is so successful as to render the threats all but eliminated—when was the last time you got tuberculosis?—people may lapse into ignorance, and thus impatience with the “nanny state.”

Which again leads to the conclusion about the beauty of libertarian paternalism: It creates a default that, according to best available evidence, will be most likely to promote the best outcome (as perceived by government). But the individual remains free to choose some different outcome using his own resources.

(Ok, admittedly, the individual doesn’t get to avoid the tax burden—unless he’s Hobby Lobby and wants to avoid complying with a policy that the court has ruled as being an legitimate exercise of Congress’s taxing power, but that’s a whole ‘nuther issue….)

read full comment
Image of nobody.really
nobody.really
on October 19, 2017 at 19:21:36 pm

"So what? Well, much of contemporary government exists to manage these kinds of threats. And when we experience the threats, we become informed about government’s role and grateful for the assistance"

And THEN, there goes the *default* position, doesn't it? should relief from a toothache be included in the gubmints default threshold? Should end-of-life counseling be included? And let us not forget governments "incentive" to minimize end-of-life medical costs. Recall that the Great and Powerful Obama remarked that "Let's give granny some vicodin..." instead of providing extended and expensive medical care.

So we are back to the same position: Where shall the gubmint's default position be set?
Then again, why should the gubmint be enabled to set a default at all. Oops, that's right - because some amongst us believe that the government ought to be responsible for everyone's healthcare. To concede this, compels one to accept gubmint default positions.
It is inescapable!!!

read full comment
Image of gabe
gabe
on October 19, 2017 at 22:23:02 pm

So what? Well, much of contemporary government exists to manage these kinds of threats. And when we experience the threats, we become informed about government’s role and grateful for the assistance….

And THEN, there goes the *default* position, doesn’t it?

Uh … no.

Hurricanes and fires have destroyed a lot of homes lately. As a default, FEMA may offer people trailers to provide temporary housing. But if you choose to leave town or remain homeless, I expect they wouldn’t compel you to live in a trailer.

Should end-of-life counseling be included? And let us not forget governments “incentive” to minimize end-of-life medical costs.

Uh … yeah. Nudge would favor giving people information that might help them make a more informed decision—especially when a more informed decision might also be a cheaper decision. Government also informs people on Medicare and Medicaid about the existence of generic versions of medications, for precisely the same reasons. Why is this a problem?

Recall that the Great and Powerful Obama remarked that “Let’s give granny some Vicodin…” instead of providing extended and expensive medical care.

Oh, now I get it: You’re concerned that government isn’t doing enough, and has a duty to provide additional care. That’s mighty compassionate of you, gabe. I’m glad to have your support for expanding Obamacare. :-)

Seriously, I have no idea what you’re talking about. For what it’s worth, Obama restricted access to Vicodin. That is, under Obama the US Drug Enforcement Agency upgraded the status of drugs containing hydrocodone (including Vicodin) to the same regulatory class as painkillers such as Oxycontin, Percocet and codeine. This meant that patients can now get only a three-month supply of the drug, and must see a doctor to get refills. If you’re trying to make some other point, maybe you could provide a link/citation in support?

So we are back to the same position: Where shall the gubmint’s default position be set?
….It is inescapable!!!

Indeed we are, and indeed it is. For example, Reagan signed the Emergency Medical Treatment and Active Labor Act, barring emergency rooms from “dumping” uninsured patients (which was causing a lot of poor women to give birth in hospital parking lots). I surmise you must hate, Hate, HATE this policy because it results in government requiring that people receive some, but not all, of the care they might want. If so, please take it up with Reagan.

read full comment
Image of nobody.really
nobody.really
on October 19, 2017 at 22:27:21 pm

So, I guess your bureaucrat friends will now be telling me I cannot fertilize my rose bushes!

No, they're just saying that if you've GOT indoor plumbing, you might as well USE it. Otherwise, you frighten the neighbors.

read full comment
Image of nobody.really
nobody.really
on October 19, 2017 at 23:49:43 pm

Just a few responses to the questions posed above:

WOW. Are you Trump’s speech-writer?

No.

Is it true that people are always unwilling to engage in risky behavior?

No.

At this point, let me assert that I’m not well acquainted with “some ordinary guy who just wants to live simply….” Let this man’s kid get a life-threatening condition, or live through a natural disaster that destroys his water supply—or, as C.S. Lewis observed, even experience a few minutes of genuine toothache—and then see how much this man values simplicity.

I think you and I have different perspectives on what "live simply means." Many people who wish to live simply have pacemakers, are on dialysis, use cell phones, fly on airplanes and have indoor plumbing. I was referring to simplicity in interactions with others and the government. I would imagine that a person with a child having a life-threatening condition would not wish life to be more complicated.

So what?

I dunno.

read full comment
Image of z9z99
z9z99
on October 20, 2017 at 11:30:47 am

Again, the "hyperbole"

Raising valid concerns about the "default" position (which you initially mentioned) is, of course to be conflated with both opposition to any gubmint assistance and (Drum roll here, please) HATE.

Yep, I MUST be a hater; after all, some nobody (as is typical) has so designated me.

read full comment
Image of gabe
gabe
on October 22, 2017 at 04:19:30 am

Good point! You may find this article of mine interesting:

Berggren, Niclas (2012). "Time for Behavioral Political Economy? An Analysis of Articles in Behavioral Economics", Review of Austrian Economics, 25(3): 199-221.

Abstract: This study analyzes leading research in behavioral economics to see whether it contains advocacy of paternalism and whether it addresses the potential cognitive limitations and biases of the policymakers who are going to implement paternalist policies. The findings reveal that 20.7% of the studied articles in behavioral economics propose paternalist policy action and that 95.5% of these do not contain any analysis of the cognitive ability of policymakers. This suggests that behavioral political economy, in which the analytical tools of behavioral economics are applied to political decision-makers as well, would offer a useful extension of the research program. Such an extension could be related to the concept of robust political economy, according to which the case for paternalism should be subjected to “worst-case” assumptions, such as policymakers being less than fully rational.

read full comment
Image of Niclas Berggren
Niclas Berggren

Law & Liberty welcomes civil and lively discussion of its articles. Abusive comments will not be tolerated. We reserve the right to delete comments - or ban users - without notification or explanation.