Bryan Caplan  

Bechhofer Immigration Bet

The Margins of Moral Weaseling... Sticky wages and sticky fed fu...
If I were in Obama's shoes, I'd give life-long deferred action to every illegal immigrant, past, present, and future.  I'd also immediately pardon all 20,000+ of the people in federal prison for immigration offenses. 

But to belabor the obvious, Obama and I are two very different people.  He's a two-time gold champion of the political Olympics.  I probably couldn't win a election for Chief Dog Catcher.  He's based his career on pandering to the American people; I've based my career on flunking the American people.  I was scoffing at voters' intelligence before it was cool.

The upshot: When I hear that Obama plans to shield many millions of illegal immigrants from the nation's draconian immigration laws, I'm skeptical.  Such an action requires the very iconoclasm the democratic process ruthlessly screens out.  Bold announcements notwithstanding, I expect him to (a) slash the numbers, (b) cave in to public pressure, and/or (c) fail to effectively deliver what illegal immigrants most crave - permission to legally work. 

GMU econ prodigy Nathaniel Bechhofer is considerably more optimistic.  Argument proved fruitless, so we agreed on a bet.  The terms:
If, by June 1, 2017, the New York Times, Wall St. Journal, or Washington Post assert that one million of more additional illegal immigrants have actually received permission to legally work in the United States as a result of Obama's executive action since November, 2014, I owe Nathaniel Bechhofer $20.  Otherwise he owes me $20.
Of course, I hope Nathaniel wins.  But hoping and thinking are two different things.

Update: The noble David Balan has taken the same bet against me for $100.

Comments and Sharing

COMMENTS (17 to date)
Massimo writes:

Are immigration laws draconian? Or the Caplan (Obama?) open border regime? Are property rights draconian and ruthless? Many children are homeless, many children are abused/neglected, many people are in slavery, etc. Why do some people live in nice houses, go to nice school districts, go to nice universities, etc? There are no immigration laws stopping white people from entering Afghanistan or walking down the streets of Newark, but you would be in serious physical risk.

Daniel writes:

Talk is cheap and so is the present value of $20 in 2017.

Kevin Erdmann writes:

I read stuff like this and it boggles my mind:
"The lawsuit argues that Maricopa County officials and Arizona legislators deliberately sought to sidestep federal laws they disagreed with by ramping up state identity-theft laws to nab workers using false credentials. It also claims that Maricopa officials refuse to prosecute the vast majority of employers who violate federal law related to hiring undocumented workers.

Federal law places more liability on employers who hire workers illegally, including civil and criminal penalties; it provides no criminal penalties for undocumented workers, the complaint states. Since 2008, the Maricopa County Sheriff’s Office has conducted more than eighty workplace raids and arrested more than 780 workers, including nine on June 13, 2014, according to the complaint. Those found using false Social Security numbers, including two women who are plaintiffs to the lawsuit, are charged with felony identity theft. During that same time frame, the county pursued actions against only four employers, with civil actions against three."

so these are people marching and protesting for immigrant rights - dedicated people. And it's important to them that people are arrested for hiring these workers. Clearly the immigrants would have a different opinion...strange.

David J. Balan writes:

If you're offering, I'll take Nathaniel's side of the bet too. I have no deep insight here, just my sense that Obama would have nothing to gain from saying he's going to do this and then not doing it. If he wasn't committed to seeing it through, he would have just skipped the whole thing.

Handle writes:

I respectfully request you explain your betting strategy.

On the one hand, you say that betting is a tax on [word adults are not allowed to even abbreviate with two letters on EconLog].

That is, people make a lot of empty or insincere claims for signalling reasons because those claims are costless. But once the cost is raised to where it would 'hurt' to be proven incorrect, then they are much more likely to either express their genuine beliefs, or remain prudently silent.

But all your public bets are far, far below the level where it would hurt either you or your counter-party.

That doesn't make much sense. An uncharitable interpretation is that you wish to achieve the sincerity-perception gains of making your public assertions in the form of a bet, but without actually having to put a meaningful amount of welfare at risk, so you can still easily afford to pay the tax in order to signal.

Is there another interpretation?

Christophe Biocca writes:


A more charitable interpretation is that if you're betting publicly on your own blog, and you followup on all bet resolutions publicly as well (including the ones you lose), then most of the harm is in the form of lost reputation.

Also even a token amount can have an impact on the loser's perception of their own position. Talk is much much cheaper than even a $1 bet, because with one dollar, you have to followup and acknowledge you made a bad call. The bet format also forces you away from generic positions and into the specifics.

I don't believe Bryan believes that Nathaniel Bechhofer is being insincere. I think he just looks forward to being able to say: "You were confident enough to bet on this, and yet I won. Adjust your priors."

Peter H writes:


There is a benefit to publicly establishing a bet for a nominal sum.

First, there are real reputational costs to both parties to being wrong. This is especially true when both parties have some level of public profile at the time of the bet, and which they wish to maintain / expand. Even a $1 public bet from Bryan has far more on the line from a reputational standpoint.

Second, the act of negotiating a bet is a clarifying exercise. It forces two people on opposite sides of a question to come to an agreement on what the exact nature of the question is. It prevents talking past one another, or obfuscating, or cherry picking sources. The parties have to agree on what facts constitute success or failure, and how those facts will be found in advance. That's actually a pretty big deal.

Yancey Ward writes:

I still think it most likely he will act boldly, even if I also think he will be acting unconstitutionally and will deserve impeachment for doing so.

Up until the last few days, I had fully expected him to pardon all the illegal immigrants and fast track them for permanent resident/citizenship status. However, the meme has started to circle in the media that he will act far shorter of that goal by only cementing the status quo (few deportations or enforcement actions)- an action that can only last until January 2017, and that can be undone immediately by the next president. I have to wonder where that meme is arising, and I have a suspicion it is from the administration itself which can see the country isn't anywhere close to being in alignment with the bolder action.

ConnGator writes:

The only thing that gives me hope that Obama will do the right thing is that he will never again run for election. So he has changed from a politician to an "elected official".

Time to step up.

Njnnja writes:

Are you sure that you would prefer to set immigration policy (or just about any other policy, for that matter) by the whims of the executive, regardless of the opinions of large portions of the population? As soon as somebody with another set of beliefs is elected, they would feel free to begin mass deportations. Especially on a repeated game such as deportation where one side requires repeated cooperation but the other only needs to defect once, getting everyone to buy in to the structure of how policy is decided is much more important than the actual policy at any given moment in time. A lot of the things in the American system of government that are so frustrating to getting things done serve an important purpose in the meta-policy realm.

Peter H writes:


The reason the meme has started that the action will only be deferred prosecution and work permits that could be instantly revoked by the next President is that it's the only plausible action within the law.

Obama can't pardon illegal aliens because being an illegal alien isn't a crime under the Immigration and Nationality Act or any other US law. And they can't be given permanent residency because the INA requires permanent residents to be in a valid visa class, to which illegal immigrants can't belong.

All that was ever going to happen was something like what Obama did with the DREAM kids, except for many more people.

Yancey Ward writes:

Peter H,

Just because that is what can be within the law is no barrier- that is the point of acting minus Congress, otherwise, why wait?

If what you describe is what he does, the groups most agitating for this will be greatly, greatly disappointed- they are clearly pushing for and expecting legal status/green cards and citizenship pathway. I think Obama has backed himself into a corner on this one.

LD Bottorff writes:

I would prefer that people who are here without documentation would receive whatever executive or legislative action is required to allow them to work peacefully in our country. However, I have no expectation that President Obama will act in the best interests of immigrants or our country. If he were so inclined, we would have had immigration reform in 2007 when his party controlled both houses of Congress and an immigration bill could have been sent to President Bush. And, executive action at this point is likely to have an extremely adverse impact on the chances for reasonable immigration reform.

Peter H writes:

While the immigrants themselves would be disappointed, their leaders would not be. They're agitating for a path to citizenship as a part of an immigration reform bill, but they know it is not possible as part of exec action.

Obama doesn't seem to be doing this stupidly. He is doing an exec action that is actually probably fully lawful, despite the kvetching from the anti-immigrant right.

Handle writes:

@Christophe Biocca; Peter H:

Thank you for your comments. I'm afraid I don't find them very persuasive regarding my specific concern. Most of the negotiating and reputation-signalling ends can be accomplished through a simple predictive statement on one's blog with good follow-up.

But if the purpose of a bet is supposed to communicate the sincere high level of confidence one has in one's public assertions (in other words, that it's not [word adults are not allowed to even abbreviate with two letters on EconLog]), then it must be for stakes that are far higher than the value of the benefit of the other signals. It should 'hurt' a lot to lose. Otherwise, it may be perfectly worthwhile to pay the small tax on [word].

Here's a clarifying example:

Let's say two of your friends were talking about investing in companies. Friend A says, "I'm pessimistic about Netflix. Everyone is getting into the on-demand streaming game now. Amazon, Hulu, HBO, etc. I think it will go bankrupt." B responds, "No way! Netflix has a loyal customer base. I'm super confident that their stock price will double."

Friend B gets on his blog and tells everyone about the discussion and says, "To demonstrate how confident I am in Netflix's prospects, me and A agreed to place equal and opposing market orders. I am putting my money where my mouth is and investing a whole $20 in the company! Meanwhile Friend A, who disagreed, shorted Netflix, also to the tune of $20."

Like me, I think most people would read that statement with a puzzled expression. Sure, people talk a lot of empty game about stocks all the time.

But when they are really confident enough to invest, they never do so with such paltry sums. If they were genuinely confident and not just blowing smoke, they would be willing to - and typically do - put a much larger amount of wealth at risk, and they would be looking forward to the large reward.

I look at a public bet for $20 with the same skepticism, and since I think my reasons for doing so are obvious to those that are making these announcements, I wonder what else is really going on.

When Caplan announces a bet for $10,000 or more, then I'll have no doubt that he is anything but entirely confident in the veracity of his assertion and has no ulterior motives.

Peter H writes:


Thanks for writing back. I don't think you really addressed my second point of the bet clarifying the terms of debate, which is a useful thing independent of the dollar value. It forces two people on opposing sides to concretely say what they believe will happen, and specify a failure condition. That raises it far above the impact of just writing a blog post, where one can obfuscate or add caveats that mean its unlikely you'll ever have to admit you were just plain wrong.

Handle writes:

@Peter H:

"I don't think you really addressed my second point of the bet clarifying the terms of debate."

That's what I meant by a 'negotiating' - exactly this sort of clarifying back-and-forth.

The ideas of [word] is that someone's actual predictive model is a very flat and dispersed bell-curve with a lot of overlap with the other person's, but they are claiming to have one with a much tighter variance, and probably at a mean that is a more exaggerated distance away from their counterparty's.

This is like lawyers arguing a hard case in an adversarial system, where their internal sense of which outcome is just may be uncertain and in reality very close, but they argue as if they are absolutely certain of polar extremes. There's a good reason the practice of trial lawyering is often associated with [word].

So, by placing bets, you are doing the equivalent of 'raising' in a a game of poker. You are suspicious that they are bluffing with regards to their degree of certainty, and you are testing the other party's hand by imposing a cost if they either (1) don't back down and fold (i.e. admit they were [word]ing) or (2) maintain a gap between their stated model and actual model.

I'm willing to believe that a nominal bet may make a marginal positive contribution to this process - especially making people think harder about what their model really is instead of expressing gut feelings - but it's probably a small one compared to the other signalling factors at play and the opportunity costs of time, research, and cognitive effort. So a larger bet would be even more beneficial, and one should still be suspicious of the beneficial effect of trivial consequences.

For your argument to hold, it would have to be the case that there is a significant improvement in clarity from the $0 to $20 (or 'nominal') level, but that then one hits the domain of rapidly diminishing marginal returns for this factor. I don't think that's accurate.

The main point is that whatever a bet does, it does even better at a higher value by incentivizing greater honesty, precommitment, clarity, and more intellectual effort. To me, the sweet-spot range is probably between a day's and a week's wages.

Comments for this entry have been closed
Return to top