Bryan Caplan  

Fixed Costs and Open Borders

Henderson on Piketty, Part 3... Ebola Bet...
Given existing border controls, mild measures to prevent serious contagious disease seem morally acceptable.  Yet the best choice, in my view, remains fully open borders - tear down the walls and make travel between countries as free as travel within countries. 

Appearances withstanding, there is no contradiction between these views.  As I explained a while back, fixed costs imply a straightforward consequentialist case for extremism:

But what if there is a fixed cost of having a carbon tax in the first place? For example, the net expected benefits could be:

-$1,000,000 + $10,000,000*p [where p=P(Al Gore is right)]

The $1,000,000 might be the overhead of the carbon tax collectors, or the costs of every tax-payer who has to fill out a carbon tax form, or what have you. Given this fixed cost, for p<.1, the net expected benefits of a carbon tax are negative. On efficiency grounds, Arnold would be correct to council inaction until p exceeds that threshold.

Every micro textbook tells us that when the price of a good gets so low that firms can't recoup their fixed costs, it makes sense to simply close up shop - or not open in the first place. The same goes for government programs.

Thus, even a consequentialist could consistently favor marginally expanding a program yet prefer the program's utter abolition: Marginal benefits can exceed marginal costs even though total costs exceed total benefits.  On the more reasonable view that government action is only justified if its benefits heavily exceed its costs, the path from fixed costs to abolition is even smoother. 

Are the fixed costs of borders really that high?  Absolutely.  Picture what a pain it would be to erect and navigate checkpoints at every border of every U.S. state.  Imagine what even token internal restrictions would do to U.S. commerce, travel, jobs, and housing markets.  If these checkpoints were already in place, telling border guards to screen for Ebola might be reasonable.  But erecting internal checkpoints to slightly reduce the risk of Ebola is crazy.  And if this judgment is obviously right for internal borders, why is it obviously wrong for external borders?

Comments and Sharing

COMMENTS (7 to date)
John Hall writes:

I think you're confusing things. The argument that we should generally have open borders is separate from the argument that we should close them during times of outbreak.

We have state-managed borders now. I don't think it is reasonable to expect us to eliminate that in the near future. So if you favor reducing the burden of this, you will probably focus on more practical things (like allowing more foreigners to travel and work here without as many burdens).

While I generally favor more immigration, I can see the logic behind closing the borders to foreigners who have traveled to affected (or is it effected) regions and to requiring enhanced procedures (possibly a limited quarantine) for Americans who have recently traveled to those regions. If not for this strain of Ebola (where we can quibble about the cost benefit analysis), then for some future infectious disease that has greater costs. It is unreasonable to expect politicians to not react to the fear that such an epidemic would put in the citizenry. I would rather work (with other countries) to ensure that procedures are well defined and made explicitly temporary.

John Smith writes:

Because the US is not a country that tends to have a significant infectious disease (that are incurable, and have no available vaccine) problem *relative* to the rest of the world.

Africa as a region is.

Jameson writes:

One thing this post does, unfortunately, is lay bare a general weakness in the argument for open borders. I say this as someone who is generally for open borders.

An argument used by at least a few apologists has been to say that if borders between countries are a good idea, why not borders between states in the US?

I think the answer to that question is really not that hard. Whatever differences we see between the states in the US--in terms of wealth, culture, etc.--relative to other places, the US is amazingly uniform for such a large region. The difference between Kentucky and New York is nothing compared to the difference between the US and Liberia.

This might sort of work as a feedback loop. The states are close enough to each other that open borders between them seems perfectly natural. So open borders happens, which actually creates movement between them, which makes them gradually all the more similar. Throw in national broadcasting and you've got a recipe for the complete uniformization of half a continent.

So at best, the argument makes it "obvious" that there should be open borders between the US and Canada. But for most people, the leap is pretty long from Canada to, say, West Africa.

Bill writes:

"Every micro textbook tells us that when the price of a good gets so low that firms can't recoup their fixed costs, it makes sense to simply close up shop - or not open in the first place."

Actually, every micro textbook tell us that when prices gets so low that firms can't recoup their variable costs, it makes sense to close up shop. Anything made beyond variable costs can be used to cover fixed costs.

MikeDC writes:

1. Yes, The costs of internal border controls are obviously higher than the cost of external border controls.

2. Lets focus on those costs (external border controls) vs. the benefits. I basically accept the proposition of free entry but for conspiracy, criminality, contagion and cost (I'm not in favor of free health care and welfare benefits for non-citizens - or citizens- except on an absolute emergency basis).

It's not at all clear to me that including the fixed cost of border controls changes the calculus. Yes, the p of, say, an outbreak of infectious disease or large-scale terrorist attack is relatively small, but the cost of either could easily run into the trillions.

And of course, p rises if it becomes known there's no border control.

Thus, I can't imagine the cost/benefit analysis would ever truly work out in favor of an open border.

David Condon writes:

I think of open borders as meaning no restrictions based on country of origin. What you're talking about is much more extreme. You mean to say we shouldn't even check for criminal history, health status, or current most wanted status, nor for any possessions in their vehicle, even if it's a large shipping vessel. Ebola wouldn't be my main concern; the Mexican Drug War would be.

Between 10 and 20 thousand people die in the war each year. Most of this violence is concentrated in the Northwest; some of it in border towns. Officially, the US Border Patrol only spent $3.5 billion, but the actual cost is much higher; around $12 billion according to one source. However, only 25% of that is spent at ports of entry, which is the high volume area. If there were a 5% spillover effect after removable of border controls (which seems reasonable) then the cost per life saved is $3-6 million. This spillover effect would be concentrated to a few cities. Violence in El Paso, for instance, would escalate rapidly. That's for just one effect. If you add in other factors (you mentioned contagious diseases) or assume cost control measures to reduce expenditure at ports of entry (say by simplifying the customs procedure, especially in non-essential areas) then that cost per life drops further.

ted writes:

Mr Caplan,

In your many, many posts advocating the economic benefits of a far more liberal immigration policy, you never talk about the cost of the social friction that comes with it. How high is this cost? Does it exceed perhaps the benefit? Perhaps by orders of magnitude?

This particular point is relevant to this post. You talk about the cost of checkpoints added to every US state. This is a silly argument: there's no reason to do this, because there's too little social friction generated by US citizens moving from one state to another. This is precisely why we have borders, border control, and immigration policies. It's not just random stuff that we make up in order to waste money.

I live in Switzerland, and I can easily imagine the social cost of importing, say, 5 million people from Bangladesh, Pakistan and Indonesia, having a combined population of ~587 million, so I'm talking less than 1%. If Switzerland were to adopt the Bryan Caplan school of thought, this would be perfectly feasible.

Do you have any reasonable expectation that the Swiss culture and lifestyle would withstand such an event, or would it be more akin to the meteorite that wiped out the dinosaurs?

Do you perhaps have any examples where millions of extremely poor, uneducated people, raised in countries where violence is commonplace, with a medieval religion asserting that non-believers essentially aren't human and have no rights, and last but not least, believers in an honor culture in which the normal answer to your daughter being raped is to kill her, moving into a 1st world liberal, peaceful, largely agnostic and democratic country populated by a different race would result in a successful cohabitation and the large economic gains that you keep mentioning?

Do you consider the British experience, which has imported - from said part of the world - a tiny fraction of population, compared to what's feasible under an open borders policy, a large success? Or a success in any way? Would letting in a lot more make it better?

Comments for this entry have been closed
Return to top