Bryan Caplan  

Geography Is Policy

Greg Mankiw's Deficient Treatm... What does Greece tell us about...
For the last twenty years or so, Jeffrey Sachs and co-authors have been arguing that institutions and policy matter less than most economists think.  The harsh reality is that geography has a huge effect on countries' economic success.  From what I've seen, the Geography Matters camp is on to something: Even after correcting for national ancestry, high absolute latitude and coastal access continue to have huge economic payoffs.  In fact, geographic effects are much more robust than the effects of national ancestry

Social scientists who accept the power of geography tend to get pretty pessimistic about development.  If poor countries adopted the institutions and policies of rich countries, they still wouldn't do very well.  Few go full fatalist.  But they do lose hope that economic reforms can quickly transform the world.

And that's where the geo-centric economists are completely wrong.  Contrary to their own self-image, their view is radically optimistic.  Consider the extreme scenario where geography is the sole determinant of national prosperity.  Is there anything mankind could do to swiftly raise per-capita GDP?  Absolutely: Move people from poor countries to rich countries.  Is that the kind of thing that policy can change?  Again, absolutely: Legalize movement from poor countries to rich countries.  How much would that accomplish?  Given the draconian regulations now on the books, such deregulation would swiftly transform the world.

In the real world, of course, geography isn't the sole global problem, so deregulating migration isn't a full-blown panacea for global ills.  But if Sachs is remotely right, this deregulation is the closest thing to a panacea we've got.  Bad geography only retards human progress insofar as humans remain in locations with bad geography.   And once it's legal, humans will vacate the bad areas on a massive scale

To be fair, Jeff Sachs has written in favor of freer migration:
When high-income, high-productivity countries close their national borders to migration, they are denying the rights of individual migrants to seek improvement in their own conditions, and are also blocking a vital channel for improved global productivity. A global migration regime should favor migration both on account of the global efficiency gains and on account of the human right of individuals to seek their preferred residences (see Carens 2013, for a cogent ethical analysis from a human-rights perspective).

The global regime should pay special attention to emigration from the world's most impoverished regions, with special attention to those suffering from intrinsic barriers to development due to geographical, ecological, climatological, or other intrinsic factors. Migrants from such regions face the greatest need to emigrate but also the greatest obstacles. They tend to be poor, less educated, and with few familial or business contacts in high-income countries to facilitate their migration. These are the boat people drowning in the Mediterranean.
But to the best of my knowledge, Sachs never quite makes the fundamental point: Migration policy is the co-factor that makes geography important.  "Geography matters a lot" does not imply "Policy doesn't matter so much."  Instead, it implies that "Migration policy matters a lot."  Geography is not destiny, but opportunity.

Comments and Sharing

COMMENTS (13 to date)
Lars P writes:

Can technology negate geographic disadvantages?

I'm thinking of how air conditioning has lead to economic upswing in the US South. When more of the economy moves to the net, ports should get comparatively less important. I'm sure there are more examples.

Luke J writes:

I am still advocating for digging a channel from the Gulf of Guinea to the Indian Ocean. Surely this can be accomplished with less money than we have already poured into Africa, and without the controversy‚Äč of immigration.

Toby writes:

I have another policy: nuke all the places with good geography! If geography matters so much and poverty can be ended with migration policy, then nuking all the places with good geography should be practically irrelevant. But if a policy of nuking all places with good geography can ruin those places, then so can migration policy. I would therefore modify the statement,

Bad geography only retards human progress insofar as humans remain in locations with bad geography.


Bad geography only retards human progress insofar as too many humans remain in locations with bad geography and too few humans come to locations with good geography from places with bad geography.

In short migration policy is a question of finding out when benefits will exceed the costs.

The fact that geography seems to matter a lot does not mean that migration policy should be more strict or less strict. That's a separate question.

John Hall writes:

No reasonable person would dispute the impact of geography. For instance, there is a lot of economic activity by rivers and oceans, but much less activity in cold, mountainous areas. It's seems like you would need to account for transportation costs. Historically, there was more development by rivers and oceans because the transportation costs were lower. Much harder to drag food up a mountain than down a river.

So geography can help explain, partially, why New York has more economic activity than Vermont. But there's a whole bunch more to explain why the U.S. is richer than Africa.

"Geography is not destiny, but opportunity."

True. However, that opportunity may not be something good. I'm treading on fine water here but what about Iraq? If we believe the argument that Bush went to war in the Middle East to safeguard America's oil then geography has been a negative for Iraq in recent memory. So to has diamonds for Botswana or a multitude of other examples. I think there can be no doubt that institutions and policy matter just as much, if not more, than geography.

EclectEcon writes:

In Canada we don't really have barriers to migration between the regions. We do, however, have policies that strongly inhibit migration. There are equalization payments from "rich" provinces to poorer ones, and there are myriad gubmnt programmes designed to help out the unemployed and the poor in the "have-not" provinces. These programmes provide strong incentives for people to remain in the "have-not" provinces rather than move.

Yaakov Schatz writes:

I do not understand how geography could be a substantial factor when we look at Hong Kong and its surroundings, Singapore and its surroundings, Israel and its surroundings, north Korea and south Korea, Cuba and Florida, etc.

Roger McKinney writes:

Mises destroyed the nonsense of geographic determinism in 1949 in Human Action. Essentially he wrote that it assumes that people only react to stimuli and can't think. It's part of the mainstream econ view of human nature that says people are just automatons. Reality is much different. People plan and choose means to achieve those plans. They think, calculate, learn, and recalculate. They overcome obstacles. And there are just too many exceptions to take it seriously.

If a regression shows support that is merely evidence of misspecification.

Michael S writes:
I do not understand how geography could be a substantial factor when we look at Hong Kong and its surroundings, Singapore and its surroundings, Israel and its surroundings, north Korea and south Korea, Cuba and Florida, etc.

Not completely familiar with Sachs' argument. But from my own knowledge of development econ, favorable geography can be viewed as a necessary, but not sufficient, condition for economic growth. You need it to grow, but having it does not mean growth will automatically follow.

Hugh writes:

Switzerland is at a major disadvantage with all those mountains.

I feel so, so sorry for the Swiss.

TexasCapitalist writes:

[Comment removed for supplying false email address. Email the to request restoring your comment privileges. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

IronSig writes:

Hugh has his finger on the contention here. I need to do more thinking about it, but the Sachs view conflicts with Hernando de Soto and Deirdre McClosksy that institutions and Liberal ideas underpinning them are necessary for civility and economic growth.

gda writes:

So if we move the population of, say, sub-Saharan Africa to New York State and the province of Ontario we should expect great (or at least better) things from that transposed population?

I think not. At least, not for 50,000 years or so, with perhaps a few non-homo-sapien hominids thrown in to spice up the gene pool.

Perhaps I am misunderstanding the geographical (aka "magic dirt") argument, but this concept seems stunningly naive. Laughable, even.

Lets recall also that the institutions of the "favourable" geographical areas (which some consider the primary difference between rich and poor countries) were also introduced into the "unfavourable" geographical areas.

Unfortunately, once colonialism became unpalatable, and the colonialists exited those "unfavourable" areas, those institutions were squandered and/or corrupted and/or dismantled for the most part.


Return to top