Musina-border-crossing-006[1]

Risking death to get into South Africa

The supposedly horrible socioeconomic consequences of South African apartheid’s abolition are sometimes used as a cautionary tale against open borders. But this story of Ethiopians and Somalians risking life and limb to get into South Africa serve as a potent example of how much people are willing to risk in search of a better life:

41 young Ethiopians suffocated to death inside an overcrowded van in Tanzania. With the aid of human traffickers, they had been hoping to start a new life in South Africa.

Some ended up paying with their lives, while those who survived will be deported back to their home country.

…Most refugees from Ethiopia and Somalia are economic refugees, says Getachew. But others flee also from war and political persecution. 32 year-old Mohad Abdul is among those who fled to South Africa because of violence in Somalia….Integration in South Africa was relatively easy for Abdul. He quickly obtained a residence and work permit. Today he is a businessman in Johannesburg and watches closely as more and more Somalis and Ethiopians flock into the country.

In no other country are there so many asylum applications. In 2011 alone, there were 100,000 applications. The authorities can scarcely keep up with processing them.

There is no accounting for such reckless risking of life without considering the place premium: the same person doing the same job in one country can earn dramatically more than he or she would in a different country. The Somalian fleeing lawlessness is almost certain to be more productive in any other society in the world, since that country will at least have a half-functioning legal system. It is not difficult to imagine that even countries in less anarchic states might not offer their citizens the institutions conducive to productivity and prosperity which do exist a country or two away.

The international wage discrimination created by closed borders is literally the worst that has ever been measured. That conclusion may sound shockingly strong, but when you consider that there are Indonesians who literally migrate to Australian jails (because to them it’s better to be in a jail in Australia than free in their homeland) or Afghans who risk being shot to death to get into Iran, what’s shocking is how blind we are to the suffering which closed borders create.

The image featured at the top of this post is of a mother with her child crawling under the South African fence bordering Zimbabwe, taken by Themba Hadebe for the Associated Press in 2010 and published in The Guardian.

John Lee is an administrator of the Open Borders website. Liberal immigration laws are a personal passion for him. See all blog posts by John.

Twitter 

11 thoughts on “Risking death to get into South Africa”

  1. For context, we should also recall how South Africa came to have such a large place premium in the first place relative to the rest of sub-Saharan Africa in the first place. Its distinctive feature was increased European settlement and sustained oppressive colonial rule by those settlers.

    South Africa has done fairly well in the last 19 years since apartheid in retaining its economic vitality (although with high unemployment, increasing inequality, and falling life expectancy) and a strong advantage over its neighbors in desirability to African immigrants, as Grieve Chelwa has discussed on this blog.

    But Zimbabwe kept pace economically for 20 years after the end of white rule before suffering catastrophic economic decline, converging with and then falling beneath the average for sub-Saharan Africa (excluding South Africa, which makes up close to 30% of SSA’s GDP). Now Zimbabwe has no place premium and instead of attracting immigrants, refugees and other emigrants are fleeing Zimbabwe.

    http://en.wikipedia.org/wiki/Economy_of_Zimbabwe

    A change in the composition of the electorate, and other power blocs like the military, in a society will tend to change its institutions. But such changes should be expected to accumulate stochastically over time, rather than an instant first-day jump to a new equilibrium. The long-run outcome for South Africa is not yet clear.

    1. I’m not sure it is clear that we should expect electorate changes to be more significant over time rather than particular situations, crises, or leaders leading to collapses of good institutions. For instance when Zimbabwean leadership attempted to hold a referendum that would change the constitution to enable a faster-tracked land reform actually failed. http://en.wikipedia.org/wiki/Land_reform_in_Zimbabwe#2000s

      In other countries, changes in the electorate don’t seem to be vital for drastic changes in institutions for worse. The Weimer Republic and post-WW2 West Germany didn’t have significantly different electorates but the former resulted in a break down in democracy and a radically expansionist regime while the latter has been one of Europe’s most successful states. Even with changes in the electorate in Europe, East Asia, and the United States over the 20th century while the expansion of regulation and the welfare state may be less than economically ideal, they’ve hardly been disastrous either. Even with the effective expansion of lower skill and quite possibly lower IQ members of the populace overwhelming the old voting demographics over the course of the past two centuries, countries with good institutions have more or less tended to keep them. Thus a much stronger force than the make up of the electorate seems to be that good institutions tend to lead to a virtuous cycle to maintain themselves while bad institutions tend to reinforce themselves as well. Situations like Zimbabwe, the Weimer Republic, or post-WW1 Italy are weird and generally seem to have more to do with particular personalities/crises than a build up of the effects of an electorate change (note: this also explains while many revolutions or coups which drastically change who’s in charge lead to surprisingly little difference in economic outcomes with exceptions like Chile again being weird not the norm).

      1. I did say “and other power blocs like the military,”.

        “Even with the effective expansion of lower skill and quite possibly lower IQ members of the populace overwhelming the old voting demographics over the course of the past two centuries, countries with good institutions have more or less tended to keep them.”

        Could you give some examples of the countries you had in mind as undergoing major falls in average levels of IQ (or education or skills measures) while retaining their good institutions over the last two centuries?

        “Situations like Zimbabwe, the Weimer Republic, or post-WW1 Italy are weird and generally seem to have more to do with particular personalities/crises than a build up of the effects of an electorate change ”

        The retreat to weirdness and unpredictable individuals has problems. Fascism in Italy and Nazism in Germany occurred at the same time, along with the growth of fascist and socialist movements elsewhere. Noise in leadership is inevitable, but what that translates into depends on circumstances. This is what I meant by “stochastic.” Mugabe may have been personally less economically sensible and more ruthless than the average African leader, but policies like expropriating the white commercial farmers were politically attractive because his base of support (a significant part of the electorate, but also in the armed forces and party thugs) and incentives were much changed from the Rhodesian government. Mugabe is able to get support from his base to stay in power even though he needs to cheat to win elections.

        And the kind of person who can take power and become popular is affected by the knowledge of potential followers (voters, party members, soldiers), and by the distribution of characteristics in the population eligible for leadership.

        1. “I did say “and other power blocs like the military,”.”

          I realize, my last note was meant to acknowledge that.

          “Could you give some examples of the countries you had in mind as undergoing major falls in average levels of IQ (or education or skills measures) while retaining their good institutions over the last two centuries?”

          Certainly. Take Italy for example. Northern Italy has been prosperous for centuries while Southern Italy has been poorer for centuries. The former is more urbanized, educated, and wealthier and likely higher IQ (though finding reliable data on that last one has been tougher for me) and yet has lived under a national government with the latter with completely open borders for nearly a century and a half. The northern regions, despite all this, haven’t really declined. Indeed in terms of economic success it’s hard to tell the difference between northern Italy and western Germany (http://epp.eurostat.ec.europa.eu/cache/ITY_PUBLIC/1-21032013-AP/EN/1-21032013-AP-EN.PDF). The Italian government right now is having problems, but blaming that on a century and a half ago unification seems questionable. Even more importantly, the rich parts of Italy have not been brought down by the poor parts. Parts of Italy that have been rich for centuries still remain some of the wealthiest areas of Europe and that honestly isn’t likely to change in the current euro-crisis.

          Another example is the United States. Even ignoring 19th century immigration, consider the increased power of the lower IQ African American population due to the civil rights and voting acts. It’s been 50 years and the US has retained strong institutions and it’s status as the wealthiest non-city-state/non-oil-dependent economy in the world. Did African Americans take over? No, but your point on averages should still mean we should still expect some relative decline rather than effectively none (and cases of other countries catching up on some metrics seem to be more based on their improvement than our decline, ex Japan and China).

          Countries seem to get into stable positive or negative equilibria and demographic or power structure changes seem less important in knocking them out of these than particular crises or leaders who seem capable of achieving these changes without prior demographic/power structure changes. The two Koreas are possibly the best example of this. Both the highly undesirable North Korean set of institutions and the rather desirable South Korean set thereof have managed to maintain themselves for decades and despite major changes in the global context which created the divide in the first place (namely the fall of the USSR and end of the Cold War).

          “The retreat to weirdness and unpredictable individuals has problems. Fascism in Italy and Nazism in Germany occurred at the same time, along with the growth of fascist and socialist movements elsewhere. Noise in leadership is inevitable, but what that translates into depends on circumstances. This is what I meant by “stochastic.” Mugabe may have been personally less economically sensible and more ruthless than the average African leader, but policies like expropriating the white commercial farmers were politically attractive because his base of support (a significant part of the electorate, but also in the armed forces and party thugs) and incentives were much changed from the Rhodesian government. Mugabe is able to get support from his base to stay in power even though he needs to cheat to win elections.

          And the kind of person who can take power and become popular is affected by the knowledge of potential followers (voters, party members, soldiers), and by the distribution of characteristics in the population eligible for leadership.”

          With the rise of socialist and fascist movements the best explanation appears to be the impact of the First World War and later the Great Depression, not any sort of demographic change (which given the locking down on border controls in multiple countries during this period were particularly stable).

          But with the Zimbabwe example I still stand behind the idea that generalizing the effects of demographic change from it is probably unjustified. If Mugabe had greater incentives to engage in a disastrous policy of forced land redistribution (as opposed to the relatively benign form that he’d been doing for two decades), does this mean his government was more likely to behave in a disastrous manner than the white Rhodesian government? I say not necessarily. The white government had it’s own potentially disastrous pit falls as well. Consider that a policy of wide-scale forced sterilization and eugenics against the black population would have been much more likely under an explicitly racist white government than Mugabe’s government. Can we infer that was a less likely eventually than Mugabe’s because it didn’t happen? Again not necessarily. As you yourself noted, two decades of policies not significantly worse than the white Rhodesian government’s didn’t prevent one major collapse. Is it truly inconceivable that the white government would have never had a leader that decided sterilization was the answer to the white population’s problems? Successful attempts by powerful minorities to begin wiping out majorities are not completely unheard of (for an example see the actions of the early white settlers in the New World, though of course they did have disease to help). Extremely bad policies are more likely because of a history of bad policies or a particular crisis. Demographic changes seem to me to have a bigger effect on changing the nature of though disastrously bad policies than changing the chance of some form of crazy policy.

          1. Thanks for the Italy pointer, I haven’t looked into that one.

            The American South is a tricky case, because after the extension of the franchise you had extremely racially polarized voting. In southern states with white electoral majorities where whites overwhelmingly vote Republican while minorities overwhelmingly vote Democratic, the Republican primary is the biggest determinant of who takes office, and resembles the old regime.

            http://en.wikipedia.org/wiki/List_of_U.S._states_by_African-American_population

            This effect of racial polarization is one of Bryan’s favorite mechanisms in arguing that low-skill immigration won’t expand the welfare state, but it is something of a “double or nothing” game: if racial polarization amplifies the impact of the ethnic majority then it reduces the electoral impact for groups that will be minorities, but increases the impact of larger scale immigration that changes the majority. And polarization varies across times and circumstances.

            ” Is it truly inconceivable that the white government would have never had a leader that decided sterilization was the answer to the white population’s problems?”

            Certainly not. But we see such disastrous activities more often in Africa between African groups than, for example, by whites against other ethnicities in Latin America, and we didn’t see it in the few observations we have in Africa.

            “Successful attempts by powerful minorities to begin wiping out majorities are not completely unheard of (for an example see the actions of the early white settlers in the New World, though of course they did have disease to help).”

            The European colonists committed horrible atrocities (although the pre-invasion levels of violence were very high too), but disease was the primary factor:

            http://en.wikipedia.org/wiki/Columbian_Exchange
            “It is estimated that upwards of 80–95 percent of the Native American population died in these epidemics within the first 100–150 years following 1492; the most affected regions in the Americas lost 100% of their population.”

            1. To Italy, you are correct a majority lives in the North (though the percentage does seem to have shifted more in the North’s favor since unification it does seem to be the case that the North has outnumbered the South even pre-unification: http://www.tacitus.nu/historical-atlas/population/italy.htm), thus if a majority is required for the major negative impacts then that example may not have much relevance.

              To the American South, you are correct on a state level, but we also have examples on a city level where in many places blacks form the majority and take over city government. Atlanta is one such example. Now the Atlanta metro has many white-controlled minor cities that make up most of the Atlanta population, but the city itself controls and has maintained or expanded some of the wealthiest (as well as admittedly some of the poorest) neighborhoods of the city. Areas such as Buckhead (http://en.wikipedia.org/wiki/Buckhead_(Atlanta)#Economy) and Midtown (http://en.wikipedia.org/wiki/Midtown_Atlanta). Black controlled city governments haven’t led to decline and disaster in Atlanta though and kind of definitive look at the effects of black versus white leadership would include more cities and more research than I am capable of doing at the moment.

              To the seeing more atrocities in black ruled Africa, have you considered the colonial period in this? The Belgian Congo, German ruled-Namibia, Portuguese rule in Guinea-Bissau, the various atrocities committed gaining rule over Africa in the first place all need to be considered against the whites as well. If white-ruled South Africa and white-ruled Rhodesia weren’t as bad as black-ruled Zimbabwe or black-ruled DRC, that doesn’t mean white-ruled Congo or Namibia were better. Certain countries seem to have gone down-hill post colonial rule, but others could be argued to have gone up quite a bit.

              Finally you are completely right disease did most of the work (though European did occasionally intentionally spread disease most of it was unintentional), but the Spanish in particular did learn that enslaving Native Americans led to even faster death rates and yet continued unabated. With a weaker unintentional disease factor the losses would not have been nearly so great, but Europeans certainly didn’t mind helping the diseases along.

              1. About states, there is a paper by McDaniel (2006) that estimates state-level IQ means and correlations with gross state product, health, violent crime, and government effectiveness as measured by the Government Performance Project:

                http://www.people.vcu.edu/~mamcdani/Publications/McDaniel%20(2006)%20Estimating%20state%20IQ.pdf

                Here is Jones (2006) for national level data for comparison:

                http://mason.gmu.edu/~gjonesb/JonesSchneIQ

                I don’t think that the relative population numbers destroy the northern Italy example, it’s still a good piece of evidence in favor of the irrelevance of education and test scores in the national population for institutions.

                Someone pointed me to the informal economy, said to be large in Italy. The Mafia are concentrated in the poor southern regions, suggesting more undercounting of GDP there, but some said most tax evasion was in the richer north, although it was not clear how much of that reflected the larger northern population. So I’m not sure about the net impact of informality. I’m also not sure about the relative power of national and regional and local governance in Italy.

                “To the American South, you are correct on a state level, but we also have examples on a city level where in many places blacks form the majority and take over city government.”

                This would be helpful, although this would be only one of three levels of government, which one would need to control for in a small sample, or do a proper large study. Restrictionists might reply with Detroit and other cities.

                “Black controlled city governments haven’t led to decline and disaster in Atlanta though and kind of definitive look at the effects of black versus white leadership would include more cities and more research than I am capable of doing at the moment.”

                It does seem like a good dataset might exist, perhaps tucked away in the tables or appendices of a paper focused on some other city regression. Running measures of urban governance against electoral demographics. Maybe another day.

                “To the seeing more atrocities in black ruled Africa, have you considered the colonial period in this?”

                Yes. A lot of murderous violence in total, but looking at Steven Pinker’s work it seems like the precolonial societies were even worse in rates of death by violence on average. Frequent murders and smaller scale skirmishes and wars added up to a higher death rate than the fewer big bloody atrocities of the colonial empires. And because of the secular decline in violence we need to grade on a time-adjusted curve. In economic terms colonialism seems to have increased colonial local total GDP, e.g. agricultural output, even if population growth meant that this didn’t translate to much per capita GDP growth.

Leave a Reply