48 thoughts on “Botswana And Zimbabwe

  1. You are comparing apples and oranges.
    Botswana received independence in 1966 while Zimbabwe was liberated from a brutal racist White rule only in 1980. The Rhodesian government was so despicable that even the Apartheid South African government refused to deal with it. Around 1980, the population of Zimbabwe was 10 times that of Botswana. Botswana is simply living off its diamond wealth. For now the diamond wealth has kept a lid on any potential social tension. There is enough to go around. But there are worries about what happens when the diamond is gone and if the one-party government is capable of finding new avenues. (refer to a recent edition of the Economist).
    Zimbabwe’s story is of betrayal by the British government and of course Mugabe himself. Nobody talks about the dispossessed black labourers who were promised the return of their stolen land. Zimbabwe’s wealth, which is its agriculture remained locked up with a small elite. Apparently, we are supposed to feel sorry for the absentee White commercial farmers. What is happening in Zimbabwe will have repercussions on many African countries and the foreign economic interests in those countries. It worries them a lot. For example, in SA, a promise was made by the ANC government about redistributing land. No progress has been made yet. http://www.johnpilger.com/page.asp?partid=484
    There is a lot of myth-making about Botswana and hate-mongering toward Zimbabwe. The truth is somewhere in the middle.
    I am not defending Mugabe. But facts must be told. Sorry about the long post. This is a slightly depressing story of another African country receiving a bad deal.

  2. I am surprised to note that I tend to agree partially with HiAgain. The political history of the two nations are very different and cannot be discounted to have economic freedom as the only yardstick.

  3. Incidentally, the fact that Botswana has diamond wealth makes its performance all the more astonishing. Diamonds, oil and natural resources in general tend to be antithetical to economic freedom, because governments can enrich themselves without depending on citizens’ cooperation. Note that the other diamond rich countries in Africa do much worse. So Botswana maintaining economic freedom for 40+ years is quite creditable.

  4. The diamond wealth, in some strange way also distinguishes the country from the very fertile Zimbabwe. White colonialists owned huge farms that did cause political resentment among locals.

    After all, Diamond mining is more complicated and is historically less sacred an activity for humans when compared farming on ground.

  5. And by the way, Ajay Shah quotes the author as,
    Botswana went from $700 to $10,000 of per capita GDP over 40 years. This was done through a strategy of `a limited government and one of the freest economies in Africa. (In its 2007 Economic Freedom of the World report, Canada’s Fraser Institute ranked Botswana’s economic freedom on par with that of Belgium and Portugal).’

    Well, Botswana’s average life expectancy fell from 60 to 35 in that period — owing to spread of HIV/AIDS. The nation’s main source of wealth is independent of its people, namely diamonds. Hence, it’s only natural that the per capita goes up.

  6. That. is. not. correct.
    1) GDP has increased in absolute terms as well as per capita terms
    2) The total and per capita GDP has been increasing continuously since 1966, and it was the fastest growing country between 1966 and 1990.
    3) AIDS did not exist before the 80s.

    The success of Botswana and the tragedy of AIDS are independent events.

  7. The population growth chart is almost linear. Now, in that same period, with a similar base, show me one other third world nation which does not have an exponential growth and achieved reasonable economic progress.

    Your argument that the population did increase and that the GDP also increased in absolute terms provide a very weak argument even if you are correct. Given that you did not provide comparative charts or relative growth rates. Botswana may actually have achieved magic, but your defense makes it seem like Jagadguru lived in Botswana for a while.

  8. Just a few pointers – I don’t really know whch side of the debate I fall on.

    36% of Botswana’s economy is mining, mainly diamond. A semi-government owned company owns 4 of the 5 main mines, of which one is now the richest in the world. Agriculture employs 80% people but contributes only 3% of the GDP. With this kind of data (http://en.wikipedia.org/wiki/Economy_of_Botswana), one would expect the Gini to be really high, and sure enough, it is among the 5 highest in the word, at 60+. (http://en.wikipedia.org/wiki/List_of_countries_by_income_equality)

    It would be hard to deny the role of an economically free government. And attributing everything to AIDS is a Niluism. However, it seems plausible that Botswana has a lot to owe to a stable-ish political past (no civil wars etc.) and abundant natural resources apart from diamonds (tourism, mostly for widllife, accounts for 12% of the GDP, for one).

    Also, I think you are blurring economic repression and repression in general. Oil rich countries of the middle east are surely not economically repressive, despite any incentive that the government/sovereign may have to be so. Why? I tend to think political pasts are at play here much more than we give them credit for. The ill-effects that Mugabe’s rule has on the economic statistics of Zimbabwe are also surely not only on account of the economic repression.

  9. Dude, Nilu, You want me to do basic maths for you? Let’s see if you can work the conclusions from these basic factual statements:
    1) Botswana had the highest growth – total growth, not per capita – in the world between 1966 and 1999. India is part of the same world.
    2) Both Botswana and India have had linear growth rates. So your assertion that the per capita growth rate is high because Botswana’s population growth rate was not exponential is simply false.

    What more do you need?

  10. Ritwik,
    There are sides? I don’t have a point of view on this. I am just not convinced yet that Economic Freedom has propelled Botswana. Or that Botswana can be compared to Zimbabwe. The 60/35 statistic is hardly a clincher — but that is just one among the several possible holes that Ajay Shah or his endorser, Ravikiran, seem to ignore.

  11. And, India had a population control program that was at its peak in the 70s and 80s.

    I did not mean economic growth.

  12. There are two Botswanas. The rich one is into diamond mining and so increases per capita income. The poor one is busy having sex, getting AIDS, dying and also increasing per capita income. No wonder this country is an economic miracle!

  13. Nilu, you are trying to send me on a wild goose chase? Your claim is that Ajay Shah made the elementary error of looking at the per capita GDP and ignoring the decline in population growth rate caused due to AIDS. This can be answered by looking at the total GDP growth and comparing with any other country, such as Zimbabwe. Nothing more needs to be done to establish the foolishness of that particular point.
    You can change the argument by bringing in the base and that can be answered by the following numbers:
    1966 per capita GDP:
    Botswana: 78
    China: 101
    India: 88
    Zimbabwe: 281
    2005 per capita GDP:
    Botswana: 5845
    China: 1698
    India: 730
    Zimbabwe: 259

    Here I can point out that Botswana’s GDP numbers look impressive not just because of a low base, but also because it actually achieved an exceptionally high absolute (per capita number)
    You can point out that these are per capita numbers. Yes they are, but there is nothing exceptional about Botswana’s population numbers. All the countries mentioned above more than doubled their population during the period under consideration. So what is your point?

  14. Ritwik, there is a reason why one compares neighbouring countries or similar countries. Botswana is a different country from China. Most of its money comes from diamond mining. Wealth from natural resources tends to get concentrated. So the remarkable thing here is that it did not get concentrated too much. No one is claiming that Botswana followed the absolute best policies possible. No one is claiming that if a country on the Sahara desert follows the best policies possible, it will be the richest country in the world. Initial conditions, natural resource endowments, etc. matter. Which is why Botswana is compared with other countries in Africa, not with the US.

  15. I did not claim anything.

    If anything, I claimed, you cannot look at Zimbabwe and probably have to look at South Africa. But South Africa is again very different and you possibly have to look at Kenya. But Kenya is not even in southern Africa.

    So, I just suggested your claim isn’t convincing. And, as implied already, the per capita GDP claim is not fully answered by looking at Zimbabwe. If you want to make a reasonable case, I suggest you look at another African country of similar size that was also stable. Quote the population figures of Kenya and possibly South Korea even. Or, Thailand.Then compare growth rates adjusted for differences in population, if any.

    If you think all that is not worth your time, which I don’t think it is, I suggest you don’t second claims made without such data.

  16. Ritwik, about the repression arguments: The whole point of the natural resource curse is to say that the presence of concentrated natural resources (like diamond or oil, not like plantations) makes it possible for dictatorships to sustain themselves without causing an economic collapse. Of course, initial conditions matter – Norway became a democracy before it struck oil and it stayed a democracy. But natural resource-rich countries find it difficult to turn democratic because:
    1) They can sustain a small ruling and military elite with their resources which will effectively repress the larger dirt-poor population.
    2) They can sustain a small ruling and military elite and throw crumbs at the larger moderately poor population to keep them from turning so dissatisfied that they will revolt.

    Which of these will happen depends on many other things. But it is instructive to note that the diamond-wealthy countries in Africa are more like 1) than like 2). Viewed against this backdrop, Botswana’s achievements are remarkable.

  17. “If you want to make a reasonable case, I suggest you look at another African country of similar size that was also stable. ”
    Why? Your argument is that the per capita GDP is skewed by people dying of AIDs causing a decline in population growth. The graph establishes that people did not die in sufficient numbers to have an effect on population growth till the late 90s. That is all we need to know to put your argument at rest. Everything else is you trying to obfuscate the issue.

  18. Ravi, I am a bit sleep deprived, so forgive me if I am thinking even less clearly than usual. I have also not gone through all the comments so I might be repeating some points.

    1. Your point about there being economic freedom despite diamonds is valid, but the causality is either reverse or orthogonal to the argument being made.
    2. What this article is doing is a rudimentary level of “matched sample analysis”. In that methodology, usually N entities are chosen. Then an effort is made to find a match for each entity. Ideally, they should be similar on all other variables apart from the focal predictor variable (here, economic freedom). So you have a sample size of 2N and then you compare the 2 N-sized samples and see if there is a difference.
    3. The matched sampling is done to avoid alternative explanations. The sample is sized N, because for one or two entities, there may be something unique or idiosyncratic driving the outcome (here, per capita GDP), and a decent sample size smoothens this out.
    4. Even if all the other variables are not matched, that can be controlled for in the analysis.
    5. In this case, we have one country versus one country, and they are not matched on variables that might provide alternative explanations. So while the discussion is good enough for a newspaper or magazine article, it does not have even an iota of rigor.

  19. The graph establishes that people did not die in sufficient numbers to have an effect on population growth till the late 90s.

    That doesn’t even need to come into play. Zimbabwe’s life expectancy also dropped to 35 (from about 56), so whatever supposed effect AIDS would have on driving up per capita GDP would act for both countries.

  20. I even tend to agree with Gauravan.

    You can’t look at Zimbabwe for economic growth because of its political history. Now, having ruled Zimbabwe out on one count, holding on to it for another seems needless. Especially since you don’t know how much to discount for economic growth even if the population growth can be considered.

    So, you need to look at some similar, stable country with a similar base for population. And then discount possible deviations in population if any. Do something similar for economy. Or some such thing.

    You may even end being correct.

  21. And is geographical proximity the best, or the primary indicator of similar conditions? Initial conditions, natural resource endowments, matter. So do political histories. Which is why Botswana is better off understodd while being compared to a country that has never had prolonged internal strife. Which pretty much was HiAgain’s point, before you latched on to Nilu and the AIDS chase. Zimbabwe has a population of about 13 million, Botswana about 1.7 million. Zimbabwe has not one diamond mine to speak of, Botswana has >25% production of diamonds in the world. All this constitues ‘similar’? Only because it is ‘neighbouring’? Botswana is natural resources (mineral ) rich, Zimbabwe has always been primarily agrarian (>60% of the economy despite a 51% contraction since 2000) And Mugabe’s main tyranny has been the land reforms – seriously, what’s the point of comparing botswana to Zimbabwe? (http://en.wikipedia.org/wiki/Economy_of_Zimbabwe)

    More importantly, my main point with the statistics was that in the case of this particular country, looking at the AVERAGE economic growth over 40 years was missing out almost ALL the important statistical nuances. When one wishes to prove/support a causal assertion using a statistical argument, is it not advisable to explore the statistics a little to check the strength of the initial assertion?

    And this – “Wealth from natural resources tends to get concentrated. So the remarkable thing here is that it did not get concentrated too much”. Not too much?? After having the 5th highest Gini in the world, more than almost all other natural resources rich country in the world. I wonder how you use the phrase ‘did not get concentrated too much’.

    There are several juxtapositions to prove that free economic policies aid growth. Zimbabwe vs Botswana seems to be an extremely flimsy one at that. If it is growth over 40 years that is the main argument, a little bit of history may be handy.

    ” 1930-1970
    As weighed growth rates, economic planning performed very well during the early and mid-1930s, World War II-era mobilization, and for the first two decades of the postwar era. The Soviet Union became the world’s leading producer of oil, coal, iron ore, and cement; manganese, gold, natural gas and other minerals were also of major importance.

    In 1930, the Soviet economy was estimated at about 10 percent of the size of the USA. By 1950, it was estimated at about 20 per cent of the size of the USA. Nevertheless, the USSR had already emerged as the largest economy in Europe then surging ahead of the UK, Germany and France in a single generation.

    In 1961, a new redenominated Soviet ruble was issued. It maintained exchange parity with the Pound Sterling until the dissolution of the USSR in 1991. By 1970, the Soviet economy had reached its zenith and was estimated at about 60 per cent of the size of the USA. ”

    Plus, Botswana has full current and capital account convertibility and is financing infrastructure through forex. We have seen in ’97 where that path leads to.

    2nd lowest life expectancy, HIV/AIDS pandemic, lagely non diversified economy, staggeringly high Gini, mismatch of asset and liability durations on the foreign capital front. The country is a strutural and fundamental mess. A couple of years of slump in the diamond industry and/or lack of control on AIDS, and we

  22. If you want to show that economic freedom leads to better macroeconomic indicators (even gini coefficients), you dont need this cumbersome matched sampling technique.

    Get data on something like the “economic freedom index” for all countries for a number of years, ideally at least 20. Let the Y’s be the macroeconomic indicators. For the X’s, use a lagged measure of economic freedom, as well as the year-by-year change in economic freedom (positive and negative and what magnitude). Run a regression of the Y’s on the X’s. (look up “panel data” on wikipedia).

    Of course, it is an obvious question with an obvious methodology so someone must have already done it.

  23. …might realise that we we were standing at the peak while whistling about the ascent.

  24. The point that Acemoglu or Easterly have made about Africa is that correlation is often mistaken for causality. You see Botswana, a tiny ass country of 1.7 million people, see a nice per capita number thanks to diamond mining, economic freedom, and jump to the conclusion that economic freedom is the medicine for Africa. Easterly´s books suggest that such simple reasoning has not worked in other countries in africa, and sometimes in the same country in the next generation.

  25. Gaurav’s suggested technique with lagged independent variables and changes rather than absolute measures is interesting, even if obvious. Especially if run on the Gini coefficient.

  26. If someone is willing to tabulate all the relevant macroeconomic indicators and the freedom index or some such measure for even 40 or so countries for 20 or so years, I will happily do the analysis. I have a WINBUGS code ready for a panel data analysis for another project anyway. Will require very little tweaking.

  27. Ritwik, I have already written the counter to all regression analysis in Africa. Paul Collier did all the regressions possible on Africa and wrote a book – Bottom billion or whatever and Acemoglu laughed at him, said he was a moron for mistaking correlations for causality.

  28. Do try and read Easterly´s “The elusive quest for growth”. When I was a youngster full of certainties about economic freedom, Paul Romer handed me the book, and I have become sadder and maybe wiser about economics and africa.

  29. avataram, all a regression does is show if there is a relationship between “some” X and Y. Cross-sectional regression can suffer from the correlation-causality trap you mention. Lagging the independent variables does reduce that risk, though does not eliminate it completely. What we seem to be talking about is a global sample, not just Africa.

    Anyway, the point seems to be that freedom leads to prosperity. Simply a main effect. That could turn out to be true for a sample of countries from all over the world, and yet the point you make about Africa would still be valid. Because we may have support for the main effect X->Y, but Africa could have a set of unique Z’s which makes the meaningful effect as being X*Z->Y, and that may not be a positive relationship. In other words, presence of high levels of Z (whatever the Z’s are, Easterly and the like might know) renders the X->Y relationship meaningless.

  30. Shit, my mega comment got deleted. I hate your blog’s formatting etc.

  31. I am shocked to find that I agree with Gauravan entirely. That brings me to my point, a study on a global sample like that is a better indicator of freedom –> prosperity than desperately hunting down a Zimbabwe-Botswana comparison.

  32. No need to be shocked. Like all senile people, you have good days and bad. Today seems to be your good day, so you actually seem to understand a) logic b) economics c) statistics, and hence agree with me. 🙂

  33. By the way, a general short ramble which in my humble opinion is useful. When I first learnt, or in fact “truly got” the following point – If an interaction effect i.e. X*Z->Y is significant, then the main effects X->Y and Z->Y are completely meaningless – it struck me as so simple and yet so profound. Sadly, I notice that 95% of the social science research papers I read seem to forget this fact, and seem to be obsessed with finding just main effects, thinking that just controlling for other factors makes it all OK.

  34. Ohh, now GS reminds of a brilliant case we had in Marketing I, with all that rambling about interaction effects. Too much. I doubt if you have ever made this much sense since you left for the US. Just reminds me that mathematics should be the be-all and end-all of all education. Kill social science.

  35. I lived in Gaborone, Botswana from September 4, 1995 to February 8, 1997, and traveled the country extensively from Francis-town to Lobatse and made many trips to Johannesburg South Africa. I offer these observations:
    Like the United States it was trying to control migration from their neighbors in Zimbabwe, Zambia and South Africa. The country that nobody wanted as the gained independence to the economic jewel of the region. As one observer put it “There is economic electricity in the air here in Gaborone.
    I attribute this radical improvement to
    1] A democratic government that has the interests of its peoples at heart. The sharp rise in housing available to the masses of people, the construction of many schools, colleges, a university and many medical clinics and hospitals
    2] The contracting of services to experts from other countries until natives of Botswana gained enough expertize to do it themselves. Our dear friend was from Ghana. He ha d a PHD in agri-eonomics and he was under contract to the Botswana government with an objective to get the most production out of the land and their cattle. His contract is now over and he has gone home to Ghana. A trained native has taken his place,
    3] When i was there some 1100 natives has been sent to the United States to gain expertize in a variety of field such as engineering and psychology. Their expenses including tuition were paid in full by the Botswana government. They had to promise when they finished they would return to Botswana and use their new found expertize in helping their own people. At that time only two of the 1100 failed to keep their promises to come home.
    3] Wise negotiations with de Beers to get the most of their diamonds as a natural resource as they came out of the ground and as they were polished and sent to market.
    Neighboring countries would do well to follow Botswana’s example.

  36. Here you have some facts that must be clarified before continuing using such wrong statistics:

    1) Botswana’s life expectancy is NOT currently the second lowest in the world. In fact it reached a minumum in the early 2000s up to 50.7 years in 2007 and is still increasing. Take into account the effects of AIDS and hence note that the standard of living for healthy people without AIDS is quite high compared to those of countries with similar AIDS-infection percentages (e.g. South Africa, Namibia, Swaziland, Lesotho, Zimbabwe).

    2) Gini coefficient stands at 63? yes, but in 1993. You have to be there in Gaborone or Francistown and see with your own eyes how properous the country has become during the last 15 years and how a large middle-class has emerged from what used to be one of the ten poorest countries in the world. Of course, poverty is still there in the surrounding slums, but you can see that improvement has been achieved during the decade.

    3) Percentage of the population living below $2 in 2003: 55.5%, ridiculous large? No, $2 are not the same in USA or Botswana. In Botswana that means that 30% of the population lives below the NATIONAL poverty line. For comparision with different countries, Belarus 27%, Belgium 15%, Colombia 49%, Georgia 31%, Haiti 80%, Macedonia 30%, Uruguay 27%, Spain 19%.

  37. I would be interested in your comments about the Zimbabwean situation now, can people even compare it to Botswana. I mean 4 million starving now, people dead, beaten u and kidnapped for supporting the opposition, no water, people literally living in houses flowing with raw sewage, no electricity, no food, no education for children, no teachers at school. The list is endless and we still are expected to hang on and blame it on Britain and other Western powers, phew and the country has massive mineral welath by the way. And you say Mugabe is receiving a bad deal, thanks man.

Comments are closed.