The wealth of nations, and democracy

Yesterday I went to a fascinating guest lecture at The Treasury, by Stephen Haber, a professor at Stanford, who is currently visiting New Zealand as the Reserve Bank and Victoria University professorial fellow in monetary and financial economics.  Haber was the co-author of the very stimulating recent book Fragile by Design: The Political Origins of Banking Crises and Scarce Credit, a comparative study of banking systems. I’ve been meaning to blog about this book for months.  To the extent that Calomiris and Haber are correct (and I’m not sure how far that is) the case for intrusive banking supervision and regulatory restrictions – of the sort the Reserve Bank is increasingly adopting  – in countries like New Zealand is materially undermined.

But yesterday’s lecture was on something quite different.  His topic was “Climate, geography, and the origin of political and economic institutions”.  It is continuing work, building on the earlier observation that stable democracies –  of which there have not been many – cluster in regions of moderate rainfall.  In the words of the abstract of a 2012 working paper:

Why are some societies characterized by enduring democracy, while other societies are either persistently autocratic or experiment with democracy but then quickly fall back into autocracy?  I find that there is a systematic, non-linear relationship between rainfall levels and regime types such that such that stable democracies overwhelmingly cluster in a band of moderate rainfall (540 to 1200 mm of precipitation per year), while the world’s most persistent autocracies cluster in arid environments and rain-forests. This relationship is robust to controls for the resource curse, as well as to controls for ethno-linguistic fractionalization, the percent of the population that is Muslim, disease environment, and colonial heritage. I advance a theory to explain this relationship, focusing on differences in the biological and technological characteristics of the crops that can be grown in different precipitation environments. Variance in the biological and technological characteristics of crops generated variance in producers’ strategies to solve problems of scarcity, giving rise to variance in the distribution of human capital and institutions associated with the protection of property rights. Democracy was more likely to thrive in environments in with a high level and broad distribution of human capital, and with institutions that protected property rights. I test the theory against a unique cross-country dataset, a comparison of democracies and autocracies in antiquity, and a series of natural experiments.

The current work takes these ideas further and builds on the work of plenty of other scholars trying to better understand what accounts for the widely divergent, and apparently deeply-rooted differences in outcomes across countries.   Haber’s claim is that climate and geography explain between a third and a half of the variance across countries in GDP per capita and in where countries stand in democracy rankings.  Here geography is not the ideas of remoteness from the rest of the world I was toying with last week, but something more about the ability to grow, store, and transport (and thus trade) food.  Places with flat land and navigable rivers or coastlines score well.  Rocky valleys don’t.

In Haber’s story, certain climates and geographies pre-condition societies to developing market-based institutions and effective but limited governments that eventually lead to greater prosperity, innovation, and democracy.  In his story, for example,  England is a place where grains can be both grown and readily stored, and transported, and where there are few extreme climate shocks that might historically have threatened whole societies. Trade requires effective enforcement of private property rights.

Others places are more naturally favourable to the development of strong central governments, which can discourage innovation.  Haber here cites both Egypt and China, and argues that the propensity to flooding naturally lead to strong central governments as a risk-management device (the biblical story of Joseph, central managing grain reserves, featured as an example of the “insurance state”). Such societies discourage any innovation that might threaten the perceived self-interest of the state.  Others places again  –  think of Pacific islands –  are prone to severe adverse climatic events, but also have climatic/geographic conditions that don’t allow the production of storable foods (eg grains), and so there are no incentives to develop the institutions that protect property rights and the development of markets.  Stealing vast quantities of grain in northern Europe would have been very valuable – it lasts a long time –  but stealing bananas in Fiji would not.

I don’t lay claim to any great expertise in this area, but for what it is worth much of what Haber had to say rang true in understanding some of the differences across some countries –  England vs Egypt/China vs Vanuatu for example.  But then again, it is not so many centuries since China was the richest (per capita) economy in the world.  Plenty of scholars try to explain the subsequent great divergence.

But I was uneasy about two things.  First, democracy is really rather a new thing, at least in its current forms.  Perhaps in a  hundred years from now it will be the established and standard form of governance everywhere, in which case Haber’s work might be useful only in explaining in which countries democracy developed first.  Then again, perhaps democracy will prove to have been a short-lived fragile flower, and the pool of countries with democratic systems could look much smaller than it does today.  After all, 80 years ago many of the countries of Europe were far from democratic, and if anything democracy might have looked to be in reverse.  Who is to say it couldn’t happen again?  Perhaps it just reflects my economics training, but differences in wealth look more persistent that differences in how much democracy there is, and is probably a better focus.  Apparently, Taiwan is less prone to adverse climatic shocks than mainland China, but the contrast –  for now at least –  between a rowdy democracy on one side of the strait, and the Communist Party’s rule on the other side, cautions against too much geographical or climatic determinism.

But closer to home,, I was uneasy was about whether his story –  whether about democracy or prosperity – could usefully explain much about a country like New Zealand (or Australia, Canada, the United States, Uruguay, Chile, Argentina).   By world standards, each of these countries is pretty well-off –  the last three less so than the others.  The first four have been among the world’s most democratic countries, and even the Latin American countries haven’t exactly been China –  Uruguay and Chile had some well-established democracies, with some brief unfortunate interruptions.

The climate and geography of these countries is much the same as it was 200 years ago, or 500 years ago –  ie in Haber’s terms well-suited to the emergence of democracy and prosperity.  And yet I don’t think there is anything in the pre-history of the territories of those modern countries to suggest that the indigenous societies in any of them had the nascent qualities that were about to lead to the emergence of societies that were among the most democratic and prosperous on earth.

Of course, it isn’t that climate is irrelevant. But the channel is different than Haber seems to recognise.   British settlers were willing to settle en masse in New Zealand or Canada because the climate and geography were conducive (by contrast, when British missionaries went to west Africa in the 19th century it was not uncommon for them to take coffins with them, so high was the mortality rate).  But what would modern day New Zealand or the United States look like if, for some reason, there had been no international migration?  Haber’s hypothesis seems to suggest that they should have been rich and free.  I rather doubt it.  Unfortunately, there are no natural experiments –  countries with good geography and climate that remained largely unsettled by Europeans.  Perhaps South Africa is the nearest example, and I wouldn’t have thought it was particularly supportive of Haber’s case.

There were opportunities in New Zealand, Australia, Canada and the United States which people from rich and successful countries (mostly the UK, but not exclusively – see Quebec, or the Spanish influence in the US) forcefully took advantage of.  The riches and success provided Britain with the military and political strength to enable new societies to “invade” and largely replace the cultures and institutions that had been in those societies previously, but which had not developed technologies that enabled them either to flourish, or to fend off the influx from Europe.  There probably wasn’t too much unique about Britain –  had the Napoleonic Wars gone the other way, more of the colonies of settlement might have been French rather than British –  but the influxes at the time when lowering transport costs made mass seaborne migration feasible were inevitably Northern European. It isn’t a particularly attractive picture, but that is what it was –  those who had developed wealth and power (and the associated successful institutions) displaced those who had not utilised the potential of those climatically and geographically favoured lands themselves.

I’ve been attracted to work of Bill Easterly in this field, who has asked “Was the wealth of nations determined in 1000 BC?” He found that differences in technology levels across countries were remarkably persistent over time, even going back as far as 1000 BC (although his focus was on differences in 1500AD).  But as his work developed, he took explicit account of the role that large scale immigration played in transplanting technology and institutions from one geographical location to another.  People make a difference.

Here is his scatter plot of the relationship between the technology levels in each country and current GDP per capita.  New Zealand, Australia, Canada and the US are in the top left hand corner: poor technology in 1500, but high incomes now.


And here is the follow-up chart, incorporating the technologies in 1500 of the peoples who now live in those countries.  Mass migration wasn’t an issue for most countries, but it certainly was for New Zealand, Australia, Canada and the United States.  If you look carefully, you’ll spot a NZL towards the top right hand corner of the chart  (the other colonies of settlement are buried in that cluster too).


I’ve often been critical of the Reserve Bank and even The Treasury on this blog. But credit should go to them for hosting a fascinating visitor such as Haber, and to The Treasury for yesterday’s open seminar.

13 thoughts on “The wealth of nations, and democracy

  1. Hi Mike

    Thanks for this blog. I was planning to go to the Haber seminar but couldn’t make it due to an unfortunate mishap on my mountain bike! Did it just cover the 2012 working paper you have linked above? Did you raise with him any of the points in your blog above? If so, what was his reply?



    • Tony

      He is still doing fresh work in the area, so it went well beyond the previous paper. He had a rich collection of fascinating material – I know a lot more about the pre-colonial Kingdom of the Congo, for example, than I ever did before. I did ask about the NZ/Aus/Can/question (what if they had been settled by N Europeans?) but we were getting to the end and I have to leave very promptly. My impression is that he has directly looked at the colonial effect only via institutions – in other words, who was the colonial power – rather than people. So when I aksed him, he agreed that he treated Sierra Leone or Gambia the same as NZ, since both had Britain as colonial country, even tho we got lots of Brits and the institutions, while Gambia really only got the formal institutions.


  2. DNA testing recently of secluded populations around the world has tracked human history from a single country in Africa. The very first wave of migrants travelled from Africa, trekked right through to Indian and those same ancient DNA marker can be tracked to the Australian Aborigines and up to east asia, mongolia and into the rest of Europe.

    Human mass migration is not anything new and is not isolated to a few countries. The world is a hotch potch of human migration.


    • No doubt true, but not that relevant to the point here. For most countries since say 1500 large scale inward migration had not been a big factor, but for Latam and the US/UK/Can/NZ it was transformative – as those Easterly charts show.


    • The first humans that left Africa was due to climate change. As the world got colder, the seas receded, Africa got very hot and dry and the deserts expanded resulting in the first human migration. Climate did play a major part and have resulted in the many different races as they adapted to their respective new environments, where it was colder and received less sun, humans lost their dark african pigments and became white.


  3. Perhaps Japan comes closest to the control you envisage of a country in the climatic zone but free of Northern European settlement and influence. You could argue this example either way to support or deny the hypothesis?


    • Yes, Japan does fit his story (altho proneness to natural disasters is a counter – Japan has lots of them. As I noted, I think his story seems to do quite a bit of useful work for the older “advanced” civilisations (for want of a better word). The ideal test would have been if, say, Ontario, New Jersey, Tasmania or the South Island had been left unsettled.


      • An unsettled South Island means that we would have the wealth of the Moa still running around but unfortunately they got eaten by the mass migration of Maori.


  4. The great thing about the British was that they brought law and order to their respective colonies. The difference with the British and other colonial powers is that the British brought with them a team of highly skilled administrators. Other colonial powers tend to rely on their generals and officers to administer and soldiers are really lousy administrators.


  5. Hi Michael,

    The geography story to explain GDP per capita differences is not new. I cannot remember, but J Sachs was probably the one who investigated it empirically. You are right it cannot explain the large differences in productivity between NZ and OZ or the US and Canada. They are close to each other geographically and equally distanced from the rest of the world. There were productivity studies about the Southern states of the United States before and after the introduction of the AC. Technology explains the differences very well. But I wonder what is so good about something that explains 1/3 of the variance, R squared of 0.33 is very small. What is the relevance of this research to monetary policy?


    • Weshah

      I’m sure it was of no relevance to mon pol or to the RB at all, but Haber was primarily here for his Fragile by Design stuff. Treasury seem to grab each professorial fellows for a guest lecture, sometimes (as this time) on a quite different topic.

      As you say, he isn’t the first to look at geographic factors but my impression is that he has now done it in more depth than most, and with a distinctive angle (the production and storability of grain, intereacting with propensity to natural disasters.


  6. It’s an interesting theory. I think Haber would benefit from talking to UC Davis economist Greg Clark, who has written about how population characteristics are a significant factor in economic/socio outcomes (eg. Clark ‘A Farewell to Alms’ 2007). Clark sets out that certain ‘middle class’ characteristics (patience, self-control, passivity, and hard work) increased in prevalence in England in the lead up to the Industrial Revolution.

    Steve Hsu, involved with the Beijing Genomics Institute, observes a similar process was likely occurring in China too based on Clark’s data.


    • Thanks. I suspect Haber would argue that those characteristics were themselves endogenous outcomes of the climate/geography etc. But to argue thus invites my response, so why hadn’t they developed in NZ, Canada, the eastern US, southern Australia, Chile/Argentina, Uruguay in the pre colonial settlement era.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s