Our Word is Our Weapon

Bjorn Lomborg is a puzzle. His main argument seems to be that trying to cut carbon emissions to reduce future global warming would be a big waste of money compared to spending hypothetical fortunes on fighting disease, hunger and poverty in poor countries. But as far as I can see, whenever he crops up in the world’s media, he seems far more interested in promoting the first part of that argument - let’s do nothing about this one global problem - instead of the second - let’s do a lot about all these other global problems.

If he was that serious about persuading us to spend tens of billions more of our own money on fighting disease and poverty in poor countries, would Lomborg have been so enthusiastically embraced by various prominent free-market fundie think-tanks who campaign against overseas aid? I doubt it, and nor have I seen any evidence of Lomborg trying to change their minds on aid - instead, he apparently just tells them what they want to hear about climate change, which is to do as little as possible. No wonder they like him.

But his argument seems pretty straightforwardly fallacious anyway. If he is as worried as he claims to be about malaria, then he should be worried about the very real potential for “many millions of cases of malaria occurring as a result of climate change which might not have occurred earlier”. If he cares about hunger as much as he says, he should be worried about the devastating impact climate change could have on the regions of the world already most beset by hunger and malnourishment. And the same thing goes for preventing armed conflict:

When the apocalyptic horsemen of famine and pestilence appear, war can’t be far behind. Decreasing pastoral lands, decreasing available tillable land, decreasing wild game, and decreasing available water all add up to more strife, Scholes says. “Subtropical dry, arid areas are going to be a huge source of conflict over the next half-century because we still have very, very high population growth rates in those areas, very low economic growth rates, and deteriorating environment,” he says.

So trying to reduce future global warming is complementary, not contradictory, to efforts to tackle, poverty, disease and so on. And in another way, too: Lomborg makes much of the estimated costs of reducing emissions compared to the costs of tackling all these other problems, as if the historic lack of funding for them was unproblematic, a matter of ignorance rather than indifference. But as anyone knows, aid budgets are notoriously inadequate and notoriously vulnerable. What we need is some sort of big, fairly steady revenue stream to fund action against poverty and so on. But what could that be? William Cline, who wrote the Copenhagen Consensus paper arguing for action to tackle climate change, has an idea:

When it comes to dilemmas for choosing between the environment and today´s poor, moreover, it seems to me the debate has missed a key consideration. A carbon tax would raise revenue, and the lack of revenue is a key obstacle to achieving many social goals. Global revenue from my optimal carbon tax . . . would raise $1.1 trillion. That can buy a lot of schooling and medicine. So rather than coming at the expense of social spending, in practice it is quite possible that carbon abatement could facilitate social spending. Because of revenue realities, action in the climate change part of the Copenhagen Consensus agenda can perhaps more realistically be seen as complementary to, and enabling of, action in the other issue areas, rather than competitive with them.

Sounds like a winner to me. But not, for some reason, to Bjorn Lomborg.

Thanks to the single most useful website in the world, my text-dump from that Dani Rodrik paper is now readable, at least in the sense that all those horrible line breaks are gone.

Just for info, I have fixed the blog design problem pointed out recently by Michael. Turns out it was a bug in Internet Explorer, so like I said just another good reason to switch to Firefox. The fix was found on the ever-useful Wordpress forums.

William Easterly’s chapter on “Freedom versus Collectivism in Foreign Aid” in Economic Freedom of the World 2006 Annual Report does not diverge greatly from his recent modus operandi, which seems to consist mostly of:

• Implying that Jeff Sachs is a communist, and
• Ignoring the large and growing body of evidence suggesting that aid is good for growth

But fortunately Easterly throws in a couple of new elements this time to keep us interested. For example, he is obviously aware of the criticisms that have been - with some justification - made of previous versions of the EFotW index, noting

there is a large problem of potential reverse causality—richer people might demand more economic freedom. Critics of the measures published in Economic Freedom of the World also might allege that they are constructed by those with strong prior beliefs that economic freedom is associated with prosperity and, hence, the indices might be unconsciously skewed to give higher scores to countries known to be success stories.

He then goes on to say how he intends to debunk these claims:

I show an instrumental variables regression in Table 2.1. Since the institutions of economic freedom originated in Europe and then spread to other temperate regions where Europeans settled (with some exceptions), I use distance from the equator as one instrument for economic freedom. Since different legal traditions (especially the British) favored economic freedom while others did not (obviously the socialist legal tradition), I use legal origin as another set of instruments for freedom. The test statistics on the validity of the instruments are mostly satisfactory, and we still show a very strong association between economic freedom and per-capita income.

This strikes me as a bit daft. Looking at it objectively, distance from the equator is a very poor instrument to use for economic freedom, since there are many examples of temperate countries which are or have been pretty poor on the freedom front, with a particularly rich crop of good examples coming from East Asia.

Of course, if you are trying to show that economic freedom is good for growth, then distance from the equator is actually a rather good instrument to use, since being in a temperate location seems to give a strong and independent boost to your growth prospects irrespective of the level of economic freedom. By contrast, countries close to the equator are more likely to be land-locked and far from rich markets, and have poorer conditions for agriculture and higher disease burdens, and so automatically experience lower growth.

But then again, there’s another good reason for Easterly to ignore the lessons of geography for economic development, namely the identity of the man who arrived at or at least popularised many of them: Jeffrey Sachs.

I’m a sucker for a beautiful chart, and New Economist is right - this one from the IMF’s World Economic Outlook chapter on commodities is a humdinger.

China does rather stand out there, doesn’t it? This seems to provide graphic support for Dani Rodrik’s argument that “China has somehow managed to latch on to advanced, high-productivity products that one would not normally expect a poor, labor-abundant country like China to produce, let alone export”.

I like this kind of ‘trend’ chart, with income or some other indicator of development on the X axis, because it means we can pick out patterns that seem common to countries at a similar stage of development regardless of when they each reached that stage (a similar illustration for trade barriers would be particularly interesting). Of course, such patterns may break down over time - for example, it might be that the inverted U-shape trend was once the rule, but the diffusion of technology means that countries like China can start to seriously industrialise at much lower incomes. So this chart needs a Z axis to represent calendar time too (at which point it will unfortunately become pretty much impossible to either construct or comprehend).

I wasn’t going to comment on the recent flurry of free-marketeer blogger silliness over inequality, but now that Tim has kindly collected it all in one place the temptation is irresistable. Here’s what he says:

A highly amusing takedown of an all too common misconception over in the US blogs over the past couple of days … Brad Delong writes, “…a good chunk of the utility the rich derive from their conspicuous consumption is transferred to them from the poor”. That is, that it is not the having a lot, but having more, which makes the rich happier, exactly the same as Layard’s contention, that having less, whatever the absolute level, makes one unhappier.

All sorts of peoplethen pile in, Greg Mankiw, Arnold Kling, Don Boudreaux, and the ever delightful Jane Galt. As Arnold does with height, Jane uses the idea of beauty instead of wealth. It is indeed true that greater height or beauty provide greater opportunities in life … so why should these not also be equalised? … Where is the difference, in logical terms, between these and taking money away from the wealthy?

It is revealing that each of these free-marketeers has concluded that the problem with inequality is that the poor are envious of the rich, because there are other good reasons why inequality might bother the poor. Chris covers two of them very well:

One is that there are genuine consumption externalities. For example, if the rich drive hummers and big SUVs, they make roads more dangerous for the rest of us. Or perhaps their demand raises the prices of positional goods, such as houses in nice areas.

The other possibility is that “envy” is, in fact, a sense of injustice. Most people don’t much begrudge a lottery winner or top sportsman his fortune. Instead, what looks like “envy” is instead a discomfort that some people’s wealth is unjust. When people look at the wealth of David Lesar or Paris Hilton, is it really just envy they feel?

I’d add a third, which is the effect of income inequality on the distribution of power. The rich can use their money to buy influence, with a powerful incentive to do so when greater inequality in a society means greater divergence of interestst between them and the poor masses. The UK and US, the most unequal of the OECD countries, certainly provide no shortage of anecdotal evidence to back this hypothesis up, but is it supported by the data? Well, in Transparency International’s Corruption Perceptions Index there does seem to be a clustering of countries with low inequality, such as the Scandinavians, Singapore and Switzerland, among those who are least corrupt. More systematically, this research finds good support for the hypothesis in cross-country analysis.

The idea that the rich rig the system in their favour might explain the apparent relationship between higher inequality and lower social mobility, and low mobility in turn justifies the perception that the rich got where they are more by being born and raised rich than by talent and hard work. Again, this is about justice, not envy.

The effect of inequality on institutions is behind the conclusion arrived at by none other than Tim’s favourite economist William Easterly that “Inequality Does Cause Underdevelopment“. And not only does inequality lower growth, it makes what growth there is less pro-poor, as the 2006 World Development Report showed:

the impact of (the same amount of) growth on poverty reduction is significantly greater when initial income inequality is lower. On average, for countries with low levels of income inequality, a 1 percentage point growth in mean incomes leads to about a 4 percentage-point reduction in the incidence of $1 per day poverty. That power falls to close to zero in countries with high income inequality. Policies that lead to greater equity thus lead to lower poverty—directly through expanding the opportunities of the poor and indirectly through higher levels of sustained development.

So, it’s not just envy - the poor (and indeed the middle-classes) have good reasons to want less inequality. Do these apply to the two phenomena Tim et al seem to think are logically equivalent, namely beauty and height?

Generally, no. Inequality in beauty or height don’t appear to corrode our political institutions and slow growth, and we don’t tend to begrudge the beautiful or tall these characteristics because they are more or less randomly distributed within a particular population in each new birth cohort. The main thing upsetting the randomness is in fact income inequality, because plain rich people often get to mate with beautiful non-rich people, which might eventually make beauty more a preserve of the rich. Likewise, one of the main sources of jealousy against the pretty or athletically freakish comes when they manage to gain incredibly riches for no other reason - again, this wouldn’t be so much of a problem if there was less income inequality.

But do beautiful people make life harder for the rest of us mingers by depriving us of potential mates? I suppose we’ve all been there, but fortunately beauty appears to be no more common in either sex, which means there is an equally large proportion of mingers of the opposite sex with helpfully lowered expectations who are therefore more likely to want to ming with you. Plus, I don’t know about you, but I quite like having beautiful people around - it improves the view.

Like Adam Smithee, I think this paper from Francisco Rodríguez is pretty important. Basically, he says that most of the cross-country growth regressions beloved of so many economists are seriously and inherently flawed, but we don’t necessarily know how flawed or in which direction. The reason, as far as I understand it, seems to be the presence of “strong non-linearities in commonly used growth data sets”, in other words the problem that a given proportional change in a particular factor will have a different effect on growth depending on the level of that factor or of income.

Rodríguez finds that “the set of assumptions necessary to justify fitting a linear function to the data is so restrictive as to practically make the linear specification the true theoretical curiosum”, and decides that the most useful conclusion to draw from all those regressions is “a decisive rejection of the linearity process”, which was perhaps not the result hundreds of econometricians were hoping to hear. I don’t think that will stem the flow, though, since this kind of research has often been an exercise in taking a long time to say very little at all anyway.

Dani Rodrik is producing some very interesting stuff these days. Below the fold are extensive extracts from his new paper Industrial Development: Stylized Facts and Policies. Trust me, it’s more exciting than it sounds. The main message, for me, is the importance to development of increasing the ’sophistication’ of manufactured exports, and the importance to that in turn of learning, spillovers and technology transfer. There’s also some good theory on why the results of import liberalisation have been so disappointing in many cases. He contrasts the failure of governments like those in Latin America who, following orthodox advice, tried to just ‘get out of the way’ of businesses and watched their economies stagnate with the deliberately interventionist (though not always in the ways you might expect) approach adopted by the more succesful East Asian countries.

(more…)

From Bloomberg via the excellent Truck and Barter, here’s a description of some interesting new research from Alberto Alesina and William Easterly [and Janina Matuszeski]. It’s serious stuff but the central finding - that ’squiggliness’ is correlated with ‘bad stuff’ - can’t help but make me picture the pair as some sort of econometric Bill and Ted [and, er, Rufus?].

Correction: In my original post I left out the name of the paper’s third co-author, Janina Matuszeski. Apologies for that - especially because, as the grad student in a team with two professors, Janina probably did all the hard work.

“Authors Alberto Alesina and Janina Matuszeski of Harvard University and William Easterly at New York University divided countries into two categories: natural and artificial. A natural state is one defined by ethnicity and geographic features such as mountain ranges. Mountains reinforce ethnic communities — if only by isolating them. Natural national borders would tend to be bumpy.

The map of an artificial state by contrast looks like it was drawn with a ruler, which it often was. Its straight borders sometimes partition ethnic communities, placing them in two countries. Other times, they place tribes that are hostile to one another in the same nation.

Most nations have borders that are a combination of lines and bumps, so the authors developed a mathematical measure to quantify the extent of border bumpiness, which they called squiggliness. Since borders on oceans are extremely squiggly, the authors controlled for that and studied only the squiggliness of national borders with other nations. Their thesis is that it is better to be natural than artificial, and that squiggliness is good for growth and stability….

Less squiggly countries, the scholars found, generally have lower income, worse public services and higher infant mortality rates. They also found that social unrest, the sort that leads to wars, was also more frequent in unsquiggly places. The net finding, says Alesina, is that artificiality is “correlated with bad stuff.'’

It turns out that squiggliness matters even among countries ranking in the middle of the squiggliness scale. “When you move from the top quarter of squiggly countries to the bottom quarter you see a serious loss of gross domestic product,'’ Matuszeski says.

There are outliers, to be sure. At No. 11, Lebanon is super squiggly, which makes the current war there seem like an anomaly. The U.S. and Canada, as stable as they come, have long straight borders and low rankings. Here the situation is different, Matuszeski says, for “a key factor is when the border is drawn.'’ If it is drawn before settlers came — as was the case in the near-empty New World — then trouble is less likely…

There are other aspects of the study to challenge here, starting with the choice of the word “squiggly.'’ (It turns out the scholars thought about “wiggly,'’ but felt that “squiggly'’ worked better.)

The bigger problem with the study is the circularity of the argument. The great powers of a 100 or 50 years ago drew the lines that created the colonies or satellite countries.

Britain for example arbitrarily constructed Iraq, and arbitrarily decided its size, which is a bit less than twice that of the U.S. state of Idaho.

“The worst thing that ever happened to Iraq was the invention of the straight edge,'’ Easterly says. “They took Mesopotamia and combined mutually antagonistic groups in one nation.'’ Colonialism or tyranny sets trouble in motion. The lines themselves came later. …“The lesson of history is respect nationality,'’ Easterly says. “For Iraq, at the very least you want to emphasize the federalism established there and strengthen it.'’ He and his partners are looking at this in a new study, on wars and squiggliness.”