Brad DeLong's Semi-Daily Journal (2004)

These were the authors of the study on the spread of the gold standard and the first wave of globalization?

The last bit about the absence of a link between petroleum exports and democracy comes as no surprise to me. What reason is there to expect greater accountability from any government that has easy access to non-tax-based revenues, thanks to oil rents extracted via state-owned parastatals?

What is *truly* interesting is that the link also fails to hold for commodity exporters. I wonder if the nature of the commodity being exported might have something to do with it?

The bit about the commodity exportation not leading to democracy in general certainly seems to have been true for Latin America for most of history up until the last 20 years. Their economies were generally based upon production of a few agricultural products, government handouts to the producers and high trade barriers. Interesting to think that this might have gone hand-in-hand with the general failure to develop democratic institutions during this period of time. Only post WWII when worldwide political stabilization and spread of organized agriculture led to competition and a decrease in commodity prices, making monoculture inviable as an economic system were these countries forced to open trade barriers, and concomitantly develop some real democratic institutions.

Pretty neat idea, but the causality of it all escapes me. What comes first, international trade or democratic governance, and why? What's the link?

They write:

" The likely endogeneity between democracy and trade is addressed via the gravity model of trade, allowing us to obtain a measure of natural openness. This serves as our instrumental variable for actual trade openness a la Frankel and Romer (1989). We use this powerful instrument to obtain consistent estimates of the causal impact . . .

Scientism, pure and simple.

George Soros notes that in Africa there is a negative correlation between resource wealth and open society. His view is that natural resources create both the means and the ends for financing civil war.

The last part comes as no surprise to me either.

"Pretty neat idea, but the causality of it all escapes me. What comes first, international trade or democratic governance, and why? What's the link?"

The authors used a 'measure of natural openness' to trade as an instrumental variable, rather than actual trade volume, precisely in order to tease out the direction of causality. I don't know the details of this study, but a gravity model typically postulates that the amount of trade depends on the distance from and size of potential trading partners.

The idea is to find a measure that is correlated with actual trade volumes, but which, unlike actual trade volumes, cannot be affected by the presense or absence of democracy in any plausible way. Then, if a relationship is found between this measure and democracy, the causality must run the other way. I don't know how convincing the data and argument are in this case, but it sounds interesting.

So: for nations that lapsed away from democracy, what proportion of the totalitarian or authoritarian movements that took power used protectionism in their early efforts to gain traction?

Frank-

Offhand I can't think of very many examples of countries which truly lapsed away from democracy. Perhaps the only one which immediately comes to mind would be the Weimar republic in post WWI Germany, which was certainly very short lived and never very stable in the face of extreme economic instability during the period of time 1920s-1933 in which it governed. Have you any other examples?

Luke-

"Scientism" is too kind.

Sam

"Offhand I can't think of very many examples of countries which truly lapsed away from democracy."

I can think of maybe a few. Argentina, Uruguay, Cuba (?), Czechoslovakia (1948), Venezuela (?). You're right though, these countries are pretty few and far between. And I doubt that some of these places (or Weimar Germany) were truly democratic in the first place.

Sam and Luke: What do you propose instead of light modelling and econometrics? Hand-waving only gets people so far in terms of understanding the world. I generally believe that it's better to have bad information than none at all. So long as we take these and any results with a grain of salt, we should encourage such work. It sure beats n-commodity general equilibrium, which I despise precisely because of its scientism.

Sam and Luke -
If I may paraphrase Churchill, scientism is the worst way of trying to understand the world except for all the others that have been tried (re that, note my nom-de-plume).

If econometrics doesn't do it for you, just take an informal look at history in the long view. Large trading sectors have always, since Mesopotamian times at least, been associated with peace, prosperity, learning and civil society. Autarchy, on the contrary, has always been associated with militarism, poverty, ignorance and tyranny.

Who gave more to the world - free trading Athens or autarkic Sparta?

By "as democratic as" do Msrs. Lopez-Cordova and Meissner mean the US bipartheid political system aka our Corporate Dumb & Dumber morality play?
Rule by millionaires, for billionaires? With the government and corporate elites controlling 96% of the National income, you and me getting our rewards by-n-by, in that great WTO in the sky?

Take a look at commodities. The falling US$ is raising the price of imports, but US producers are just raising their prices to sop it up. The little guy, the average joe working 2-3 jobs and making $21,450 a year (yup, that's the AVERAGE US wage earner), is getting *clothes-lined*.

Can ya feel me, playah?!

Interesting, but doesn't sound very convincing. You could also correlate democracy with the use of cars and obtain a pretty good result. Just because it happens at the same time doesn't mean there is a causal relation. Of course the study tries to demonstrate the causal relation, but the fact that countries with commodity based economies didn't followed the model suggests there is something else at work here. ¿Maybe what the country produces is as important as trade openness? Or maybe the degree of trade openness and diversified production simply mean developed economies, in which case the study says something a bit obvious.
In the case of Uruguay, by the way,the "democracy lapses" are associated more with economic crisis than with trade openness. The first in the XX century was in 1931 (the Depression) and the second was in 1971 after "worldwide political stabilization and spread of organized agriculture led to competition and a decrease in commodity prices, making monoculture inviable as an economic system". The anomaly of commodity dependent countries is easy to explain. In these economies land is the most valuable resource and it tends to be highly concentrated which in turns produces an extremely powerful economic elite and little social mobility. Not a great recipe for democracy. The relative difference of democratic Uruguay in early XX century in Latin America was caused by the political weakness of the landowners which allowed a urban "social democratic" government to build a modern, more equalitarian country on the basis of the taxes paid by the rich landowners.
The study seems a bit ideology driven, if you ask me. It's pretty unlikely that something as complex as political institutions can be caused simply by trade policy. It may be a factor, yes, but one of many.

Keynesian Theory is like Darwin's Theory, an ancient side-saddle, well worn, gathering dust.

The only Gravity Model in play these days is an exponentially increasing one, wherein, once global trade ceases to become bilaterally regulated (and therefore self-limiting) and instead becomes a spiral of mutually aligned free-trade players, one providing currency or capital, one providing natural resources, and one or several providing finished manufactured goods and steel bottoms, and if the self-regulating effects of tariffs and currency exchanges are removed by WTO and price-pegging, then the model for trade imbalance become asymptotically vertical ... the natural world energy model is the Forest Fire or the Typhoon (or the Black Hole of Steve Hawkins.)

Whose economy will be completely immolated first?

Read
Tschirhart's "Ecological Transfers in Non-Human Communities - Parallel Economic Markets in a General Equilibrium Ecosystem Model"

"Nash Equilibrium and Economic Theory", http://home.uchicago.edu/~rmyerson/research/jelnash.pdf

and McCain's, "Zero Sum Games", http://william-king.www.drexel.edu/top/eco/game/game.html

Then move slowly towards the exits....

"Offhand I can't think of very many examples of countries which truly lapsed away from democracy."

"I can think of maybe a few. Argentina, Uruguay, Cuba (?), Czechoslovakia (1948), Venezuela (?). You're right though, these countries are pretty few and far between."

Let me add a few to that list where since 1950, democratically elected governements have been displaced for significant periods of time:

Chile, Brazil, Peru, Bolivia, El Salvador, Dominican Republic, Nigeria, Iran, Pakistan, Burma, Phillipines, Turkey, Greece and increasingly Russia. Of course, in the US we had an interesing judicial coup.

Some, it is true have come back and we seem to be in a good period, but that is no guarantee of anything.

Justin Duckie:

"..like Darwin's Theory, an ancient side-saddle, well worn, gathering dust."

With this bit of idiocy, you have just thrown away any right to be taken seriously.

Lighten up Grep!

"... like an ancient side-saddle" was an interesting, almost poetic, bit of fun.

I don't know what it means, but it provides relief from the relentless, almost mindless political chatter polluting this site.

Even more amusing is his inability to see any stabilizing feedback mechanisms in the global economy. Nothing guaranteed, of course. Certainly not happy futures for the USA.

Doomster predictions of the Global Economic Forest Fire are always fun to read. I have a collection of such books in my office.

Let me add a few to that list where since 1950, democratically elected governements have been displaced for significant periods of time:

Chile, Brazil, Peru, Bolivia, El Salvador, Dominican Republic, Nigeria, Iran, Pakistan, Burma, Phillipines, Turkey, Greece and increasingly Russia. Of course, in the US we had an interesing judicial coup.

Posted by: Eli Rabett on March 15, 2004 08:59 PM

First of all as a former resident of South America I would dispute all of the above Latin countries as having had stable democratic governments which then went to autocratic regimes spontaneously. There is a big difference between calling a government 'democratic' and actually having democratic institutions in place that function. While there were some fits and starts, notable Chile et al., as it turned out throughout latin american history a constant factor until 1980 was that whenever a popularly based government would spring up promising a hint of self determination, it would be immediately (within a few years) crushed by the old dominant elitist hierarchy. With or without the CIA being involved, this was the standard old story of all of the famous latin 'republics'- crushed by local caudillos.

Things have been different since then, though, and remarkably so. Not in all of latin america, but certainly Venezuela, Chile, Uruguay, Argentina, and Brazil. Even in the case of Venezuela, recently, with Hugo Chavez, the caudillo's nightmare, could not be dethroned because the local elites basically decided they would rather stick out a Chavez than go back down the military dictatorship route, even with the meddling CIA. The same is true in Brazil, where a populist, former socialist president has been fully accepted by the whole society- noone wants to start with the autocratic crap again- they have had enough.

So, is it free trade, is it experience, is it that even the elites feel poor now compared to the Gates' et al. of the wealthier world? Hard to say, but my sense is, is that things really have changed, and there is no going back. Good for them, good for us, good for the world!

Leaving aside what "democracy" means, perhaps more realistically it would correlate with the higher degree of functional differentiation in an economy, since this would seem to generate the functional need for a more complex institutional means of coordinating different interests. Openness to trade would then be a function of the functional differentiation in the economy in the first place.

Add to the list of relapsing countries Italy in the 1920's.

And Japan and Spain in the 1930's.

Ever reliable John H. The debate over what constitutes democracy may not go away when we begin discussing national institutions, but it certainly becomes more precise. I'm sure the authors employ a definition for democracy that allows them to sort countries somehow, though it is not mentioned here. Very likely, we would disagree over some of the inclusions. When we get down to institutions - courts, laws, civil service, schools and the like - then the link between democracy and trade may be clearer. We might also speculate that causation runs not from trade to democracy, but from proper institutions to both trade and democracy. Or that there may be some virtuous interaction between the lot of them.

It does seem a bit naive to write, straight-faced, that "a one standard deviation increase in trade with other countries could bring countries like Indonesia, Russia or Venezuela to be as democratic as" the US, UK or France. What we seem to have is members of the free trade religion rubbing a couple of data series together and then claim that trade is the democratic philosophers' stone. Is the world really that simple?

Let me agree with all of the above, especially my colleague the non economist. The problem with doing so however, is that before about 1950 you could count the number of stable democratic countries on your hands (maybe including toes), and even today you would only have to include the digital appendages of a friend or maybe two if you were a real optimist. Progress to be sure.

The authors claim that "one standard deviation increase in trade with other countries could bring countries like Indonesia, Russia, or Venezuela to be as democratic as the U.S., Great Britain, or France" doesn't ring true. It implies a direct causality between trade and democracy.

This doesn't take into account other driving and enabling factors in a country's path to democracy.

Singapore? Not mentioned so far. Democratic?

Mainland China?

I agree with the "scientism" criticism. Has the powerful gravity model of trade contributed anything to our understanding which we didn't already had -- i.e. the anecdotal evidence adduced by d. derider? In particular, anything to justify the causal speculations about the effect of one standard deviation change?

Especially because you have to throw out natural-resource trade to begin with. And probably fine-tune your idea of democracy, and history of democracies, to fit your theory, too.

And then people pitched in with talk about everything else being handwaving and useless. Seemingly those of us outside economics / econometrics are going to have to sit on the ashheap of history and take things on trust from here on out.

But I'm not impressed with the ability of economists either to be right all the time, to take everything relevant into consideration, to be adequately self-critical, or to be completely frank about doubts within the profession. That's not especially a critique of economists, just the normal way to feel about anytone who claims transcendant wisdom.

Well, we certainly are talking about an endogeneity problem regardless of what we think about causation. That's the point of the whole two-stage least squares procedure. Whether or not this is successful is another story.

I think that the authors overstep a bit when talking about causation; openness to trade is no doubt a proxy for other good institutions like economic freedom. Also for GDP (which is a problem). Again these are correlated with democracy but they aren't the same thing; think of illiberal democracies like India squashing growth or liberal nondemocratic regimes like the late British rule in Hong Kong encouraging it.

So we have increases in GDP apparently being associated with better institutions. Tell us something we don't know. Econometrics is a useful thing (as I mention above) but this is just not a terribly interesting or well-done result. We have to include some sort of income variable in our system in order to see whether income causes trade and democracy, or whether trade causes democracy. Even then we have to be careful. Autocorrelation, anyone?

Sorry, I'm picky about this sort of thing. Cross-country econometrics is very hard precisely because of all this endogeneity. But so long as we're skeptical enough about the results it's better than hand-waving. Or we can listen to people with compound-complex sentences about some sort of impending doom, modern day street-corner prophets. The end unfortunately never seems to come before my credit-card bill is due.

Is anyone else reminded of the movie, The Gods Must Be Crazy, when reading this thread? Folks, it's only an empty coke bottle.

I wonder if some of what has set of the objections here (mine among 'em) is an incautious bit of writing, rather than bad thinking. Surely, these aren't the first bunch to think that democracy and trade have a number of overlaps and feedbacks. But to tell us that the gravity model is a "powerful instrument" runs directly afoul of Samuel Clements urging - "When you catch an adjective, kill it." There guys are selling a little too hard. The importance in this "powerful instrument" is the consistency of the estimates it produces, by the authors own account. Consistency is important, yes, especially for the researcher to have confidence in the results. But then these guys go on to assert that they have nailed down causality. Daniel Lam makes the point that the authors have made appropriate scholarly efforts to make a case for causality, but there has to be more to causality than simply getting one's instrumental variables right. So this advertising blurb has not won us over to the product. There's no link, so can anyone (Brad) tell us if there is more to the article than what is advertised in the blurb?

Are those the two forms of discourse about human history, economics and hand-waving? The arrogance of economists at times does seem boundless.

There's democracy and then there's democracy. Among the advanced countries mentioned, France might qualify, but certainly not the U.S. or Britain.

I hope the article is better than the abstract.

"The *likely* endogeneity between democracy and trade is addressed via the gravity model of trade, allowing us to obtain a measure of natural openness."

I'm glad he found what he thought was there, now explain this to the Chinese.

First of all, there's trade and then there's trade. It makes a great deal of differnce whose terms it's on. When a "sole remaining superpower" intervenes constantly in the global market to make sure trade takes place only on terms favorable to TNCs, and Third World countries are as a result turned into giant authoritarian sweatshops, it's hard to see how "democracy" can come out of that.

For that matter, there's democracy and then there's democracy. What the chattering classes mean by the term is not direct, participatory, and decentralized control over the decisions that affect working people's lives. They usually mean, rather, what the neocons call "rule of law": spectator democracy, with the "professionalization" of policy and a heapin' helpin' of Weberian bureaucratic rationality. In other words, a centralized national government with two parties, a half-inch to the right and left of center, respectively, run by men in suits who take turns awaiting orders from the IMF.