Last week I reported on a study of global trade balances, using data published by the CIA in its World Fact Book, which showed a clear link between population density and the per capita balance of trade in manufactured goods. The following is the final chart from that article which shows the correlation:
Today we’ll look at this from a different angle. If population density is a factor in international trade, then what’s really important is not a nation’s population density relative to the rest of the world, but relative to those countries with which it actually engages in trade. No country trades equally with every other nation. Most countries’ trade is concentrated with just a handful of other nations. This is important because a nation with a high population density relative to the rest of the world might actually have a trade deficit in manufactured goods instead of a trade surplus if it trades mostly with other nations who are even more densely populated.
So I’ve used the CIA data for each nation’s major trading partners to calculate the difference (or disparity) in population density between each nation and a weighted average of its trading partners. For example, over half of all U.S. exports are sent to only six nations: Canada (20.1%), Mexico (11.7%), China (5.5%), Japan (5.1%), Germany (4.2%) and the U.K. (4.1%). And well over half of all of U.S. imports come from these nations too. In the case of the U.S., the difference (or disparity) in population density is -79.8 people/sq km. That is, while the U.S. has a population density of 31.26 people per square kilometer, the weighted average population density of its major trading partners is 111.06. So the density of the U.S. is less than that of its trading partners by 79.8.
My expectation was that this approach would reveal an even stronger correlation between population density and per capita trade balance in manufactured goods. But this approach is not without problems. First of all, the CIA data on trade partners includes all trade – not just manufactured goods. Also, the CIA only provides data for each nation’s top trade partners, accounting for 50-75% of trade for each nation. Also, it should be remembered that free trade is not universally practiced around the world. The World Trade Organization (WTO) actually enforces protectionist trade practices in favor of two thirds of its member states for the purpose of helping undeveloped and developing nations. (The United States is not one of them.) Such uneven trade practices will tend to obscure the effects of other parameters like population density.
So what did I find? Here’s a scatter chart of the results for all nations with per capita purchasing power parity greater than $25,000 per year (in other words, for all developed nations).
Once again, we see that Qatar is a statistical outlier. Also, Ireland is nearly one as well. If we remove these two data points, the relationship between population density and per capita balance of trade in manufactured goods becomes quite strong.
Here we can see that the data points conform more closely to the trend line, indicating a stronger relationship. A couple of other interesting observations can be made:
- Note that the trend line, which was generated by the computer and not drawn in by me, intersects exactly the zero point of both the x and y axes. This means that the data indicates that trade between nations equal in population density will, in all likelihood, yield a perfect balance of trade in manufactured goods.
- Nations that are more densely populated than their trading partners can, on average, expect a trade surplus in manufactured goods of about $700 per person for each increment of 100 people/sq km by which their population density exceeds that of its trading partners. The numbers are reversed for nations less densely populated than their trading partners by that amount. Returning to the U.S. as an example, our population density disparity of -79.8 would predict a per capita trade deficit of about $560 per capita. The actual figure is $1,344 per capita. This deviation from the prediction is likely due to U.S. trade with China, who benefits from WTO-enforced tariffs.
So what does all of this mean for trade policy in general? Clearly, if you’re a very densely populated nation, it’s in your best interest to advocate free trade with nations less densely populated, which would be nearly everyone. Such a policy will almost guarantee a surplus of trade. (Note that, of the ten developed nations that are more densely populated than their trading partners, only one has a trade deficit in manufactured goods.)
On the other hand, for more sparsely populated nations, while free trade with other sparsely populated nations is just fine, free trade with more densely populated nations is almost a sure-fire loser. (Only seven of nineteen nations in this situation have a trade surplus.) If such nations are interested in maintaining a balance of trade, some mechanism is needed to to counter-act the population density effect: either import quotas or tariffs.
The effect of disparities in population density between a nation and its trading partners is real and significant. For nations engaged in trade with others more densely populated, some mechanism to counteract this effect must be factored into their trade policies if trade deficits are to be avoided.