Warning: A bit wonky
Mercantilism and protectionism have come back into vogue lately, not just in the US but across the world. Although there is all kinds of literature out there on trade, I thought I’d add just a little of my own. However, a disclaimer is necessary: the results I will be discussing should be taken with a grain of salt. This analysis has not passed though any other form of review other than myself (which is why this is a blog post as opposed to academic paper). I’ll leave you, the reader, to judge their value for yourself.
With that said, let’s get started.
Method and Data Sources:
The goal of this analysis is to attempt to answer the question “Does international trade help or harm an economy?” President Trump, as well as many others, have claimed international trade harms the US economy. Economists maintain the opposite view. In order to test this claim, I have taken a simple approach. I have run a regression on GDP per Capita (logged) on the Fraser Institutions Freedom to Trade Internationally score, holding “institutions” constant (more on this in a moment). The standard regression model was applied:
LogGDP=α+βFreedom to Trade Internationally+γInstitutions+ε
LogGDP = The natural log of GDP per Capita by country for 2014 (World Bank)
Freedom to Trade Internationally = A ranking put out by the Fraser Institute measuring a country’s tariff level, trade barriers, and capital & labor controls.
Institutions = An author-developed category using the Fraser Institute Legal Systems and Property Rights ranking. This category includes measurements on a country’s: Judicial independence, impartiality of courts, strength of property rights, military interference in politics or rule of law, integrity of legal system, legal enforcement of contracts, regulatory costs of the sale of property, reliability of police, and business cost of crime.
Of the 155 observations in the data, they were broken up into 4 categories based upon their Institutions score. This was done in order to hold institutions constant across the nations and try to prevent any negative/positive feedback from institutions that could influence our outcome.
First, some summary statistics for each category:
|GDP Per Capita; Note: this data is not logged|
And the results of the regression:
Model: LogGDP = 2.26+.327Intl+1.28Inst+ε
At this particular point, I want to avoid talking too much about the hard numbers. There are likely a number of issues with my model (not the least of which is only one observation in Group 1) and once this gets cleaned up, I suspect the actual coefficients would change. Rather, I want to focus on the sign of the coefficients. We see a positive coefficient for the Freedom to Trade Internationally variable. This indicates that, as a nation’s freedom to trade internationally rises, so does the log GDP per capita. This is contrary to the claims of President Trump and his cabal of political speakers. What’s more, this effect is fairly consistent; the R-squared term of 0.55 tells us this is a decently strong positive correlation.
The next question I’d like to address is a claim some other have made, along the lines of “trade is fine so long as we’re trading only with people with similar rules and cultures as us. Currency manipulation, export subsidies, and the like actually hurt America!” This is a far more interesting question than the above, but it will also be much harder to design an experiment and test (if any of you have any ideas, please feel free to share!).
In conclusion, this has been a quick and dirty regression on the topic of international trade. If you’ve any questions, comments, suggestions, I’d love to hear them.
In case any of you are the visual type, here is a graph of the data