The Gravity Model of Trade (see Feenstra, page 192-196) predicts that trade increases with the partner country’s GDP: the bigger your partner country’s GDP, the more you trade with that country. The model also predicts that trade decreases with trade costs, which can be proxied by the distance to the partner country, along with other variables. Specifically, the further away your partner, the less you trade with that country. In this assignment you will test these hypotheses.
Step 1: Collect your data. You will need data on trade flows, GDP, and distance between your country and its 20 largest trading partners. Note that this data is helpfully collected for all trading partners for the years 1948-2019 at CEPII (http://www.cepii.fr/cepii/en/bdd_modele/presentation.asp?id=8 However, the dataset too large to open in excel, thus I would only recommend it for students more comfortable with other statistical tools. All the documentation for the dataset can be found at the CEPII link above. You can also collect your own data from the following sources:
- Trade Flows. Use the United Nations Comtrade database to download total imports and exports between your country and all of its trading partners in 2019. Once you have this data, calculate total trade as the sum of exports plus imports associated with each partner. Note that this data is in nominal U.S. dollars (See helpful comtrade hints).
- Nominal GDP Data. Use the World Bank’s World Development Indicators (https://databank.worldbank.org/source/world-development-indicators ) to download nominal GDP in US dollars for each partner country in your dataset.
- Distance data can be found from CEPII at the website:
http://www.cepii.fr/anglaisgraph/bdd/distances.htm . The file dist_cepii.xls is a spreadsheet file in Microsoft Excel format which contains distances between countries in kilometers.
The country codes are three-letter ISO codes. You can find a list of ISO codes here: https://www.iso.org/obp/ui/#search
Step 2: Choose a sample country to Visualize your data. Make a scatter plot of your data that allows you to visualize the size of the country (GDP) as a determinant of trade. It should have GDP (as a percentage of the GDP of all the partner countries in your dataset) on the horizontal axis and total trade (as a percentage of total trade of all the partner countries in your dataset) on the vertical axis.
Step 3: Estimate the coefficients of the gravity equation: (see attached file)
You should estimate coefficients b and c using multiple regression analysis, which can be done in a spreadsheet package or in a statistical software package. The following website has a nice summary of doing regression analysis in excel for those of you who aren’t familiar with this tool:
Write your report. Start with a general discussion of the gravity model, how you will test it, and what data you are using. Carefully report the results from your scatter plot and interpret the coefficients you estimate. Include your scatterplot and a Table of your coefficient estimates. Conclude your report by analyzing whether the gravity model seems to fit the trade data from your country. You should also turn in the statistical package output or spreadsheet that you used to analyze your results.