Long-run Economic Growth & Convergence

Extract the data for GDP per capita of all sovereign countries for selected years of 1960, 1990 and 2015 from global databases (like the IMF or World Bank) and perform a convergence data analysis based on the following questions.

use the programing language Gretl, accessible for free download at http://gretl.sourceforge.net .

3. Data Analysis

Using data of GDP per capita provided in the dataset, plot histograms of GDP for all the years between 1960 and 2015 and describe how the distribution of income evolves. If there are any outliers, omit them. In a case where you encounter any outlier, then provide economic intuition for evaluating these countries as outliers.

Calculate the Growth rate of GDP per capita for each of the countries in the years under consideration. Show boxplots of incomes for all the countries in each year. What happened? Is there any evidence for convergence? Discuss the notions of conditional and unconditional convergence in this regard (Hint.: Use logs instead of levels).

Split the data into two groups of poor and rich. The split should be done according to the median income in 1990. Then show boxplots of incomes for both groups in each year. What happened? Is there any evidence for convergence? (Hint.: Use logs instead of levels. Start with splitting the sample: poor in 1960, 1990 and 2015, then the same for the rich. Plot boxplots for each).

4. Extract the GDP, investment, education, political stability or other socio-political and economic factors for all the economies. Run the regression analysis both for the conditional and unconditional convergence for a sample of developed countries and developed countries, each separately and an aggregate analysis for the global economy. The selection for the development level categories can follow regional divisions or international organization membership, etc.... (For e.g. OECD, Western Economies, OPEC, BRICS, etc) . Plot the respective convergence analysis scatter plots depicting the convergence behaviour. (Use the average per capita GDP growth between 1990 and 2015 as a dependent variable and GDP growth rate in 1960 as the independent variable for the unconditional regression analysis. For the conditional convergence, use additional explanatory variables in addition to the GDP growth rate in 1960 ).

NB:

a) Unconditional convergence: (e.g. for the years between 1960 - 2000)

GDP2000-1980 = + GDP1960

lnGDP2000-1980 = + lnGDP1960

b) Conditional convergence (e.g. for the years between 1960 - 2000)

GDP2000-1980 = + GDP1960 +X1960

where X1960 is a set of country-specific controls (education, fiscal and monetary policy, competition level, etc.) - we compare countries with similar starting characteristics