Lecture 14

5 0 5 prt lso grc esp hun bra pan mwi hti pry npl 1 0

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ndex -.5 0 .5 PRT LSO GRC ESP HUN BRA PAN MWI HTI PRY NPL -1 0 1 Change in Log GDP per Capita (Penn World Tables) 2 Figure 4 Democracy Growth and Income Growth 1870-1995 Change in Polity Composite Index .2 .4 .6 .8 PRT HUN URY BRA NOR ITA AUT SWE NLD DNK VEN ARG MEX JPN GBR CAN BEL USA 0 GRC CHE CHN 1 1.5 2 2.5 3 Change in Log GDP per Capita (Maddison) 3.5 The Statistical Model Panel data regressions with …xed e¤ects show no evidence that income per-capita is positively correlated with democracy no matter how you look at the data. Our basic regression model which we use for the post-war and for the 100 year sample is: dit = αdit 1 + γyit 1 0 + xit 1β + µt + δi + uit , (1) where dit is the democracy score of country i in period t . The main variable of interest is yit 1 , the lagged value of log income per capita. The parameter γ therefore measures whether income has an e¤ect on democracy. All other potential covariates are included in the vector xit 1 . In addition, the δi ’ denote a full set of country dummies and s the µt ’ denote a full set of time e¤ects, which capture common s shocks to (common trends in) the democracy score of all coun...
View Full Document

Ask a homework question - tutors are online