19
each country, and because it portrays voters as sophisticated individuals with the capacity
to compare past economic performance with present economic performance. It assumes
that voters posses long term vision. These variables I will call Non Myopic Growth, and
Non Myopic Inflation.
14
The first variable measures the difference between the mean
growth rate of GDP per capita observed since the moment that the party took power until
new elections and the mean growth rate of the decade preceding that government (i.e. and
historical mean)
15
. As one can see, this is a long term variable because it compares the
economic performance of the current government with an historical growth rate. This
measure also controls for the specific economic situation of each country.
Non Myopic
Inflation is measured in exactly the same way as Non Myopic Growth. The second set of
economic variables is: Myopic Growth and Myopic Inflation. These variables measure the
average of the GDP per capita growth and inflation in the two years preceding the
election. These variables are not as myopic as the standard variables used in the PALW
model that just consider the inflation and growth rates of the quarter when the election
took place; however, compared to my long term measures, they can still be taken as short
term measures.
16
14
Although I also constructed a Non myopic version of unemployment, I will not use it because it contains
a lot of missing data, due to the fact that there was little unemployment data for the sixties and seventies. I
have very few observations, thus I will use the “myopic” quarterly unemployment measure of Palmer and
Whitten.
15
For example, if we were to study the non -myopic GDP per capita growth effect on the 2000 US election,
we would measure the difference between the average mean rate of GDP per capita growth from 1997 to
2000, and the average mean rate of GDP per capita growth from 1986 through 1995.
16
I will only use these myopic measures in the models that I will run with the expanded data set described
in the next section, because I do not have the quarterly inflation rates for all the cases in this data set.
Convention