The effects are quite small because we estimated the short-run effects, we expect them
to be larger in the steady state.
Finally, how does our model fare in terms of forecasting and reliability? As might be
expected from such a rare event, it puts the probability of Estonia’s adopting the flat tax
in 1994 at .01. Russia’s probability of flat tax adoption in 2001 is .39; Albania’s for 2005
is also .39, Romania’s is .17, and Georgia’s is .99. The erratic nature of these predictions
indicate that there is still much to be done in terms of model specification. That said, we
might mention that the predicted probability for all countries in our sample of adopting
the flat tax in 2005 is 75 percent. Overall, the model predicts 72.8 percent of observations
correctly: if Pr > 0.5 and y is 1, and if Pr < 0.5 and y = 0.
5 Conclusion
We hope to have demonstrated that, when it comes to extraordinary politics, economic
forces - either domestic or international - do not seem to be sufficient to get reform off
the ground. In the presence of the right ideological environment, rational policymakers
learn from other countries’ successes with the flat tax and move to adopt the reform
themselves. Both political constraints and peer pressure from neighboring countries are
a powerful influence.
If diffusion is a strong force, though, it provokes an interesting policy question: though
we have shown diffusion to matter intraregionally, could the flat tax revolution jump over
to Western Europe as well? Eastern European countries for the most part kicked off the
1990s with ambitious reform agendas to reshape and open their economies. Policy reform
in the West may be more difficult. But competition from the East may prove too strong
a force. As Estonian Prime Minister Laar put it, “if ‘old Europe’ is to compete effectively
27