And wrong.

4.3. Assessing the Robustness of the Results The robustness of the results obtained was assessed using alternative models.11 The GDP growth variable (gdp_%_it) was replaced by GDP per capita based on constant prices (gdp_pcap_%_it). This helped to shift the research focus from a country’s economic strength to its level of development. This variable reflects changes in the well-being of the population. The analysis was based on the assumption that a lower or negative growth in real GDP per capita promotes a higher vote share for far-right or far-left parties (Brückner – Grüner 2010). The unemployment rate variable (unemp_dif_it) was replaced by employment rate variable (emp_rate_dif_it). Low employment (see Hungary and Portugal) suggests that a significant part of the productive-age population is not able or willing to work. This economically inactive labour force may also comprise people who gave up trying to find a job after a long period submitting futile applications (i.e. discouraged workers). The third economic indicator used in the analysis is the share of general government gross debt in GDP (debt_gdp_it). Some of the EU countries included in this research became so severely hit by the economic crisis that they were unable to maintain a sustainable fiscal policy and were forced to ask international institutions (EU, IMF, and the World Bank) for a bailout. These countries recorded significant growth in indebtedness. The general government gross debt for all the EU countries included in the research in 2012 reached an average of 71.8% of GDP. The assumption is that growing indebtedness increases the vote share of far-right or far-left parties, due to either expected, or already implemented restrictive measures. The control variables (controlit) in the estimated models remained unchanged. The results of the alternative models were similar to the results of the models found in Table 1 and Table 3 in the case of the far-right parties. The models revealed that a deterioration in economic variables (represented by growth in GDP per capita, employment rate, and general government gross debt per GDP) did not result in a higher vote share (or share of seats) of far-right parties. The explanatory variable which influenced far-right voters was social benefits. In the Tobit model for vote share as a dependent variable, the immigration growth variable (immig_%_it) was statistically significant at a level of 10%. This proved that far-right voters feared competition from an immigrant labour force. On the other hand, a 1% decline in the employment rate increased the vote share of far-left parties by 1.68 percentage points. The share of seats for far-left parties increased by 2.63 percentage points. The growth in GDP per capita variable (gdp_pcap_%_it) and general government gross debt variable (debt_gdp_it) were not statistically significant in any specifications of the model. Far-left voters were sensitive to changes in the openness of the economy. The statistical significance of the openness of economy variable (open_econ_dif_it) was at a regularly sustained level of 5%. When far-left parties participated in elections in West European countries they usually won a higher vote share and entered the lower house. The legacy of communism prevented similar success for far-left parties in the former the Eastern bloc. However, the number of seats which western farleft parties gained in lower houses did not differ from that of far-left parties in post-communist countries. Decline in voter turnout increased the share of seats of far-left parties in the period observed. The mean value of the electoral district (mean_district_it) has always been without statistical significance. The results obtained can be considered to be robust. The analysis proved that far-right parties’ electorates were influenced by factors other than the main macroeconomic indicators, while those voting for far-left parties were sensitive to the situation in the labour market, as well as the development of international trade. Neither growth in GDP per capita, nor general government gross debt seemed to exercise an influence on electoral decisions on both sides of the ideological spectrum. The empirical analysis showed that economic performance in the set of EU countries influenced the electoral success of far-right and far-left parties in different ways. In the case of the far-right electorate, economic development measured by means of basic macroeconomic indicators proved to be of no significance.