- 10 -
well informed on political issues and to go to the ballots to vote sincerely. The disappointment of this
expectation leads to a loss of prestige. These private costs of being politically uninformed arise with
respect to single political decisions, but not with respect to political parties. Private costs of political
participation in direct democracy increase the individual incentives to vote according to their own in-
terest, such that the extent of expressive motivation is lower in direct than in representative democ-
racy.
Second, the hypothesis that representatives have higher civic virtue than those whom they represent
can be fairly well challenged on the basis of experience with most democratically elected govern-
ments and assemblies in OECD countries. Brennan and Hamlin (2000, p. 178) may be right in as-
serting that the relevant virtues need not coincide with prejudices. It is perhaps more important that
representatives behave virtuously in the sense of not exploiting citizens, instead of subscribing to a
certain morally informed sexual behaviour. However, the cases of corruption in France, Germany,
the U.K., Italy and so on in the last two decades as compared to Switzerland and U.S. states illus-
trate that representative democracies are not able to select more virtuous representatives; the amount
of corruption does not seem to depend on whether the citizens have direct political rights or not.
Somehow the authors seem to hang onto the old dream of selecting statesmen instead of politicians.
German speaking economists often bemoan that a statesman like Ludwig Erhard, the first German
Minister of Economics after the War, has not been in sight for years in Germany.12 That voters might
fail to select more virtuous representatives may result from the false implicit assumption that they can
more easily judge people than policies. People may more easily appeal to individual emotions than
policies ever can. If for example beauty matters as much for the election of representatives as it does
for hiring in the labour market, beautiful representatives have ceteris paribus less restrictions to ex-
tract political rents.
Third, the authors exaggerate their expressive voting argument. It is not fully convincing that citizens
express their opinion at the ballots in the same fashion as cheering at a football match. Mueller
(2003, p. 321) notes that it is at least equally likely that citizens use the opportunity to express more
noble sentiments. “Norms that govern conduct toward others might be expected to be particularly
likely to come into play when individuals vote.” (Mueller, 2003, p. 321). Kirchgassner (1992, 1996)
argues that voting is a low cost decision by which citizens do neither affect their own instrumental
well-being nor the well-being of other individuals but the collective outcome. In such a low cost deci-
sion it is cheaper to act along moral sentiments than in market decisions where moral behaviour is
potentially punished by utility maximising individuals. The theory of low cost decisions hence provides
a competing argument why people vote and additional arguments why voting is guided by moral
codes. The possibility that the discourse preceding the referendum decision at the ballots may lead to
a revision of preferences over policies plays a special role in that respect. Citizens may act more al-
truistically in referenda and initiatives than in economic decisions. For example, Pommerehne and
Schneider (1985) presented empirical evidence that Swiss people vote for redistributional pro-
grammes despite the fact that these result in a loss in their own wealth. In decisions at the ballots,
voters therefore may follow an ethical position to a larger extent than could be expected on the basis
of the preferences that guide their behaviour in the market place. This effect can be traced back to
12. See, e.g., Vaubel (1988).