open Secondary menu

Why is Turnout Higher in Some Countries than in Others?


4. Turnout and Electoral Administration

Tables 3A and 3B show the correlation with electoral administration variables once the effects of socio-economic and macro-institutional variables have been neutralized. It turns out, first, that the variables related to registration (whether obligatory or not; whether it is possible to register the day of the election; and whether the government is responsible for taking the initiative to register electors6) have hardly any effect on voter turnout. The ability to register on the day of the election even seems to have the perverse effect of lowering turnout as calculated on the basis of the population of voting age (Table 3B). But here, too, we think that the more credible data are those on turnout as a function of the number of registered electors, and these data show a correlation that is not significant (Table 3A).

4.1 Lists of Electors

It is surprising that the procedure for registering on the lists of electors does not seem to have a clear effect on voter turnout. Traditionally, the low turnout in the United States has been attributed to the difficulty of registering on the lists of electors (see in particular Wolfinger and Rosenstone, 1980; Powell, 1986). The various measures taken in the U.S. to facilitate registration have had, in fact, a modest but real effect on voter turnout (Knack, 1995; Brians and Grofman, 2001). It should be pointed out that the American studies are based on more detailed information than we had on the way in which the assorted measures were applied in various U.S. states. It is, perhaps, because our data were not sufficiently precise that we did not observe any significant effect. However, we should also point out that the registration procedures in the United States are so peculiar to that country that the applicability of American studies is necessarily limited.

4.2 Polling Day

We also wanted to check the hypothesis that turnout can be increased by holding the vote on a holiday. Franklin (1996) presented results that seemed to confirm this hypothesis, but his study, in contrast to ours, did not have any controls for socio-economic and geographic factors. Our own data indicate that this does not seem to be a really significant factor.

4.3 Ease of Voting

Finally, we created an "ease of voting" variable that indicates whether it is possible to vote by mail, in advance or by proxy.7 This variable has a positive, significant coefficient when turnout is calculated on the basis of registered voters (Table 3A). Our results suggest that turnout is indeed higher when the electoral legislation facilitates the exercise of the right to vote. More specifically, all other things being equal, turnout is about 10 percent higher in countries where it is possible to vote by mail, in advance or by proxy, than in countries where none of these options are available.

It has been suggested that the length of the election period can have an effect on voter turnout. Electors may be bored by campaigns that last too long and therefore go to vote in smaller numbers. On the other hand, campaigns that are too short may not be able to arouse voter interest. So far as we know, these hypotheses have never been tested in a systematic fashion, and we do not have complete data that would enable us to do so.


6 The government is deemed here to take the initiative in the registration process when a public agency takes the initial steps to register people who have just gained the right to vote, and not the people themselves. In Canada, the enumeration procedure (until 1997) and the register of electors procedure (since 1997) both fall into the category of registration systems in which the government takes the initiative. Since enumeration is not very common in the rest of the world, it is difficult to distinguish its specific effects from those of other methods for which the government takes the initiative.

7 The variable assumes the value 0 when none of the options is available, and 0.33 when only one option is available under the electoral legislation. The variable assumes the value 0.66 when two options are available and 1 when all three are available.


Table 3A – Determinants of Voter Turnout:

Socio-economic, Geographic and Macro- and Micro-institutional Factors
Independent Variables Dependent Variable: Turnout by Number of People Registered on the Lists of Electors

 Regression Coefficient

 (Error)

Eastern Europe
0.18
(3.85)
Population (log)
-3.34***
(1.22)
GDP per capita (log)
8.28***
(2.93)
Compulsory voting with penalties
13.28***
(2.80)
PR and mixed compensatory
4.20*
(2.19)
Compulsory registration
3.79
(2.45)
Registration on polling day
-4.80
(3.07)
Government initiative
0.85
(2.43)
Holiday
-0.26
(2.72)
Easy to vote
11.04**  
(4.35)
Mali
-33.42***
(11.14)
Constant
55.54***
(13.74)

Number of cases: 119
Adjusted R2: 0.46
*     significant at 0.10 (two-tailed test)
**   significant at 0.05 (two-tailed test)
*** significant at 0.01 (two-tailed test)



Table 3B – Determinants of Voter Turnout:

Socio-economic, Geographic and Macro- and Micro-institutional Factors
Independent Variables Dependent Variable: Turnout by Population of Voting Age

 Regression Coefficient

 (Error)

Eastern Europe
4.16
(4.71)
Population (log)
-3.12**  
(1.56)
GDP per capita (log)
8.93**  
(3.52)
Compulsory voting with penalties
5.59
(3.43)
PR and mixed compensatory
7.84***
(2.74)
Compulsory registration
4.37
(3.16)
Registration on polling day
-9.18**
(3.82)
Government initiative
-1.24
(3.06)
Holiday
-0.97
(3.39)
Easy to vote
8.10
(5.45)
Mali
-23.51*
(12.03)
Constant
51.27***
(16.73)

Number of cases: 89
Adjusted R2: 0.40
*     significant at 0.10 (two-tailed test)
**   significant at 0.05 (two-tailed test)
*** significant at 0.01 (two-tailed test)