77
makes the static assumption more complicated to be held at the mayoral level. Thus, the
proxies for understanding patterns of ambition are less precise. Nevertheless, it is
possible to observe some tendencies. Taken all the periods and districts together, and
based on the available data (11,789 observations for 2,238 municipalities between 1983
and 2007)43, I can compare the rate of effective incumbency across years. I am not
considering candidacy here, but only whether the mayors have remained the same
across time.
Using these data, sixty percent of mayors have not served more than one term, as
showed in Table 6.5. While a quarter of the sample served two terms, just 14% stayed
three periods or more. Such a distribution would suggest that ambition might not be as
static as for the case of governors. However, as mentioned, missing observations abound
for periods before 1991. Thus, the likelihood that longer careers are underrepresented is
substantive, especially for smaller municipalities.
Table 6.5: Tenure of Mayors
|
Tenure |
Freq. |
Percent |
|
1 |
4,294 |
59.78 |
|
2 |
1,763 |
24.54 |
|
3 |
756 |
10.52 |
|
4 |
251 |
3.49 |
|
5 |
93 |
1.29 |
|
6 |
22 |
0.31 |
|
7 |
4 |
0.06 |
|
Total |
7,183 |
100 |
43 A naive calculation would consist in multiplying the N of municipalities (2238) by the seven time
periods. If this was the theoretical sample, my sample would equal a 75.25% of the real data. However, as
mentioned, districts have been created and mergedacross time. Thus, I think my data share is actually
bigger than .75.