Study of Undervotes in Recent Santa Monica Elections
Amy Connolly, Julie Walters
March 25, 2004
1 IntroductionMarch 25, 2004
Santa Monica currently uses what is known as Block Voting to elect its city council. In this system, each voter gets to cast as many votes as there are seats to be filled, and the winners are the candidates with the most votes. Although this system is better than the single-member district plurality system used in most American elections, it does share some disadvantages with electing councilmembers by district. If Santa Monica were to adopt Choice Voting for its city council elections, these disadvantages would be eliminated, while the most important advantages of the current system, including city-wide representation, would be retained.
One of the potential problems with Block Voting is that voters are compelled to vote strategically. For instance, a voter, having cast one vote for his or her favorite candidate may be reluctant to cast his or her remaining votes, since those votes may decrease the likelihood of their favorite taking office, voters tend to "undervote," that is, they often do not cast all of their votes.
This study, performed by Santa Monica Ranked Voting, lays out the number of allotted votes that were not cast by Santa Monica voters, and the number of voters that did not cast all of their votes. This study finds that at least a third of Santa Monica voters consistently undervote. This means that voters are not giving a full picture of their political preferences under the current system. Choice Voting would render a more directly proportional result than with the current system, helping ensure that all aspects of our community are fairly represented.
2 How It Works
We obtained the number of ballots cast and number votes cast in the last three city council
elections from the city clerk's office. The statistics for the 1998 and 2002 elections may be
found on the web:
1998 resultsThe results of the 2000 election were obtained via private communication with the city clerk's office and may be found at Santa Monica Ranked Voting's website, www.smrankedvoting.org. For each election, we obtained the number of ballots cast and the number of votes cast. Since each voter is entitled to the same number of votes as there are seats to be filled, the total number of votes that could have been cast is given by:
2002 results
# possible votes = # ballots cast x # seats
Then, the average number of votes cast per ballot is:
# votes per ballot = # votes cast / # ballots cast
The number of votes that were "missing" when votes were tallied (i.e., the number of allotted votes that were not cast) is:
# missing votes = # possible votes -- # votes cast
Now let's say that all undervoters cast only a single vote. Then each undervoter is responsible
for a number of missing votes given by
Then, the number of people that undervoted is:
Note that this scenario is an extreme case, and gives the smallest possible number of undervoters. If anyone who undervotes casts more than one vote, then the missing votes will be distributed among more voters.
3 What We Found
Table 1 and 2 show our results.
Table 1: This table shows the average number of votes cast per ballot in each of the past three Santa Monica city council elections.
Table 2: This table shows the minimum number of undervoters in each of the past three Santa Monica city council elections.
4 Conclusions
This report shows that consistently at least one third of Santa Monica voters are undervoting in city council elections. We contend that this is due to the voting system currently in place, and we recommend that Santa Monica consider Choice Voting for its elections. Choice Voting eliminates the need for strategic votes and maximizes the effectiveness of each vote.
Now let's say that all undervoters cast only a single vote. Then each undervoter is responsible
for a number of missing votes given by
# missing votes per voter = # seats - 1
Then, the number of people that undervoted is:
# undervoters = # missing votes / # missing votes per voter
Note that this scenario is an extreme case, and gives the smallest possible number of undervoters. If anyone who undervotes casts more than one vote, then the missing votes will be distributed among more voters.
3 What We Found
Table 1 and 2 show our results.
year | seats | ballots cast | possible votes | votes cast | votes per ballot |
1998 | 3 | 33,026 | 99,081 | 74,654 | 2.26 out of 3 |
2000 | 4 | 42,659 | 170,636 | 115,800 | 2.70 out of 4 |
2002 | 3 | 30,853 | 92,559 | 70,777 | 2.29 out of 3 |
Table 1: This table shows the average number of votes cast per ballot in each of the past three Santa Monica city council elections.
year | seats | missing votes | minimum undervotes | minimum % that undervoted |
1998 | 3 | 24,427 | 12,214 | 37.0 |
2000 | 4 | 54,836 | 18,279 | 42.8 |
2002 | 3 | 21,782 | 10,891 | 35.3 |
Table 2: This table shows the minimum number of undervoters in each of the past three Santa Monica city council elections.
4 Conclusions
This report shows that consistently at least one third of Santa Monica voters are undervoting in city council elections. We contend that this is due to the voting system currently in place, and we recommend that Santa Monica consider Choice Voting for its elections. Choice Voting eliminates the need for strategic votes and maximizes the effectiveness of each vote.
The Winner-Take-All Problem
“Winner-take-all” is a term used to describe single member district and
at large election systems that award seats to the highest vote getters
without ensuring fair representation for minority groups. In the United
States, these are typically single-member district schemes or at-large,
block-voting systems. Under winner-take-all rules, a slim majority of
voters can control 100% of seats, leaving everyone else effectively
without representation. Problems this leads to include:- Severe under-representation of women, communities of color, third parties, young people, and major party backers stuck in areas where another party dominates. Winner-take-all election systems do nothing to provide representation to any group making up less than half of the population in a given voting district, and the high percentage of the vote needed to win election can be a severe barrier to minority candidates.
- Since many areas are dominated by a single political viewpoint, winner-take-all voting systems will often result in no-choice elections where one party has a permanent monopoly on power, and the winner is effectively predetermined. In the United States, two in five state legislative races go uncontested as a result, and nearly 99% of congressional incumbents win reelection by large margins.
- High percentages of “wasted votes” (that is, votes cast for candidates who do not win). Winner-take-all elections frequently result in more than 50% of votes being wasted. More voters will be represented by someone who they did not help to elect than under any other system.
- Undervoting.
Under at-large systems in particular, voters who feel strongly about a
single candidate will be likely to “bullet vote” (that is, use only one
of their votes) to help their preferred choice win election. In this
way, winner-take-all discourages voters from expressing their full
range of political preferences.
- Decreased voter turnout. With limited choice, and little chance of influencing the outcome of an election under winner-take-all rules, many people will unsurprisingly choose not to participate.
- Divisive campaigns that fail to address challenging issues and ignore entire constituencies. Under winner-take-all, there is no incentive to reach out to opponents or build cross-party support. Negative campaigning is often a sensible and effective strategy.
[more]