City of Vancouver

Chapter 5.
Results of the survey


In response to Council's request for information on government structure in other cities, the Office of the City Clerk prepared a questionnaire on systems of electing city government. In early January 1996, the survey was faxed to over 50 cities in North America reflecting a range of population size, ethnic makeup, and geographic regions. To date, 39 responses have been received, often with accompanying documentation as well as the completed survey form.

The basic data from the returned questionnaires is presented at the end of this report in spreadsheet form. The cities are ranked by their population, and Vancouver has been included in the data. (The survey was also sent to six cities in Australia, England and New Zealand. Their responses are not included in this report as they were not part of Council's request for information and represent very different political cultures.)

It should be remembered when reading the survey data that the population size which gives each city its ranking is that of the reporting municipal government. This is why, for example, Calgary and Edmonton are ranked above Toronto: while most people think of Toronto as a city of over three million people, the actual municipality of the City of Toronto includes only 15 percent of the metropolitan population and an even smaller proportion of the geographic area. Readers should also bear in mind that the data reflects the information returned by the respondents and has not been verified through other sources.


Survey questions

The survey asked the following questions, most of which included subsections for more detailed responses: The survey form also asked whether any studies or reports on the election system had been written recently, and requested that the City Clerk be sent a copy. Several of the municipalities sent back additional material in response to this request, and some of this material is reproduced or excerpted in Volume II of this report. The survey finished by asking for any comments, insights and experiences that the respondents might wish to offer about methods of selecting municipal officials and voting practices. (The survey also included questions regarding voting procedures, but these are not reflected in this report as they are beyond the scope of Council's request.)

Additional information was received from a smaller number of cities regarding area and population density, and regarding the costs associated with municipal governance. This information is summarized in two additional spreadsheets: "Cost of supporting city council in 11 North American cities" and "Major Canadian cities compared by area and population density." These follow the main survey spreadsheet.


System for electing city council

Of the 40 cities in the survey (including Vancouver), a majority of 25 currently use a pure ward system. A further 8 use a mixed system which elects some councillors on a ward basis and some on an at-large basis. Seven use an at-large system.

A closer look at the figures shows a regional pattern: of the seven cities using an at-large system, six are Pacific Coast cities (Huntington Park, Portland, San Francisco, Seattle, Vancouver, and Victoria). This pattern has been observed in larger-scale research on American cities. As one analysis puts it:

The differences by geographic division are generally consistent with the conventional wisdom regarding the types of municipalities in which the municipal reform movement was the strongest. At-large elections are most prevalent in Pacific Coast (81.3%) and suburban cities (68.4%). Mixed systems are most likely to be found in West South Central (51.7%) and Central cities (49.9%). District [ward] elections are most prevalent in West North Central (22.5%) and Central cities (16.0%). (Renner, p. 67)

Another pattern noted in the same analysis is that larger American cities are more likely to have a mixed or pure ward system, while smaller municipalities are more likely to have at-large systems.

One city, Kansas City, uses wards for primary elections, then runs the top two candidates from each ward in an at-large general election.


Number and population of wards

In the 33 cities with wards or mixed systems, the average number of wards was 14.7. There are wide variations however, from New York and Montreal's 51 wards each down to the 5 reported by Arlington, Texas.

Using population statistics supplied by the respondents, it can be seen that there is a wide variation in ward sizes. The average size was 40,047. Three of the four largest American cities in the survey have wards with over 130,000 people in them, and Edmonton stands out among Canadian cities with wards of over 100,000. On the other hand, one very large city, Montreal, has wards of just over 20,000 people, and two mid-sized American cities, Nashville and Baltimore, report wards of approximately 14,000 and 28,000 respectively.


How ward boundaries are determined and revised

A variety of methods are used to determine ward boundaries. Population is almost always a factor. Some municipalities such as Berkeley and Fort Worth strive to make ward populations as equal as possible, while others permit variances of 10 percent or more (Montreal and Saskatoon). In many cities, boundaries may also take into account such factors as "community of interest," history, geographic factors (transportation arteries, natural boundaries) and population trends.

Most ward boundaries are revised on a regular basis, notably in the United States where many cities revise their boundaries every ten years using the population data provided by the national census. Since boundary-making can be very controversial, Minneapolis stressed in its survey form that impartial or bi-partisan bodies are used in ward revision, while Pittsburgh refers it to the court system.


Age of the present system

Well over half of the cities surveyed have changed their system of election in one way or another in the past 25 years, and four have done so in the past 5 years. In a majority of cases, the change has been to a mixed system. This fits with Renner's study of U.S. municipalities, which notes that the number of at-large systems has been falling since 1981, while the number of mixed systems has been rising. Pure ward systems have stayed static in that time.

An examination of how the present systems came into effect indicates that 17 were brought in by legislation (i.e., through the action of a senior level of government), while another 20 came in through local referendum. These data should be approached with caution, however, since some jurisdictions pass legislation enabling the creation of wards before local referenda or by-laws actually make the change.


Election of the mayor

All but one of the cities surveyed elect their mayor at large and independently of council. The exception is the small city of Huntington Park, California, where one councillor is elected mayor by the other members of council.


Systems at issue

The survey asked respondents if there had been significant debate in the last decade regarding the system used to elect city government, and the results confirm that Vancouver is not alone in its discussion of this issue. Two-thirds of the cities responded "yes" to this question and reported a wide variety of topics under debate. These include ward/at-large controversies, consolidation with regional or county governments, downsizing of council, boundary disputes, and changes to alternative voting systems.

It is striking that three of four major cities on the Pacific coast - San Francisco, Seattle, and Vancouver - are putting the question of wards to voters in 1995 and 1996. Seattle has recently rejected changing its at-large system (see Volume II of this report for the amendment proposal that was put to Seattle voters). The survey form received from the fourth city, Portland, indicates that the issue may soon come up there too.


City population

As noted earlier, the survey results have been ranked according to the population size of each municipality. The largest is New York with over 7.5 million people, while the smallest is Huntington Park, California, with just under 60,000. A closer look at cities similar to Vancouver in size is found later in this chapter.


Number of voters per representative on council

While the number of people per ward was described earlier in this chapter, it is also worth looking at the number of people per representative on council. To arrive at this figure (see page 7 of the survey spreadsheet) we divide the city population by the number of council members not including the mayor unless (as is the case in only one city) the mayor is a council member who is chosen by the rest of council. This is done in recognition of the very different status of the mayor and council members in most municipalities.

The results are similar to those calculating the number of people per ward, with a few outstanding cases such as Portland with its very low number of council representatives per population. Vancouver is a little above the average, with a little over 50,000 people per council member.


Voter turnout

The participation levels of registered voters as represented by turnout in recent elections ranges from a high of 57 percent in New York to a low of 7.8 percent in Arlington, Texas. The average turnouts in the past two elections were 38.5 percent and 42.8 percent respectively.

There is no obvious correlation of voter turnout with the voting system or size of wards. Voter participation is likely affected strongly by factors such as how frequently elections are held, whether the elections are staggered, and (in the United States) if the election of city council coincides with state or federal elections. For instance, the data shown for San Diego apply to staggered elections for city councillors only. Turnout for mayoral elections tends to be twice as high.

Vancouver's voter turnouts in the 1993 and 1990 municipal elections were 35 and 52 percent respectively, and the average in the last ten elections has been 42 percent.


Vancouver and its peer cities

Ranked by population, Vancouver is number 14 of the 40 cities, with a little over half a million people. It can be classed, somewhat arbitrarily, with a band of 12 medium-large cities containing populations in the 400,000 to 800,000 range, including Calgary at the top end and Atlanta at the lower end.

Viewed from this angle, it can be seen that one-third of these cities elect their councils at large, with half using pure ward systems and one-sixth using mixed elections. The average number of wards (eight cities) is 16.8, and the average population per representative on council is 50,931 - almost exactly the figure for Vancouver.


Area and population density

In comparison with the 15 other Canadian cities included in the survey, Vancouver ranks sixth in population but twelfth in area, with 44 square miles included within the municipal boundaries. Toronto, with a larger population, is even smaller at 39 square miles. Some municipalities are extremely spread out in comparison, notably Halifax with over 2,000 square miles and three major prairie cities with over 200 square miles each. (See the spreadsheet "Major Canadian cities compared by area and population density.")

When these figures are translated into population density, the three largest cities in terms of population - Montreal, Toronto and Vancouver - also turn out to have the highest density, ranging from Toronto, with over 16,000 people per square mile, to Vancouver's 11,545. Victoria, the only other at-large city in Canada included in these figures, also has a small land area and relatively high population density at 8,470 people per square mile. The range of densities of cities electing their councils by ward is very wide from a low of 153 (Halifax) to 16,282 (Toronto).


Costs of supporting city councils

In carrying out this survey, an attempt was made to collect information permitting comparison of the costs of supporting different municipal councils in cities roughly comparable to Vancouver in population. This information, presented in the spreadsheet entitled "Cost of supporting city council in 11 North American cities," should be approached with caution. Each city has its own way of accounting for the staff and budget dedicated to the mayor and councillors; moreover, councillors' responsibilities and working schedules vary widely from city to city. Finally, only 10 cities provided figures for this part of the study.

No particular pattern is visible between at-large and ward cities. The most important factors affecting costs are the number and type of staff provided to support the activities of an elected mayor and councillors. Some city councils are supported not only by secretarial and clerical personnel, but have their own research, legal, and policy staff. Four of the cities provide one or more executive assistants to councillors.

An at-large city, Portland, had the highest cost per citizen ($7.95) and the second highest number of staff dedicated to these activities (47). It must be noted that Portland's commissioners are full-time and have both legislative and administrative responsibilities for a variety of portfolios. It can be assumed that some of their costs would, in another city, be included under the budget of specific city departments. Toronto, with a ward system, has comparable costs ($7.67), due in part to the number of staff provided to support the individual councillors and the constituency offices which the 16 councillors maintain in their wards.

With 9 staff and costs of $2.21 per capita, Vancouver's expenditure on support dedicated specifically to Council is relatively low.


For further information

For details on particular cities, readers may wish to consult the second volume of this report, which contains documentation sent by several cities in response to our survey.

Again, a broad overview of trends in municipal governments in the United States can be found in Tari Renner and Victor S. DeSantis's "Contemporary Patterns and Trends in Municipal Government Structures."


[Previous Page] [Contents] [Next Page]

Comments or questions? Please contact info@city.vancouver.bc.ca
[City Homepage] [Get In Touch]

Last Updated: June, 1996
(c) 1996 City of Vancouver