Local governance and the local political career in Canada: a sample dataset.
Cette note de recherche presente et decrit un nouvel ensemble de donnees d'acteurs politiques de la ville de Kitchener en Ontario. L'ensemble de donnees comprend 348 carrieres politiques completes, dans quatre institutions locales, allant de 1912 jusqu'a aujourd'hui. Cette note de recherche vise a montrer que les bases de donnees, a grande echelle, d'acteurs politiques locaux peuvent nous permettre de comprendre : a) les carrieres politiques locales; b) la dynamique des institutions locales formelles et informelles; et c) la nature et les limites de la gouvernance multiniveau locale. En conclusion, cette note offre des conseils pratiques a quiconque desire etablir des ensembles de donnees similaires dans d'autres villes canadiennes.
The local political career in Canada: An overview
While political careers at the federal and provincial levels have received some attention from Canadian political scientists (Docherty 1997 and 2011, Loat and MacMillan 2010, Tremblay and Stockemer 2013, White 1998), we know almost nothing about the local political career. Beyond a few important but basic elements--such as the electoral strength of local incumbents (Kushner, Siegel, and Stanwick 1997), the distinction between full-time and part-time councillors (Sancton and Woolner 1990), and the representativeness of local politicians (Gidengil and Vengroff 1997, Tremblay and Mellevec 2013)--we have little more than anecdotes to guide us. Among the many issues that might occupy our research attention, students of Canadian local politics appear to believe that the local political career is low on the list of priorities.
My own interest in the local career began with an attempt to assemble a rather traditional "collective biography" of local politicians in the city of Kitchener, Ontario between 1890 and 1930, a dataset that I needed in order to compare the development of political careers on municipal council to those on local special purpose bodies (Lucas forthcoming). As I expanded this research over a longer period of time, the smaller dataset slowly grew into a complete collection of everyone ever to have been elected to municipal council, the public school board, the water commission, and the hydroelectric commission from 1912 (when Kitchener became a city) up to the present--a dataset that I used, once again, to try to understand differences between general-purpose and special-purpose local institutions and the long term development of both. (1) The dataset contains 2980 observations, each representing one year of service on one of the four bodies, allowing us to assemble complete career sequences for each of the 348 individuals who have been elected to serve on one or more of these institutions. (2) This is, to my knowledge, the first full dataset of local mayors and councillors in Canada--and it is almost certainly the first such dataset that includes both special-purpose bodies and the general-purpose municipal council within a single, comparable list. (3)
Basic institutional comparisons
What can we do with the dataset? The most obvious place to begin is with a summary of the local career itself. Figure 1 provides a visual summary of average career lengths for each of the four institutions; the line plots the mean total years of service for the individuals who were actually serving on a given year. Most striking, perhaps, are the terms of office--often well over twenty years--of those who sat on Kitchener's two public utilities commissions. The figure for Kitchener's city council suggests a very gradual increase in average career lengths over time (the sharp drop at the end is best ignored; it reflects the fact that the careers of current municipal councillors are truncated at the present). School board careers, while roughly comparable to those of city council, show less evidence of a gradual increase, having hovered between an average of ten and twenty years for much of their history.
These longitudinal averages hide a great deal of internal variation. Figure 2 collapses all of the careers for each institution into a single boxplot: in each case, the line within the shaded area is the median career length, and the entire shaded area captures the interquartile range (the distance from the 25th to the 75th percentile) for each institution. The variation in every case is wide, from one year at the lower end up to thirty, forty, even fifty years at the maximum. Once again, however, the boxplots suggest that we can divide the institutions into two rough categories: the public utilities commissions, with median service lengths of six to seven years, and the more well-known council and school board, with median careers of four years. (4)
Overall, then, local political careers have varied in Kitchener across time and across institutions--variation that is significant enough, in some cases, to suggest real differences among the institutions in policy continuity, administrative capacity, and long-term leadership. (5) In terms of policy continuity and administrative capacity--if not electoral competitiveness--public utilities commissions appear to have enjoyed real advantages, through much of their histories, when compared with the general-purpose municipal council.
The long-term increase in service lengths among municipal councillors is also noticeable in the dataset, an increase that is likely related to the slow but steady post-war expansion of the local political career from the realm of the part-time amateur to that of the full-time politician. (Sancton and Woolner 1990). What is equally interesting, however, is what these increasing career lengths suggest about the informal rules that political institutions can develop to overcome their formal limitations. Consider the example of municipal council. Until 1962, when term lengths were increased to two years, members of Kitchener's city council were elected annually (term lengths increased to three years in 1982, and then to four in 2006). What is clear from the figure, however, is that councillors tended to enjoy lengthy political careers before these formal changes came into effect in the 1960s. While formal terms of office were short, in other words, voters in Kitchener seem to have had no expectation that politicians would serve only for a single term, and municipal councillors were often re-elected year after year for many years. These political expectations, combined with the well-known power of incumbency in Canadian municipal elections (Kushner, Siegel, and Stanwick 1997), meant that defacto term lengths had expanded before de jure term lengths began to catch up. The rules of the game in local government--as in any government--are always the result of a complex interplay between informal and formal rules, and large-scale datasets, such as the careers dataset that we are exploring here, enable us to probe the relative importance of these rules in the local context over time.
Sequences of political careers
The careers dataset also allows us to examine political careers as a whole. Table 1 provides an overview of the twenty-five most common career sequences in Kitchener; since local careers come in all shapes and sizes, the list captures just 70% of all of the sequences in the dataset. The most common career is a two-year stint on the school board; second most common is two years on city council. Because school boards and councils were larger than utilities commissions, and thus represent a larger portion of the dataset, this list is dominated by council and school board careers. The most common careers for local politicians in Kitchener have been relatively short periods of service on city council or the local public school board.
Table 2, which captures the order in which local elected officials moved through Kitchener's political institutions, is a bit more informative. Most common, once again, is exclusive service on either the city council (representing 43% of all careers) or the public school board (34% of all careers). For those who have served on more than one institution, however, the table shows that city council has been an important stepping-stone to further political service; of the 63 individuals who have been elected to more than one of the four bodies, more than half (36 individuals) began their career on council. The most common foundation of multiple-service careers, in other words, is service on city council; in Kitchener, at least, public utilities commissions were indeed, as the nickname once had it, the "senate of local government." (6)
Relations among civic institutions
The Kitchener careers dataset can also be used to explore the relations among political institutions that are created by local political careers. Scholars of multilevel governance and institutional autonomy have suggested a number of ways that we might examine the relations among political institutions: grants and other fiscal relationships, reporting and accountability relationships, shared staffing, powers to appoint members of the board or to veto legislation, and so on (Hooghe, Marks, and Schakel 2010). But another way that institutions can be linked to one another is simply by the careers of those who govern them. If a school board, for example, contains members who have served on the water commission or the city council, that school board may be better equipped to understand what goes on within those other bodies. A dataset of local political careers could therefore reveal important clues about the distance between local institutions, and about the potential implications of those distances for our understanding of local policymaking and inter-institutional cooperation.
Figure 3 provides an overview of these relationships, in the Kitchener case, in the form of a network graph. Each of the nodes (the shaded circles) in the figure represents a single local politician, with the size of the node capturing the total number of years that the individual served. The circles are shaded according to the institution on which each individual did his or her principal service; if an individual served ten years on council and just two years on the light commission, for example, their principal service would be marked as "council." The edges (or lines) between the circles represent years of shared service--thicker edges represent a higher number of shared years. To organize the figure, a computer algorithm iteratively organizes the nodes such that they are "attracted" to those by which they are attached by an edge, and "repelled" by those to which they are not attached. (7) I have rotated the layout so that it can be read roughly chronologically, with earlier politicians to the left and more recent politicians to the right. The figure provides a single, comprehensive, visual overview of all 356 careers in the Kitchener dataset.
What can we learn from the figure? Most obvious, perhaps, is the size of the light-gray and medium-gray circles, representing the two public utilities commissions; as we already know, many of those who served on those commissions did so for years--even decades. The gradually increasing size of the dark-gray circles (representing those who did their primary service on city council) also captures the increasing career lengths on city council that we have discussed above. What may be most interesting about the figure, however, is the distance between the dark gray circles, which stretch out along the top of the figure, and the white circles along the bottom. City council and the public school board appear to have had few crossover politicians in Kitchener--and the number of crossovers appears to have declined over time.
This possibility merits a bit more investigation. Table 3 provides a summary of career "elements" in Kitchener from 1912-2014; it captures, in other words, the institutions on which each person served without attention to the order in which the service occurred. Once again, the most common careers involved exclusive service on city council or the school board; just 10% of those who served on council or the school board have served on council and the school board. Since the end of the Second World War, however, these "crossover" politicians have declined significantly--just eight individuals have served on both city council and the school board since 1945, and just a single individual has done so since 1962. To put this another way, more than two thirds of those who have served on both council and the school board did so in the first thirty years of the dataset. These relational data suggest that school boards and local municipal councils have become, since the end of the Second World War, two separate political worlds.
The source and extent of this shift from "one world" to "two worlds" of local policymaking will become clear only in the context of further comparative research. Nevertheless, the Kitchener experience suggests two reasons for the divergence between the world of municipal services and the world of schooling. The first is the opportunity cost created by the increasing length of post-war political careers; as service lengths extended in the postwar period, aspiring local politicians may have needed to make a clearer choice between the "municipal track" and the "schools track" at the local level. (8) The second is the gradual expansion in the geographic scale of Ontario school boards after the Second World War (Gidney 1999), a process that began in Kitchener with the Kitchener and District School Board in 1967 and continued with the even larger Waterloo County School Board in 1969--changes that put considerably more distance between Kitchener's municipal system and its school system. (9) These factors appear to have been important in the Kitchener context; testing their importance, and the extent of the divergence between schools and municipalities, should be a priority for comparative careers research in the future.
The lessons--and potential--of local actor datasets
In our analysis above, we have used a complete dataset of local political actors in Kitchener, Ontario, to survey the careers of 356 distinct individuals across time. I have briefly shown that these data can be useful for understanding the local political career itself: the length of such careers, the differences in typical careers across institutions, and the movement of individuals from one institution to another. We have also used a network presentation of the data to explore the links among local institutions that are implied by political careers.
All of this is possible with a single dataset from a single city. And having assembled the data, it would be possible to extend them in a number of ways: for instance, we might use the dataset to select a smaller set of representative careers (or, alternatively, very unusual careers), and to use newspaper and archival sources to understand how actual local actors made decisions about their political careers. Such research would help us to understand how politicians' individual decisions aggregated into the patterns that we have explored here. In my view, however, the greatest potential for these datasets is in the opportunity for comparison across both cities and levels of government. Comparison of career datasets would allow us to explore some of the foundational questions of local government in Canada: to what extent are changes to local political institutions determined internally, and to what extent are they determined by outside forces such as provincial legislation? How do relationships among local political institutions differ across cities and across time? How can we best describe the "worlds" of local governance within the wider local state, and how have these political worlds evolved? How do different local careers, and the stability or instability of those careers, affect local policymaking? Local actor datasets enable us to address these questions in a systematic and directly comparable way.
A second extension of the dataset would be to incorporate additional levels of government. Many provincial and federal politicians begin their careers at the local level, and some have argued that the experiences of these politicians at the local level have an important effect on the policies that they develop once they have moved into provincial or federal politics (Ibbitson 1997: 221-222). By adding data on provincial and federal careers to these local datasets, we could explore the kinds of local service careers that members of provincial and federal parliament undertake before or after their parliamentary careers begin (Barrie and Gibbons 1989). These integrated datasets would be particularly useful for comparing careers by gender, allowing us to describe not only the relative proportion of political positions that are occupied by women but also the nature and length of their overall career paths in comparison to men. (10)
There are, of course, no "magic bullets" for any research question, no all-encompassing datasets that enable us to answer all of our questions about the nature and development of the local state. Local careers datasets provide just one more angle on these questions. Nevertheless, much recent work in the study of urban politics has emphasized the need--and the enormous potential--for comparative urban research in order to address large questions about institutional change, governance, and policymaking (Mossberger, Clarke and John 2012). To carry out such comparisons requires that we develop reliable means by which to actually compare very different cities to one another. Datasets of local political careers can help us to do this work.
Conclusion: The practicalities of data assembly
Assembling career datasets is simple--but it can also be time-consuming. The central practical goal for anyone who is interested in carrying out such work should therefore be to begin with the sources that most easily provide reliable lists of names, and proceed to more difficult sources only when necessary. In my experience, the best place to begin is with annual reports, which typically exist only for public utilities commissions and are usually available in local or provincial archives. Except when considerable distance is involved, the time cost of travelling to the archive is well worth spending given the accuracy and simplicity of the annual reports, which frequently include the names of the commissioners in the first few pages. When such reports exist, it is possible to compile a century's worth of names in less than a day.
When annual reports are unavailable, the next best source is the newspaper. Election results for council, school boards, and other races are prominently featured in local newspapers, and while it can take time to locate the results, the information in the news reports is generally clear and accurate (some fortunate researchers, in places like Toronto, will be able to skip the microfilm and consult searchable digital copies of the newspaper; others may discover compiled scrapbooks containing articles on local elections in their city's local archive). Researchers should be aware, however, that acclamations were very common in municipal politics; when results seem to be missing from the newspaper's coverage of the election, the researcher will have to consult the report on the municipal nomination meeting, which usually occurred several weeks before the election, in which local acclamations are discussed.
The vagaries of microfilm scan quality mean that newspapers occasionally fail to provide the necessary election results. In this case the third option is to consult the minutes for the institution in question. These minutes are usually easy to access in local archives, and they are generally accurate and easy to follow. Nevertheless, researchers who consult institutional minutes must be attentive both to attendance and name spelling. Because of absences, it is often necessary to consult the minutes for five or six meetings in a year before one has a complete list of the individuals for that year, and recording secretaries, for all their diligence, were often rather creative about the spelling of names in the minute books. The time involved in decoding the faded, barely legible handwritten scrawl in early minute books, and in flipping back and forth among several months' worth of meetings in order to compile a full list of names, makes recorded minutes most useful only when other sources have been exhausted.
Once a dataset is compiled, researchers should check the list for accuracy with a particular emphasis on name spellings--it is important that each distinct individual in the dataset is actually distinct, and not simply a variant spelling of the same name. Statistical analysis of the completed dataset can be accomplished with a range of software packages and techniques, including many--such as sequence analysis (MacIndoe and Abbott 2004)--that I have not described in detail here. Researchers who wish to carry out analyses similar to those that I have done above can download the original datasets, Stata do-files, and Gephi network files from the data archive on my personal Website (11).
The systematic accumulation of careers datasets is straightforward and simple--though occasionally time-consuming--work. It is the kind of task that lends itself well to teamwork--and even, with proper quality controls, to "crowdsourcing." I have written this note in an attempt to demonstrate the range of research questions that can be tractably addressed using such datasets. In this comer of urban politics, as in so many others, our work has only just begun.
(1) Unfortunately, this dataset does not include the Kitchener Separate School Board. While I have a complete list of separate school board members from 1912-1927 and from 1945 to the present, the Kitchener Separate School Board Archives are missing the board minutes from 1928-1944. Despite extensive searching, I have not been able to fill in these missing years using local newspapers, which do not appear to contain reports on Separate Board elections during this period.
(2) To avoid unnecessarily truncating the data, the dataset includes the full careers for everyone who was a member of one of the four institutions in 1912 or thereafter, including their years of service before 1912. Some observations in the dataset are therefore for years before 1912.
(3) I have consulted the following sources to construct this dataset: Kitchener News Record and Daily Telegraph 1890-2014; Kitchener Council Minute Books (Kitchener Corporate Archives); Kitchener Water Commission Annual Reports (Kitchener Corporate Archives); and Kitchener School Board Minute Books (Waterloo Region District School Board Archives). The closest existing dataset to this one, to my knowledge, is that of Kathryn Kopinak in London, though it covers a considerably shorter period of time; see Kopinak 1985.
(4) It is important to remember that these figures include all careers in the dataset, and that there are many careers in the early period that lasted just a year or two. Median career lengths for school boards and councils would be higher if we were to examine only the period after, say, 1980 or 1990.
(5) In theory, rapid turnover would not necessarily reduce continuity and capacity, provided that those politicians shared similar goals and possessed similar levels of policy knowledge. In practice, however, I think that the stability and continuity provided by very long political careers, together with the increased knowledge gained by the politicians themselves, extends the capacity of local institutions. There is little question in the Kitchener case that the long-serving members of the hydro and water commissions became the dominant voices on those boards.
(6) See for example Kitchener-Waterloo Record, October 25, 1978.
(7) The algorithm is "Force Atlas 2," with edge weight influence set to the default value (1.0).
(8) There is some evidence for this increasing professionalization and time pressure in the newspaper articles of the period. See, for instance, Kitchener Waterloo Record May 4, 1963; October 25, 1965; and February 14, 1967.
(9) For details on these changes, see Kitchener Waterloo Record April 7, 1967; November 15, 1967; December 10, 1968.
(10) Unfortunately, the dataset does not currently include a gender variable. To generate the variable would in most cases be quite easy, but a small but persistent minority of cases are challenging because (a) some names are ambiguous as to gender and (b) in earlier years, women were sometimes listed in minute books by their husband's name (for example, Mrs. John Doe, but sometimes without the "Mrs."). Nevertheless, the need for more research on gender and the municipal career is clear (see Tremblay and Mellevec 2013) and I hope to address the ambiguities in the dataset and pursue this issue in a future essay. For a good start on this front, see Kopinak 1985, p. 404.
Barrie, Doreen, and Roger Gibbins. 1989. "Note: Parliamentary Careers in the Canadian Federal State" 22 (1): 137-45.
Docherty, David. 1997. Mr. Smith goes to Ottawa: Life in the House of Commons. Vancouver: UBC Press.
--. 2011. "The Canadian Political Career Structure: From Stability to Free Agency." Regional & Federal Studies 21 (2): 185-203.
Gidengil, Elisabeth and Richard Vengroff. 1997. "Representational Gains of Canadian Women or Token Growth? The Case of Quebec's Municipal Politics." Canadian Journal of Political Science 30 (3): 513-537.
Gidney, R D. 1999. From Hope to Harris. Toronto: University of Toronto Press.
Hooghe, Liesbet, Gary Marks and Arjan H. Schakel. 2010. The Rise of Regional Authority. New York: Routledge.
Ibbitson, John. 1997. Promised Land: Inside the Mike Harris Revolution. Toronto: Prentice Hall Canada.
Kopinak, Kathryn M. 1985. "Women in Canadian Municipal Politics: Two Steps Forward, One Step Back." Canadian Review of Sociology and Anthropology 22 (3): 394-410.
Kushner, Joseph, David Siegel, and Hannah Stanwick. 1997. "Ontario Municipal Elections: Voting Trends and Determinants of Electoral Success in a Canadian Province." Canadian Journal of Political Science 30 (3): 539-553.
Loat, Alison and Michael MacMillan. 2010. The Accidental Citizen? Toronto: Samara.
Lucas, Jack. Forthcoming. Fields of Authority: Special Purpose Governance in Ontario, 1815- 2015. Toronto: University of Toronto Press.
MacIndoe, Heather and Andrew Abbott. 2004. "Sequence Analysis and Optimal Matching: Techniques for Social Science Data." In Handbook of Data Analysis, edited by Melissa Hardy and Alan Bryman. London: SAGE, pp. 387-406.
Mossberger, Karen, Susan E. Clarke and Peter John. 2012. "Studying Politics in an Urban World: Research Traditions and New Directions." In Oxford Handbook of Urban Politics, edited by Peter John, Karen Mossberger, and Susan E. Clarke. Oxford: Oxford University Press, pp. 3-8.
Sancton, Andrew and Paul Woolner. 1990. "Full-time municipal councillors: a strategic challenge for Canadian urban government." Canadian Public Administration 33 (4): 482-505.
Tremblay, Manon, and Anne Mellevec. 2013. "Truly More Accessible to Women Than the Legislature? Women in Municipal Politics." In Stalled: the Representation of Women in Canadian Governments, edited by Linda Trimble, Jane Arscott, and Manon Tremblay. Vancouver: UBC Press, pp. 19-35.
Tremblay, Manon and Daniel Stockemer. 2013. "Women's ministerial careers in cabinet, 1921-2010: A look at socio-demographic traits and career experiences." Canadian Public Administration 56 (4): 523-541.
White, Graham. 1998. "Shorter measures: the changing ministerial career in Canada." Canadian Public Administration 41 (3): 369-394.
Jack Lucas is Assistant Professor, Department of Political Science, University of Calgary.
Caption: Figure 1. Mean Years of Service, By Year
Caption: Figure 2. Career Lengths by Institution
Caption: Figure 3. Service Relationships Among Elected Local Politicians, 1912-2014
Table 1. 25 Most Common Career Sequences in Kitchener, 1912-2014 Career Sequence N % SS 30 8.43 CC 23 6.61 C 20 5.75 CCC 16 4.6 CCCCC 16 4.6 SSS 15 4.31 SSSS 15 4.31 SSSSSS 11 3.16 CCCC 10 2.87 CCCCCC 10 2.87 CCCCCCC 8 2.3 SSSSS 8 2.3 SSSSSSSSS 7 2.01 CCCCCCCCCCCCC 6 1.72 CCCCCCCCC 6 1.72 CCCCCCCCCC 5 1.44 CCCCCCCCCCC 4 1.15 CCCCCCCCCCCCCCC 4 1.15 S 4 1.15 SSSSSSSSSS 4 1.15 SSSSSSSS 4 1.15 CCCCCCCCCCCCCCCCCCC 3 0.86 CCCCCCCC 3 0.86 SSSSSSSSSSSS 3 0.86 SSSSSSSSSSSSSS 3 0.86 C = Council L = Light Commission S = School Board W = Water Commission Twenty-five most common career sequences in Kitchener between 1912 and 2014. Each letter under "Career Sequence" represents one year: the most common sequence, for instance, is a two-year career on the school board (SS); the second most common is a two-year career on council (CC); and so forth. (N) is number of distinct individuals with each career; (%) is percentage of all distinct individuals. Table 2. Career Order in Kitchener, 1912-2014 Order N % C 148 42.53 s 120 34.48 CL 16 4.6 SC 13 3.74 L 11 3.16 CS 8 2.3 w 6 1.72 cw 5 1.44 CSC 2 0.57 cwc 2 0.57 ses 2 0.57 SL 2 0.57 LC 2 0.57 WC 2 0.57 C = Council L = Light Commission S = School Board W = Water Commission Order of local career service in Kitchener from 1912-2014, ignoring the number of years served in each institution. The most common career "order" is to serve exclusively on council; the second most common is to serve exclusively on the school board; the third most common is to serve first on council and then on the light commission; and so on. (N) is number of distinct individuals; (%) is percentage of all distinct individuals. Table 3. Career Elements in Kitchener, 1912-2014 Elements N % C 148 42.53 s 120 34.48 cs 25 7.18 CL 20 5.75 L 11 3.16 CW 9 2.59 W 6 1.72 CSL 2 0.57 SL 2 0.57 LW 2 0.57 CSW 1 0.29 CLW 1 0.29 sw 1 0.29 C = Council L = Light Commission S = School Board W = Water Commission Elements of local political careers, ignoring the order in which those elements occurred. Most common career in Kitchener is to serve only on council or a school board; third most common career is to serve on council and the school board; and so on. (N) is number of distinct individuals; (%) is percentage of all distinct individuals.
Please note: Illustration(s) are not available due to copyright restrictions.
|Printer friendly Cite/link Email Feedback|
|Publication:||Canadian Public Administration|
|Date:||Dec 1, 2015|
|Previous Article:||Alternative routes: intergovernmental relations in Canada and Australia.|
|Next Article:||The J.E. Hodgetts Award Le Prix Roland Parenteau.|