This is an insight of the research project I am conducting at Berkeley Geography about the Bay Area demographics and politics. I am interested in the relation between residential choices and political beliefs. I am looking at the political significance of urban (or non-urban) settings as both the cause (the reason for moving in) and the result (as a consequence of local politics) of the idea of society that leads to a residential choice. As residential choice is not always a free choice, I also tackle the issue of gentrification, particularly contentious in the Bay Area. Please find below more about the issues at stake and the full project here.
The Urban Pattern of Votes
The level of urbanity seems to be a consistent explanatory dimension to spatial patterns of electoral geography. Most democratic countries demonstrate a dichotomy in voting behavior between city centers on the one side and less urban areas on the other side. This dichotomy is for instance patent in 2012 Obama election where city centers tended to lean more democratic than their outskirts. Most federal ballots in Switzerland follow the same pattern. The French 2012 presidential election also exhibits such an urban/suburban pattern. This pattern was especially salient regarding the far-right party Front National whose best results occurred in the suburbs of cities. Along with social and demographic characteristics, the distance to central places seems therefore to be a valuable predictor of election outcomes.
Homogenization of Neighborhoods
Along this center/suburb spatial dichotomy, the spatial political pattern in the United States is also characterized, since the 1970s, by the political homogenization of communities. The increasing number of landslide uncompetitive elections reveals the polarization of neighborhoods (Bishop, 2009, §174). Abrams and Fiorina (2012) have criticized the relevance of this empirical observation for explanatory purposes, and especially the hypothesis of the “big sort”. Their main criticism is that this approach only relies on federal elections, whereas local politics do not necessarily exhibit so clear a pattern. Nevertheless, the homogenization bolsters the idea that the place an individual chooses to live in and his voting behavior are related:
“As americans have moved over the past three decades, they have clustered in communities of sameness, among people with similar ways of life, beliefs, and, in the end, politics.” (Bishop, 2009, §95)
These empirical observations do not identify clearly the causal relation between residential choice and political views, but they demonstrate a relation between them.
The Political Significance of Residential Choice
Van Diepen and Musterd (2009), as well as Lévy (2013) have measured “ways of living” in the city as levels of urbanity. Urbanity can be measured considering three component of places: density, social diversity and functional variety.
The level of urbanity of a place is a proposed synthetic indicator comprising these three components and characterizes places from urban to infra-urban (Lévy, 1994). They can also be thought of as different ways of relating to the other: whereas city centers are characterized by density, social diversity and functional variety and tend to house “allophillic” individuals; suburbs, less dense and less diverse, tend to house “allophobic” individuals (Ourednik, 2010). Places also differ on their access to mobility. City centers are served by public transportation, while suburbs rely more on private motorized transportation. Thus, “dwelling” somewhere is not anecdotal of one’s way of thinking its place in society.
Urbanity “is a concept that cannot be restricted to what happens in cities. It is a mode of social relations” (Boudreau, 2010, 55). Louis Wirth also recognized urbanity as a set of sociological characteristics more than just a spatial arrangement of realities. The comprehension of the city and urbanity as a “way of life” (Wirth, 1938) is followed by the work of many scholars in the field of geography and urban sociology who deem the city as an innovative place (Jacobs et al., 1970) that attracts talent and human capital, what Florida calls the “creative class” (Florida, 2006). The mode of social relations included in urbanity could explain the spatial polarity between political parties in the United States.
Residential Choice, Political Choice: Which Comes First?
Two competing models have been devised to explain ur- ban pattern and homogenization of votes: the “big sort” or “friends-and-neighbors influence”. Either people move to places where people are already like them, or they are influenced by their neighbors once they are settled.
The big sort: The big sort model states that people move to live with their own kind (Bishop, 2009). Individuals have different preferences and don’t share the same expectations in term of local public goods, therefore they will move to the place where local public goods services meet their expectations (Tiebout, 1956). Individuals who deem their local government does not align with their preferences favor moving to another community that fits best their political views over trying to influence local policies. This is what Tiebout (1956) calls “voting with one’s feet”: instead of casting a ballot, people move. For him, this is a spatial solution to the “free rider problem”, people benefiting from public goods without sharing the costs (Olson and Olson, 2009). Individuals might want to have a greatest impact on politics and choose therefore a community where their opinion will be more predictably taken into account, where their chosen policies will be actually delivered.
Friends-and-neighbors influence: The “friends-and- neighbors” hypothesis states that people tend to imitate the political behaviors of their neighbors. Once they settle somewhere, they tend to slowly adjust their view of society and political preferences towards the ones of their neighbors. Over time, the community homogenizes its political preferences, exhibiting a “community effect”. If an individual feels good in the place he lives in, he shall credit local politics for it and therefore the dominating political party in the community (Zuckerman, 2005) or simply be “directly influenced by what [his] neighbors are saying at the moment” (MacKuen and Brown, 1987).
Is Residential Choice a Free Choice?
“Big sort” and “influence” are not necessarily mutually exclusive but go together in an evolving circle process. These two hypothesizes both posit residential free choice. Under this postulate, people are not assigned a place to live by external condition but can make free and informed choice of a residential place. The expansion of commuting, in particular, frees people from having to live close to their workplace. However, this choice is not totally free. Dieleman (2001) identifies external constraints limiting the number of residential opportunities available to households (housing prices, local housing policies, etc.), as well as constraints based on their own trajectories (mobility rate, stage in the life circle, size of the dwelling unit, life events, etc.) (Clark and Dieleman, 1996).
The validity of the hypothesis of a causal relation between political views and residential choice is therefore heavily contingent to housing freedom of choice.
The Urban Fabric of Land Value
Housing affordability is constrained by housing market prices. Factors influencing these prices are manifold and differ on two main dimensions: economic and geographical. Following Alonso’s seminal article extending the agricultural theory of rent to the urban case (Alonso et al., 1964), urban rent can be defined as the advantage of places that are spatially close to city centers (CBD, urban amenities...).
In neoclassical models of urban rent, the real-estate market is driven by offer rather than demand. William Alonso states that urban land market depends on the ability of the owner to allocate land to its most profitable use (Alonso et al., 1964). By contrast, marxist theorists contend that urban ground rent is not related to demand but to speculation of investors (Rose, 2013). The economic debate surrounding the “urban rent” is a yet unsettled dispute between neoclassical and marxist economists (Ball, 1985), but neither the neoclassical nor the marxist approach have included space in their explanatory models.
Land valuation can not be considered as if land parcels were isolated the ones from the others (cf. housing “upgrading”) because potential ground rent is a “neighborhood scale phenomenon” (O’Sullivan, 2002, 255). Housing market value of a neighborhood cannot be reduced to a global economic phenomenon in order to explain why reinvestment becomes profitable.
The lack of space in economic models is peculiarly salient when considering the phenomenon of gentrification. Gentrification refers to the phenomenon by which higher income residents replace the lower income residents of a neighborhood, following renovation and “upgrading” of dwellings. Furthermore, the informal term super-gentrification has been coined in connection to New York and London where high incomes of finance people have changed neighborhoods dramatically (Butler and Lees, 2006). The recent return of Silicon Valley rich elites in the city of San Francisco is usually credited for the dramatic changes in the housing market, accelerating gentrification, and thus making San Francisco the model of super-gentrification.
Gentrification is more a descriptive term than an operative analytical tool. The causes and potential effects on public policies are subject to controversies. Gentrification thus makes a good case to explore rent value because it makes easier to decide the answer to the question: What do high housing prices indicate? Speculation? Neighborhood effect? Or something else?
Marxist theorists contend that the speculation that lead to gentrification then creates spatial injustice in the city (Harvey, 2010). The rent gap is the difference between the devaluing of capitalized ground rent and the potential ground rent the owner could achieve by upgrading housing accommodations (Smith, 1996). Gentrification thus corresponds to the filling of the rent gap.
Both neoclassical and marxist economic models of ground value have lead to the exclusive understanding of gentrification as an exogenous creation of value building on the location rent (distance to city center). Gentrification, according to Smith (2002), is enshrined in a global neoliberal strategy to valuing urban “command centers” (Sassen, 1991), an idea also at stake in Beckmann (1969) statistical model determining rent, density and income as power functions of distance to the CBD:
“[T]he mobilization of urban real-estate markets as vehicles of capital accumulation is ubiquitous.” (Smith, 2002, 446)
A geographical approach can be related to the two hypothesis explaining the homogenization of political communities. Sorting and neighbors’ influence provide hypothesizes to include local, actor-based spatial variables in housing market models, and expand them beyond the sole economic scope. Space cannot be left out of the equation for land valuation models.
Economic and geographical micro-scale approaches to gentrification describe the phenomenon but fail to examine why housing prices suddenly rise. The location of newly revalued neighborhoods remains at the same distance to downtown: “they have always been there” as phrased by O’Sullivan (2002, 255). Moreover, housing prices augmentation does not always follows reinvestment. On the contrary, Freeman (2011) points out that gentrification might occur in the reverse order than usually depicted: high income people move to low income neighborhoods, which changes its reputation and the housing market, which in fine attracts speculation. Empirical observations in cities like New York, London (Atkinson, 2000) or Paris (Clerval, 2008; Préteceille, 2007) tend to demonstrate that gentrification relies on actor-based logics of action rather than on economic-driven rehabilitation phenomenons (Sabri et al., 2012). People’s move and housing prices augmentation follows the search by individuals for a quality of life that may be urbanity.