Abstract
This paper considers the importance of age in delineating urban space, the latter operationalised as high-density living. Many cities have experienced an increase in inner city living contributing to gentrification. Today, inner cities contain more amenities, public transit and housing options than in the past but there are also growing affordability concerns owing to rising prices. Especially young adults, sometimes dubbed Millennials, are making location decisions in a context of lower employment security, higher costs and continuing high-density re-development that now extends into suburban areas in some cases. The analysis in this paper shows evidence of a youthification process that results in an increasing association of high-density living with the young adult lifecycle stage. The higher density areas remain young over time as new young adults move into neighbourhoods where there are already young people living, and they move out if their household size increases. Youthified spaces have become characterised by small housing units that are not generally occupied by households with children. Additionally, some areas are exhibiting generational bifurcation as both older and younger adults live in some higher density areas. Youthification is driven by a combination of lifestyle, demographic, macro-economic and housing market changes that require further investigation. The youthification process is not replacing, but occurring alongside, gentrification and points to young age as a delineator of high-density living becoming more important over time. However, immigration, measures of social class and household size still remain the most important explanatory variables of high-density living.
Introduction
In earlier work I have argued that urban space is becoming increasingly generationed (Moos, 2014a, 2014b; also see McDaniel, 2004). Concentrations of young adults in central cities have grown dramatically since the onset of an ‘urban renaissance’ starting in North America as early as the 1960s (Atkinson, 2004). This has left urban space more visibly defined by generation and age (Chatterton, 1999). Research in sociology, geography and economics points to the importance of adding age as a more explicit variable in analysis of social differentiation as generational differences have grown because of ‘social acceleration’ (Constable, 1996; Rosa, 2013).
In this paper, I expand on the idea of generationed space via an empirical test of a process I call the youthification of urban space at the intra-urban scale. Youthification occurs as young adults increase in share of the total population in specific neighbourhoods. The youthification process differs from gentrification – an increase in social status of a neighbourhood – in that the former is not as explicitly a class-based process, although the two are not mutually exclusive. Gentrification, when viewed as a series of stages involving ever slightly wealthier but more risk averse in-movers (Clay, 1979; Kerstein, 1990), arguably has set the stage for a broader segment of the population to move into cities.
Youthification, then, is in part ‘the city going mainstream’, as young people across the income spectrum are found in urban spaces – some are gentrifiers, others are renters with more modest means. Brahinsky (2014: 52) argues that ‘[g]entrification is capitalism playing out in the landscape. It is essentially our economy’s urban form’ (also see Newitz, 2014). Several scholars would argue that gentrification is the urban form of social, cultural and structural changes, or a combination thereof (Ley, 1996; Skaburskis and Moos, 2008; Smith, 1979). I argue here that youthification is the contemporary lifestyle, demography and socio-economic condition of the young ‘playing out in the landscape’ but leave the debate as to whether this is a demand- or supply-side driven process aside (also see Townshend and Walker, 2010). This line of inquiry clearly would benefit from consideration of economic versus amenity-based explanations of location decisions, as one of the reviewers noted. I do not view these as mutually exclusive, however; and will deal with the motivations behind young adults’ location in more detail in forthcoming work.
Inspired by Ley’s (1986) work that examines the factors best explaining gentrification, in this article I will consider the relative importance of age as an explanatory variable of urban living vis-à-vis other demographic (lifestyle) factors associated with density. In the press and public policy discourse, higher density developments are often equated with gentrification because of the replacement of older and cheaper housing or industrial land by condominium apartments, a process Lehrer and Wieditz (2009) and Walks (2014) call ‘condofication’, which is essentially the result of ‘new built gentrification’ (Lees et al., 2010). However, not all condominium apartments are owner-occupied, nor are they necessarily inhabited by gentrifiers. Increasingly, young adults may rent condominium units or share them with roommates because of the prohibitive cost of housing or to avoid a long commute (Clapham et al., 2014; Johnson, 2011).
In Toronto, Montreal and Vancouver – the case studies in this paper – the percentage of young adults living in the highest quartile tracts by density is 36%, 29% and 40%, respectively (based on 2006 Statistics Canada data), as compared with 27%, 21% and 29% for the total population. The higher share of young adults in high-density areas, relative to the total population, warrants further consideration aside from the often assumed link to gentrification.
I begin below with a conceptualisation of the youthification process in the context of the literature exploring gentrification and urban restructuring in North American cities. While the empirical analysis solely focuses on the link between young age and density at the intra-urban scale in three Canadian cities, the conceptual discussion of youthification also considers the broader North American context and the inter-urban dimensions.
The three Canadian cities were chosen for analysis as they are known for their history of gentrification (much like large US cities), and particularly in Toronto and Vancouver the dramatic growth of condominium apartments in the inner city and transit-serviced suburbs that have experienced an influx of young adults (Kern, 2010; Quastel et al., 2012). Toronto has the largest number of new high-rise condominium apartments under construction in the world, exceeding New York, which still has the highest total number of high-rise condominium apartments (Yelaja, 2012).
From gentrification to youthification
In the 1970s and 1980s, several researchers started to ask about factors leading to the return of artists, and later higher income earners, to the inner city through gentrification (Ley, 1986). While gentrification continues today and has spread to new places (even suburban areas), new variants on gentrification such as ‘condofication’ (Lehrer and Wieditz, 2009) are being observed. The growth of condominium apartments that characterises several of North America’s largest metropolitan areas differs from traditional gentrification. Historically, gentrification was associated with the upgrading of existing housing by a ‘new middle class’ using sweat equity, and at times even led to declining densities (Ley, 1996). Condofication has a corporate dimension, is commonly ‘state-led’ and significantly increases densities because of its high-rise forms (Lees et al., 2010).
The growth of condominium apartments is at least in part related to a changing regulatory environment, with planning frameworks across North America increasingly emphasising higher density (densification) and more compact developments as a means to stem suburban sprawl (Filion and Kramer, 2011; Quastel et al., 2012). Densification is at times linked directly with gentrification – however, as we will see in the analysis, youthification is better able to account for high-density living than gentrification alone, although neither are as important as immigration. As inner city areas became more expensive (Skaburskis and Moos, 2008), and planning frameworks emphasised compactness and polycentricity, more apartments also started being built in the suburbs, particularly in Vancouver (Moos, 2012).
I am obviously not the first to consider age in urban analysis. The gentrification literature has long considered young age as one indicator of gentrification, and age is still being used to operationalise gentrification in contemporary studies, along with educational, occupational and demographic variables (Ley, 1986; Skaburskis, 2012; Zukin, 1982). There is also a literature that has considered the growing diversity of households resulting from the growing non-linearity and delay of lifecycle stages – for instance, getting married, having children, buying a house – which used to be more linearly associated with age (Townshend and Walker, 2010).
While there is thus certainly a literature that considers the importance of age in shaping societal change (see Hopkins and Pain, 2007; McDaniel, 2004; Vanderbeck, 2007), age has not received as much attention as traditional measures of urban socio-spatial differentiation, for instance immigration, ethnicity, gender and class. The literature on the ‘creative class’, which is now commonly equated with gentrifiers (even if not always accurately), often refers to creatives as young in age but in general does not deal with the age dimension explicitly either (Florida, 2002). Today, age is less frequently considered in gentrification studies because the process has become associated with people in different age ranges, and in some cases actually prices younger people out of some central neighbourhoods (Kennelly and Watt, 2012; Lees, 2000).
The definition of young adults has also changed, increasingly including people in their early 30s whereas in the initial stages of gentrification, young adults were generally considered to be in their 20s (Moos, 2014a). The aging of the young adult definition has occurred as the youthful lifecycle stage has become extended (Chatterton and Hollands, 2002) – this is an outcome of complex interrelationships among growth in smaller households, increasing educational attainment, growing female labour participation, delay in childbearing, and the promotion of inner cities as residential and recreational spaces (see Moos, 2012, 2014a, 2014b; van Diepen and Musterd, 2009).
Geographies of youthification
Previous research has already established that at the inter-urban scale, there is a geography emerging of younger and older cities (Rosenberg and Wilson, 2010). Glaeser et al. (2001) found that the young adult labour force is attracted to cities with positive economic prospects and the presence of lifestyle amenities. Table 1 shows the ranking of US and Canadian metropolitan areas with populations over 1 million in terms of the share of their young adult population. As expected based on the previous research, cities with larger shares of young adults tend to be those with natural amenities, arts and cultural scenes, night life and are experiencing economic growth such as Salt-Lake-City, Austin and Calgary. There may also be a link to the location of post-secondary institutions, as one of the reviewers noted – something that deserves further inquiry in future research.
Percentage of young adults in North American metropolitan areas with populations greater than 1 million.
Source: Calculated using census data: 2010 in US and 2011 in Canada.
Note: The largest three metropolitan areas in US/Canada are shaded.
But note that the cities with the highest living costs (and with a high representation of post-secondary institutions) such as New York, Boston, San Francisco, Toronto and Vancouver are not at the top of the list, at least hinting at the importance of lower housing costs in attracting young adults. However, low housing costs alone are not sufficient. Cities at the bottom of Table 1 are those that have experienced economic decline due to the restructuring of the economy that led to the internationalisation of the production process, and the loss of manufacturing jobs in North American cities such as Pittsburgh, Cleveland and Detroit. Thus, at the inter-urban scale, the presence of young adults appears to be higher, at least from this initial descriptive consideration, in places that are amenity-rich, with high economic prospects and relatively lower housing costs than in the largest global cities.
At the intra-urban scale, the focus of this paper, there is a similar story to be told about the presence of amenities, public transit and a vibrant inner city economy to explain the residential location of young adults, and how this geography varies with the economic structure of a city. For example, cities such as New York, Seattle and Vancouver see highest concentrations of young adults in central areas whereas cities such as Detroit see no such concentration. This is likely due to the absence of a thriving urban post-industrial economy in Detroit, a condition that has been linked to central city re-development and employment growth in the service sector, as in cities such as Toronto, Montreal and Vancouver (TMV) (Hutton, 2008; Ley, 1996; see Figure 1). We also know from previous research that across North American metropolitan areas more generally, young adults are most concentrated in central areas or in high-density areas near public transportation corridors expanding into the suburbs (Moos, 2012, 2014a; Walter-Joseph, 2015). Again, this kind of concentration is generally occurring in cities that have experienced inner city gentrification, have well-developed post-industrial service economies and expansive public transit systems (Moos, 2014a).
Geography of young adults in select North American metropolitan areas.
Factors contributing to youthification
As I have already established elsewhere (Moos, 2012, 2014a), the presence of young adults in central cities and other high-density areas is at least a manifestation of four interrelated factors: (1) preferences for an urban lifestyle, (2) changing demography and household composition, (3) increasing housing costs, and (4) changing economic prospects. It is not plausible here to distinguish exactly the causal relationship between youthification and each of these factors. This is in part because prior research has not examined these in relative terms, and because some of the factors are inherently interconnected (e.g. urban lifestyle and demography) (Van Diepen and Musterd, 2009). Below I summarise the limited research that does indeed explore the role of these four factors in contributing to youthification.
Townshend and Walker (2010) explain the linkages between the emergence of a higher density urban form and societal change in detail, pointing in particular to the delineation of urban space by lifecycle stage and demographic groups as an outcome of urban consumerism and lifestyle (also see Champion, 2001; Chatterton, 1999). Some researchers are also pointing to decreases in driver’s licenses among young adults, and increases in transit use, walking and cycling as evidence of growing urban lifestyle preferences, although economic factors likely also play a role in this (Moos, 2014a). Demographic changes are connected to the changes in lifestyle: Despite a growing share of young adults staying at home with their parents longer (Beaupré et al., 2006), younger adults in general live in smaller households than in the past, have children later or not at all, and live nearer to transit and urban amenities (Moos, 2014a).
Importantly, the presence of young adults in central areas also reflects the new economic reality that land in cities has become more expensive, housing generally less affordable (Moore and Skaburskis, 2004), young people experience higher ‘risks’ and less certainty in the labour market because of the growth of ‘flexible work arrangements’ (Beer et al., 2011; Furlong and Cartmel, 2007), their incomes are lower than in the past (Moos, 2012), and their incomes lower relative to older generations (Boudarbat et al., 2010). The combination of these factors leads to the youthification of central and other higher density areas as young adults can purchase or rent smaller spaces at a lower total cost (than, say, buy a single-family home in the suburbs), and they can trade-off higher per area housing costs for a shorter, less auto-oriented commute and access to lifestyle amenities (Glaeser et al., 2001; Meligrana and Skaburskis, 2005).
Therefore, the process of youthification at the intra-urban scale can be conceptualised, in some ways similar to the process of gentrification, either as a stage model (Kerstein, 1990) or as a process of shifting capital investment patterns (Smith, 1979) – contributing to ‘decline’, ‘transition’ and ‘success’ in terms of the share of young adults over time (see Table 2). Initially, young people move into trendy and still affordable neighbourhoods that have previously experienced decline – as in the gentrification literature, whether or not it is capital or people that initiate the process remains to be debated (Smith, 1979) although likely it is a combination of the two (Skaburskis and Moos, 2008).
The youthification process.
The more ‘entrepreneurial’ cities expending public investments on urban infrastructures and amenities (Harvey, 1989; Peck, 2005) are particularly successful in attracting the young adults to central, amenity-rich areas, but later in the process private capital is key in furthering growth of condominium units that are commonly occupied by young adults (Walks, 2014). These neighbourhoods, and cities, later become more expensive and costly as housing demand increases with growing popularity. At the intra-urban scale, the stage model is a useful conceptualisation of change but has of course its own limits in that it assumes linearity of process (Rose, 2013).
While we see an increase in social status of some central neighbourhoods through gentrification, a wider segment of the young adult population, spanning across the income spectrum, is accommodated in the central city and high-density suburbs in ever-smaller housing spaces and shared accommodations (see Kern, 2010; Moos, 2014a). This occurs because young adults who are struggling with high housing costs shift to renting and/or sharing apartments, while the higher income earners continue to gentrify (Beer et al., 2011; Chatterton, 1999).
This contrasts with working class families displaced by gentrification as they cannot downsize into a studio unit if their rental apartment undergoes a tenure conversion; or an existing older homeowner whose tax burden suddenly exceeds their fixed income – examples that are iconic of gentrifying neighbourhoods (Ley, 1996). Since youthification is at least in part a manifestation of young adults’ declining economic prospects and rising housing costs (Moos, 2012, 2014b), the changes have the potential to impact young adults’ long-term ability to accumulate equity and attain homeownership (The Conference Board of Canada, 2014).
Metropolitan context
TMV experienced some decline associated with the post-Fordist restructuring of the economy in the 1970s; however, never as dramatic as in some of the US metropolitan areas (Bourne and Ley, 1993; Hutton, 2008). Toronto and Montreal have retained a sizable manufacturing sector, and Vancouver’s economy was always more dependent on the resource sector (Bourne et al., 2011). Today, TMV have all generally been successful in supporting the growth of new economy sectors, contributing to gentrification (Meligrana and Skaburskis, 2005). But Montreal has had the most difficult time recovering from the decline in manufacturing jobs (Germain and Rose, 2000).
Industrial decline set the stage for gentrification in TMV beginning as early as the 1960s in some neighbourhoods (Ley, 1996). Initially, artists and the ‘new middle class’ of quaternary-sector workers upgraded existing housing, industrial areas and peripheral spaces. The downtowns have since transitioned to become post-industrial spaces with recreational, leisure, cultural and retail amenities. Especially Vancouver has experienced an almost complete transition to residential uses in some downtown areas – a trend now also visible in parts of Toronto.
Neoliberalisation reduced the supply of social services and housing, although to a lesser extent in Montreal (Germain and Rose, 2000). However, in all three metropolitan areas have labour markets become less secure because of the growth of more flexible work arrangements (Bourne et al., 2011; Walks, 2014). The share of the population experiencing housing affordability concerns has increased while the price of inner city living skyrocketed (Peck et al., 2014; Moore and Skaburskis, 2004). With the increase in prices, subsequent waves of inner city developments have brought about smaller and smaller apartment sizes, arguably to keep the total selling price within reach of the target demographic, which in many cases is young adults (Harris, 2011; Kern, 2010).
Immigrants who traditionally entered through inner city neighbourhoods upon arrival now commonly move directly into suburban areas (Ley and Murphy, 2001). The suburbanisation of immigrants is in part due to the increasing social status of recent immigrants who are accepted into Canada on a points system. However, a large share of immigrants is still found in higher density housing, particularly high-rise social housing of which some is located in the suburbs (Hiebert et al., 2006). Lower income immigrants have been practically priced out of the inner cities. Instead, the downtowns have become increasingly attractive to young adults who, although ethnically diverse, are less likely to be new immigrants. Contrary, older adults have been increasing in concentration in suburban neighbourhoods in recent years (Patterson et al., 2014).
Methods
The analysis has to confront at least three methodological tasks. First, there is a decision to be made about how to define the ‘urban’, since part of the question is whether young adults are moving into urban neighbourhoods. There are many ways the urban, and the suburban, have been defined in the literature (Forsyth, 2012; Patterson et al., 2014). Here, I operationalise the urban using dwelling density. This approach follows Walks (2013) and Moos and Mendez (2015) who argue that urban ways of living, insofar as density is concerned, exist in urban and suburban places. It allows us to see whether there is a youthification of the urban occurring and not assuming a priori that the urban equates solely to centrality (Walks, 2013). Conclusions are the same when population density is used instead of dwelling density.
The second methodological task is to decide how to examine the relationship between density and age, and to assess the relative importance of age vis-à-vis other factors associated with higher density living. This warrants a multivariate framework. Linear regression is used with dwelling density as the dependent variable and age (25–34, 35–44, 45–54, 55–64, 65+), household income, household size (which correlates highly with the presence of children), the share of potential gentrifiers and the share of immigrants as independents. The analysis uses Statistics Canada census tract data from 1981, 2006 and 2011 Census. Owing to data availability, the multivariate analysis only uses 1981 and 2006 data. Census metropolitan area (CMA) boundaries are used to define the TMV metropolitan areas.
All variables were normalised and robust standard errors were used to address issues arising from heteroscedasticity. Spatial regression models were tested but these did not alter conclusions. Linear regressions were thus used to reduce complexity. Gentrifiers were operationalised using Ley’s (1996) indicator of the ‘new middle class’, which includes people working in the quaternary sector of the economy as a measure of social class.
Using these variables directly in the regression produced issues with multicolinearity. Principle component analysis (pca) – a commonly used variable reduction technique – was therefore used to turn the age variables into a smaller number of components (analysis not shown for brevity). The pca produced a new component variable for each of the three metropolitan areas that is an indicator of young age. The pca also produced components indicating the location of older generations, and in Montreal and Vancouver these are in some years correlated with the presence of young adults as well. The regression coefficients are shown in standardised form so that their magnitudes can be compared over time.
Intra-metropolitan geographies of density
While some North American metropolitan areas still have density gradients that declined with distance from the inner city, we have seen the emergence of polycentric cities (Shearmur and Charron, 2004). Density maps show that polycentricity is more pronounced in Vancouver, and to some extent Toronto, as compared with Montreal (figures not shown for brevity). In Vancouver, a provincial agricultural land reserve that acted as a growth boundary, a regional growth management plan that designated secondary centres, and an elevated light-rail transit system that connects these centres, have contributed to the current spatial structure (Harcourt and Cameron, 2007; McGee, 2001; Moos, 2014a). Toronto’s urban development pattern is also shaped by a provincial growth plan that restricts development in a greenbelt and designates growth areas but this plan is more recent than the one in Vancouver (Kipfer and Keil, 2002). In Montreal where growth management has been less prominently featured on political agendas until more recently (Fischler and Wolfe, 2012), densities have remained highest in and around the inner city on the Island of Montreal, which includes both inner suburban and inner city neighbourhoods.
In Toronto and Vancouver, the high densities in the downtown are largely a product of a condominium apartment boom (Harris, 2011; Kern, 2010; Lehrer and Wieditz, 2009; Walks, 2014), which has spread to some suburban municipalities in conjunction with improvements to public transit infrastructure. The condominium boom is in part fuelled by foreign investors and wealthy migrants, particularly in Vancouver (Ley, 2010). Montreal has fewer new high-rise condominium apartments but historically higher inner city densities in the form of walk-up apartments and low-rise apartment buildings (Germain and Rose, 2000). All three metropolitan areas also have a legacy of higher density inner city and suburban apartment buildings built in the 1960s and 1970s as social and rental housing.
Density and age
Table 3 shows the correlation between the percentage of each age group in a census tract and dwelling density. In all three metropolitan areas, there is a very evident increase in the correlation between density and the presence of young adults over time, although in Vancouver the correlation was already higher in 1981 than in the other cities. Changes over time for the 35–44 year olds are also interesting to note. While in all years, correlations for this age group are low (except for 2011 in Vancouver), there was a shift from negative to positive correlations over time. It is an interesting, if just preliminary, indicator of the further extension of the young adult lifestyle to slightly older adults.
Correlation coefficients, age and dwelling density.
Source: Calculated using Statistics Canada census tract data.
For the 45–54 and 55–64 year olds, the correlations also changed but in the reverse direction, indicating the growing association of these age groups with lower density living. This finding is in line with prior research that finds older generations staying and moving into suburban environments, which on average tend to be less dense (Patterson et al., 2014). For the 65+ group, the correlations with density were modest but positive in 1981. By 2011, these changed to very low but negative correlations with density. The correlations are likely low as there are some seniors downsizing while others are remaining in their single-family homes – a trend that would not be detected by analysis of linear correlations alone.
Signs of urban youthification
The results of the regression analysis point to youthification as one of the processes delineating urban space even when we account for other factors that would influence higher density living, such as household size, immigration status, income and status in the ‘new middle class’ (Table 4). Only in Vancouver was the relationship between young age and density statistically significant in 1981. By 2006, the magnitude of the regression coefficients increased and became statistically significant in all three metropolitan areas. For the older adults, the data tell a mixed story. In Toronto, the coefficient lost statistical significance over time. In Montreal, there is a negative association in 2006, pointing to the suburbanisation of the older population. On the contrary, in Vancouver, the older and younger generations are found in higher density areas; and it is the middle-aged, which are the base in these regressions, that are more likely to reside in low-density areas. The Vancouver case points to a youthification and also a generational bifurcation as older and younger people are found in higher density areas.
Linear regression analysis, dwelling density as a function of tract characteristics.
Notes: ***p < 0.0001, **p < 0.01, *p < 0.05.
Source: Calculated using Statistics Canada census data.
Also of note is the increasing magnitude of the R2 value. The variables included in the analysis have become ‘better’ predictors of higher density living, which points to the growing demarcation of high-density areas by demographic variables (Rose and Villeneuve, 2006). Household income is negatively associated with density, perhaps not surprising given that household size is generally smaller in higher density areas, and thus there are fewer earners per household. The association between household size and density is negative, as expected, however this relationship has weakened somewhat over time. It suggests we are seeing somewhat larger households moving into higher density areas.
The presence of a ‘new middle class’ is also associated with higher density living, and this relationship is strongest in Toronto and Montreal. The presence of immigrants remains associated with higher density. However, in Toronto and Vancouver this relationship has weakened as well, likely in part due to the growing number of immigrants moving into lower density suburban areas in these two cities, whereas in Montreal immigrants have remained more centralised in higher density areas (Hiebert et al., 2006).
Overall, immigration status is the strongest predictor (in terms of the magnitude of the standardised coefficient) of higher density living. Even though there is a share of new immigrants moving into single-family homes, clearly many immigrants are housed primarily in the higher density neighbourhoods of Canada’s three largest metropolitan areas (Hiebert et al., 2006). Some of these neighbourhoods are in the inner city but in general are found in the inner suburbs, built during the post-Second World War period, consisting of high-rise and mid-rise apartment towers. There are programmes underway currently in the City of Toronto to deal with the deterioration of these housing developments, many of which are located in lower income neighbourhoods with severe socio-economic issues (ERA Architects et al., 2010). It is no surprise that immigration is an important dimension of the urban social geography but it is striking to see in statistical analysis the relative importance vis-à-vis other factors such as household size and income.
In terms of the importance of age and the existence of youthification, the evidence is clear: (1) Age is a separate dimension of socio-spatial differentiation; and (2) the relative importance of young age in delineating higher density areas has increased over time. Since the higher density areas have become more strongly associated with young age, we can refer to the process that has led to this as one of youthification. While age has been considered in analyses of the socio-spatial nature of cities for a long time (Shearmur and Charron, 2004), this analysis shows an increasing importance of young age as a determinant of density over time holding other factors constant. Youthification is evident from the data but clearly this is occurring alongside other changes that are more important determinates of higher density living, particularly immigration.
Discussion: The generationed city
I have argued here that the context of socio-economic restructuring and demographic changes occurring since the late 1970s worked to increase the share of young adults in urban areas. I operationalised the urban using density, recognising this as only one among various approaches. I argued that young adults are more likely to reside in higher density areas than other age groups, and more likely today than their young adult counterparts in the past. This, I posit, is an outcome of at least four factors – changes in lifestyles, demography, the economy and housing markets. I use data for three specific Canadian cities to demonstrate that young age has become a more important determinant of higher density living over time, calling the process that has led to these changes as youthification. The process is not separate from gentrification but highly dependent on it to set the stage for a broader segment of the population to reside downtown. While Canadian cities are used as examples, the descriptive analysis suggests similar processes unfolding across major US and Canadian metropolitan areas (also see Walter-Joseph, 2015).
Youthification needs to be understood in the context of historic and ongoing urban transformations: Early gentrification was occurring in a context of undervalued inner city land (relative to the suburbs) (Smith, 1979), and a cultural rejection of the suburbs and modern planning (Ley, 1996). Gentrification initially had a very specific housing form that varied by metropolitan area, such as the brownstones, industrial lofts and even single-family dwellings, all of which were in line with the gentrifiers’ aesthetic (Ley, 2003). More recently, in part because of rising prices, gentrification and ‘super gentrifiers’, it has been linked to the corporate development of condominium towers (Kern, 2010; Lees, 2003), and even conversions of less conventional spaces such as churches, bars and schools, which has been described as a process of class distinction as the original gentrified districts have become ‘too main stream’ (Lynch, 2013). The housing supply in the central city, particularly in the form of high-rise condominium apartments, has also dramatically increased (Kern, 2010).
While some young people are certainly complicit in the gentrification process, there is a broader process underway, arguably more aptly labelled youthification because of the increasing association of young adults with higher density areas where prices and rents are already high. While gentrification was originally linked to the upgrading of single-family and other low-rise buildings, youthification is more directly connected with a higher density urban form such as high- and mid-rise apartment dwellings: The growth of condominium apartments has allowed home ownership to increase in a context of increasing affordability concerns (Kern, 2010) but we should not forget that more than half of young adults are renters.
Just as gentrification is a product of context specific socio-economic conditions, youthification arose in the past 20 years along with demographic changes and growing preferences for urban lifestyles, increasing housing costs and changes in the economy that led to growing employment precarity and neoliberalisation (Moos, 2014a). As explored earlier, these factors all worked to produce increasing demand for smaller apartments in urban settings. What Lehrer and Wieditz (2009) and Walks (2014) have called ‘condofication’ is in many ways the materialisation of societal changes impacting housing markets on the supply side (capital moving into cities) whereas youthification is the outcome on the demand side (people moving into cities).
While there is also some evidence of the aging of lower density areas (Patterson et al., 2014), this relationship is not nearly as pronounced as the link between young age and higher density living – hence, the need to conceptualise a process of youthification in addition to thinking about urban space as generationed or gentrified. Nonetheless, it is important to remember that other factors such as immigration are more important explanatory variables than age as far as high-density living is concerned. The purpose of this paper was not to argue that young age, and youthification, are superseding other processes. But rather the goal was to provide evidence that young age has become more important over time in delineating high-density living. The process that brought about these changes is labelled youthification because it results in a higher share of the young adult population living in high-density neighbourhoods than in the past.
A limitation of this research is that the data only demonstrate the relationship between density and age in specific time periods (as opposed to measuring change). But we know from previous research that the percentage of young adults has increased more than the metropolitan average in the highest density areas and the areas of the inner city experiencing re-development (Moos, 2012; Walter-Joseph, 2015). This could come about through an influx of young adults, a loss of older populations or both. Preliminary findings suggest that both processes are occurring. There also appears to be an association between the development of new housing units and the presence of young adults in Vancouver and to a lesser extent in Toronto but not in Montreal. This suggests that youthification is linked to re-development in some but not all cases. These issues are outside the scope of this paper and will be considered in greater detail in forthcoming research.
One conceptual question remaining is whether youthification is a wholly appropriate term to describe the processes that lead to increasing association between young adults and higher density living. The trends observed could just be a momentary phenomenon as society as a whole is urbanising and thus moving toward higher density living (Champion, 2001; Glaeser et al., 2001). However, the empirical evidence suggests that at least in the periods analysed here, the neighbourhoods with high shares of young adults in one period also had high shares of young adults 10 and 20 years earlier (see Moos, 2012, 2014a).
These data point to neighbourhoods that are potentially ‘forever young’, which could be counterproductive in regards to contemporary planning goals aimed at densification (Skaburskis, 2006). The increasing delineation of urban space by age can lead to difficulty in retaining young adults with children in high-density areas, and can also work to dissuade older adults from downsizing into smaller apartments (Bromley et al., 2005). Further research is required to see whether today’s young adults will remain in high-density areas in larger numbers as they age, and in what ways age segregation and ageism may occur as a result of youthification (Vanderbeck and Worth, 2014).
Clearly, there are signs of a process of youthification underway by evidence of an increasing association between young age and high-density living. I invite others to help unravel the complexities of youthification, and its variations and limitations as a concept across time and space, in future research.
Footnotes
Acknowledgements
The author would like to thank the journal s editor and the reviewers, as well as Pierre Filion, David Ley, Tara Vinodrai, Ryan Walker, Robert Walter-Joseph, Mark Williamson, Elvin Wyly and participants at the 2014 ACSP conference in Philadelphia for comments that helped improve and refine this work. Any remaining errors or omissions are my own responsibility.
Funding
This research was funded through research grants obtained from the Social Sciences and Humanities Research Council of Canada, and from the Province of Ontario’s Early Researcher Award s Program.