#scenescapes

Introduction

Where and how might efforts to revitalize neighbourhoods by integrating or enhancing the arts succeed? Exploring the impacts of art establishments on neighbourhood income is valuable for learning where and how policies and programmes to spur community development, for example, creative placemaking, arts districts, or cultural quarters,1 might be effective in achieving their goals.

Several nationally salient initiatives where artists have led development projects in low-income areas include Project Row Houses in Houston, started in 1993; Theaster Gates’ projects in Chicago since 2012; and the Art + Practice Foundation in South LA, led by Mark Bradford in 2014. Grodach and Silver (2012) assembled international case studies of arts/community efforts. These illustrate distinct mechanisms for using arts to energize disadvantaged neighbourhoods, without displacing current residents.

This is the first US national study of how the impact of arts establishments varies across all US zip codes, divided by income and minority status. In the process we use multiple definitions of arts and disadvantage and combine social science with aesthetic and case study work to assess how generalizable the effects of arts activities are as amenities in revitalizing neighbourhoods. The zip code level analysis departs from the common case-study approach as we seek to observe more generalizable patterns of arts impact that inform debates at the intersection of arts, urban regeneration and equity.

Background: the “arts-drives-growth” question

The arts and prosperity have been linked, at least in the West since the Renaissance, where trade and imports of exotic goods sparked local creativity in clothing, architecture, and painting. Later Balzac wrote that artists needed distinct neighbourhoods to be free from bourgeois lifestyle constraints to paint and write creatively (Harvey, 2003). Jane Jacobs (1961/1992) held that artists and bohemians were core drivers of creativity, and their neighbourhoods within cities drove the new creative economy. Schumpeter (1942) stressed the creative destruction of old ideas as central for economic growth. The New Urbanism added pedestrianism and street life. David Brooks (2000) added money to transform bohemians into bobos. Richard Florida (2002) showed that these same processes worked in factories, corporate offices, universities, Economists such as Edward Glaeser (2000) stressed dense urban areas as concentrating amenities, people and economic growth.

We build on these ideas but extend them to low income minority neighbourhoods. For example, Harlem and Bronzeville, the black centres of New York and Chicago from the 1920s onward, fostered Black enterprises like dress and shoe designers, professionals like dentists and ministers and artists like jazz musicians, painters and novelists. The key black political leaders were Congressional Representatives, city council members, and mayors from black neighbourhoods; many favoured racial segregation to solidify their voting base. Harlem and Bronzeville declined economically after 1933 when Prohibition ended. But young African Americans continue to invent musical types from drill rap to hip hop to house even if the clubs are less concentrated in the twenty-first century and internet downloads and social media rise in salience. Jazz, blues and gospel thrive globally, and leading artists, much less constrained by racial discrimination, travel continually even if they retain homes in Harlem and Bronzeville. Harlem supports major bus tours of international tourists today. Meanwhile strong Hispanic areas of Los Angeles, Miami, and Chicago feature murals, Day of the Dead and Cinco de Mayo festivals, and parades. Wherry (2011) details how these artistic activities transformed the Philadelphia Barrio from a slum into an arts-driven tourist centre with guitar strummers on tour buses and more. Chicago’s prosperity in the twenty-first century, relative to most old Midwestern cities, is arguably driven by four months of music festivals and McCormick Place tourism, which continue the art-drives-income tradition of neighbourhood clubs from the Al Capone years (Clark, Lloyd, Wong, & Jain, 2002; Spirou & Judd, 2016). Hunter, Pattillo, Robinson, and Taylor (2016) explore place making via specific, newish arts activities.

Comparative modelling

About a dozen studies have explored these issues comparatively, mostly using cities and neighbourhoods in the US, finding that the arts grow where people concentrate – measured by population size, growth rate, or density (Grodach, Currid-Halkett, Foster, & Murdoch,, 2014; Kushner, 2013; Murdoch, Grodach, & Foster, 2016; Patterson & Silver, 2015; Schuetz, 2014). Only one study examines arts growth specifically in disadvantaged neighbourhoods in NYC: Murdoch et al. (2016) report that organizations locating into such neighbourhoods are the exception not the rule, and tend to be younger organizations, to target local audiences, have smaller budgets, and rely on part-time volunteers.

Where the arts grow, findings suggest that in the US they generally improve housing values (Grodach, Foster, & Murdoch, 2014; Noonan, 2013; Stern & Seifert, 2010; Woronkowicz, 2016) and income (Grodach, Foster, et al., 2014; Noonan, 2013; Schuetz, 2014; Woronkowicz, 2016) in urban and nationwide contexts. In Canada, however, Silver and Miller (2013) find that arts relations to income depend both on the type of arts that grow and type and strength of the cultural scene. Grodach, Foster, et al. (2014) similarly find that in the US the type of arts that grow affects the type of neighbourhood change. The “scenes” project’s other studies2 find generally positive associations between local arts activities and population, income, and job growth in China, Korea, France, Spain, Canada, and the US (below and Clark et al., 2014).

Method

We examine the impact of arts establishments on income across the entire US and among disadvantaged neighbourhoods. Our empirical analysis employs linear regressions predicting median household income in 2008–12 (American Community Survey 5-year estimate) at the zip code level for the entire US. Over 20,000 usable postal codes (Census zip code tabulation areas or ZCTAs) have relatively stable boundaries. These boundaries are not coterminous with the multiple meanings of “neighbourhood” or “community”, but they provide a far more nuanced analysis than national, metro, county- or city-level data. The large numbers are far better for multi-causal analysis than most past arts studies.

Estimating arts impact generally confronts concerns about endogeneity (e.g. Noonan, 2013), especially if the growth of arts (as a luxury) follows economic prosperity, and even more so if policymakers and planners target areas of rising affluence for arts growth. Our analysis mitigates endogeneity concerns by not pooling all neighbourhoods together (which could generate results from wealthy neighbourhoods driving arts growth), but instead examines the relationship of arts and income varying within and across subtypes of neighbourhoods.

We have elsewhere explored many other variables and models specifying, for example, relative feedback effects of arts on income and income on arts activities, summarized in Silver and Clark (2016).

Operationalizing arts activity

Our key explanatory factor is arts activity, measured as the number of arts establishments from the US Census’ Business Patterns (“bizzip”) at the zip code level in 2001. Measuring for establishments rather than jobs more effectively captures visible arts activity and opportunities for conspicuous consumption. Our “narrow” working definition of arts activities includes entities directly producing and distributing the arts, and includes a simple count of art dealers; museums; fine arts schools; theatre companies and dinner theatres; promoters of performing arts, sports, and similar events; dance companies; musical groups and artists; other performing arts companies; and independent artists, writers, and performers. This fits most discussions of the arts. Our “wider” definition captures the production and consumption of the arts via broad networks of direct and indirect participants (Becker, 2008). We create a broad measure of 37 North American Industry Classification System (NAICS) codes. This includes the narrow definition and adds others such as musical instrument and supplies stores, historical sites, and amusement parks. Grodach, Currid-Halkett, et al. (2014) summarize other broad measures.

Controls

Controls include factors that past research (e.g. Glaeser, 2008; Silver, Clark, & Graziul, 2011) suggests shape income or innovation: population (density in 1990), racial composition (the proportion of non-White residents in 1990), general policy environment (county-level proportion of votes for the Democratic presidential candidate), and cost of living (county-level mean median gross rent rate in 1990). We also include a measure for urbanity using bizzip data, measured in 2001 as the earliest year available. Other control variables analysed but dropped in results shown here due to multicollinearity include proportion below poverty, with a bachelor’s degree, married, and unemployed. We add controls for 1990 as initial conditions relevant to arts activity: proportion living in the same house for five or more years, to see if more established neighbourhoods with more character matter; proportion of households with children aged 0–17 hypothesizing that young families have less time for the arts; and the average commute to work time, expecting lower arts participation with longer commuting.

Operationalizing neighbourhood scenes

We also examine how neighbourhood scenes mediate relationships between the arts and income. To summarize, a “scene” refers to the atmosphere or cultural life of a place. We take it to include less tangible activities and practices, but amenities provide a window into the type and range of experiences available.

Bohemia implies an unconventional lifestyle and can be at play in neighbourhood vitalization efforts with the arts. Bohemia’s role may differ in wealthy and poor neighbourhoods. Understanding how bohemia shapes the relationship between arts activity and income may therefore provide clues for arts activity among disadvantaged neighbourhoods. Our Bohemian Scene index follows Silver and Clark Scenescapes (2016, p. 341). It measures how closely a zip code resembles an ideal-type bohemian scene, defined using classical writings on Bohemia including Benjamin (2002) and Wilson (2000). Bohemian theorists imply that more bohemianism should generate more innovation and thus income.

Tests have been few, so we sought to go further.

Artists, bohemia and scenes have been broadly discussed as overlapping concepts for decades. The Scenes Project contribution is not to ignore, but to systematize these three past artistic terms to help them become social science concepts and methods. We, thus, developed a list of 15 distinct scenes dimensions by codifying major related efforts from past work, including Hegel, Wagner, Max Weber, Levi-Strauss, Inglehart and Welzel (2005), and related survey research on basic value dimension like the World Values Survey, the General Social Surveys and International Social Survey Programs (detailed in Silver & Clark, 2016). Figure 1 outlines the 15 dimensions.

Figure 1. Scenes 15 dimensions. Source: Silver and Clark (2016).

To measure the 15 we used 143 individual industrial categories to characterize each zip code in the US. The 143 consumption-related amenities are from electronic bizzip data by NAICS codes. Each of the 143 is scored 1–5 for each of the 15 scene dimensions, using a handbook for coders defining each dimension. We computed reliability measures among coders, sharpening our Coder’s Manual of criteria, and applying the method to143 bizzip and over 300 Yellow Page amenities types. The amenities data were used to generate scenes performance dimensions as shown in Figure 2.

Figure 2. Scenes performance score construction example.

The performance scores combine scores assigned to each amenity with data on the number of each type of amenity located in a zip code. Suppose Zip Code #1 has five total amenities: four body piercing studios and one Catholic church. Suppose also that body piercing studios were scored 5 on transgressive theatricality while Catholic churches were scored 1. Multiply the number of each type of amenity (4 body piercing studios, 1 Catholic church) by that type’s transgression score (5 and 1). Sum the product and you get 21. Now divide that total output by the total number of amenities in the zip code (in this case, 5). The result of that division, 4.2, is Zip Code #1’s transgression performance score. A different zip code, say, Zip Code #2, with four Catholic Churches and one body piercing studio, would thus have a transgression performance score of 1.8.

The same procedure was repeated for each zip code, generating a score on each of the 15 scenes dimensions for each zip code area.

For this paper we created an ideal bohemian pattern, defined using classical writings on bohemia including Benjamin (2002) and Wilson (2000) in terms of our 15 dimensions, shown in Figure 3. Then we subtracted the score of each individual zip code from this bohemian ideal and took the absolute value of the difference; we reversed the sign so that a high value indicated a more bohemian zip code. This distance from a bliss point is widely used in public choice analyses of political party loyalty of individual citizens. This bohemian index was used in the regressions, showing the interesting negative relationship with income.

Figure 3. Bohemian ideal “bliss point” scores on the 15 scenes dimensions.

As a result, a zip code is scored as more bohemian if it has more amenities included in the 15 dimensions with positive weights in transgression (breaking conventional style), charisma (promoting extraordinary qualities and accomplishments), ethnic (undiluted by homogenizing, deracinated, abstract global monoculture) and self-expression (actualizing individual personality); and fewer amenities with negative weights in rational (emphasizing intellect, exercise of reason), corporate (defined by mega-corporations), state (defined by the nation-state), neighbourly (personal networks, face-to-face intimacy), egalitarian (human equality), utilitarian (instrumentalizing a situation with respect to profit), and traditional (connecting with the past and a historical narrative). The remaining dimensions are weighted neutral in the case of an ideal-type bohemia – glamorous, formal, exhibitionistic, and local.

The Bohemian Scene index is thus much broader than any index to date, such as Florida’s (2002) bohemian index which simply counted and summed census data category jobs like artists, writers, and performers – thus assuming that artists are bohemian. Because artists include (possibly) non-bohemian web designers, advertising staff, and amateur watercolour painters, we measure artists and bohemia separately. Our reanalysis of Florida’s data for gay and bohemian indexes as tolerance indicators and job drivers are in Clark (2004). Our Bohemia index correlates significantly (Pearson r = .16) with arts activities in 2001, illustrating the importance of not assuming the two are identical. The mean Bohemia score is .064, ranging from .046 to .091, (standard deviation .002, N 35,675). In Chicago, for example, Bohemia raw scores in 2001 include Bucktown (.065), Wicker Park (.065), Humboldt Park (.064), and Logan Square (.065), all then commonly perceived as lead bohemian/hip neighbourhoods (Lloyd, 2010; Redmond, 2008), despite later changes.

The scenes’ scores provide continuous measures for all zips; we do not select just a subset of high-scoring neighbourhoods but retain all. Of the 143 amenities included, tattoo parlours, nightclubs, and liquor stores were examples of NAICS industry codes scored 5 (high) on transgression (as a behavioural not a legal concept). Including this Bohemia Scene index in our national regression analysis assesses the impact of bohemian local scenes on income (distinct from the arts and control variables). This shows how important a bohemian ethos is rather than assuming that artists are all equally bohemian. The results show how this matters.

Selecting disadvantaged neighbourhoods for analysis

We conduct separate regressions in two national contexts. First is the national context, of all US zip codes for which we have data on all of the variables in each regression model. Second, we repeat the same models within each quantile of “disadvantage”. For this, we create a zip code-level disadvantage score using only median household income in 1990. We add two alternative composite disadvantage scores: one combining income and proportion of non-Hispanic African Americans, and another combining income and proportion of Hispanics (both in 1990). We rescale income and reverse the race or ethnicity measure so both variables have a minimum of 0 and a maximum of 1. High indicates low-income and a high proportion of Blacks or Hispanics. All three scores are normally distributed.

We divide all zip codes into quartiles of disadvantage. We re-estimate the regression initially four times, for each subsample. To assess robustness, we repeated using quintiles, sextiles, and septiles (Table 1).

Table 1. Descriptive statistics by income-only disadvantage quartile.

Results and discussion

We report detailed results for one illustrative set of models, then summarize main findings of others.

Table 2 shows ordinary least squares regressions of zip code income on an arts index and a variety of control variables. Control variables (in the Method section) are omitted from the tables here, but are available upon request from the authors.

Table 2. Narrow arts index, national and within Quantiles of income-based disadvantage, Model 1 and 2 results.

In Model 1 regressions, zips with more arts establishments show higher income for all US zips combined, and in three of the four subsets of neighbourhoods. The strongest effects are for the least-disadvantaged quartile, but second-strongest is consistently the most disadvantaged. Results for the wide arts index (not shown) are similar. Again, income rises with the arts index especially in the least disadvantaged followed by the most disadvantaged neighbourhoods.

Model 2 adds the strength of the bohemian scene to the model, thus measuring both arts and bohemian effects in a single model with controls. The results are dramatic. By separating arts activities from Bohemia, we find the opposite of the Jane Jacobs/Florida creativity hypothesis. More bohemian zips suppress income, controlling other income drivers in our model – the opposite of the positive arts-income effect. These contradictory coefficients provide a new perspective on these two opposing effects which are combined in many historical accounts and case studies like Jacobs’ Greenwich Village, or Florida’s national (mostly metro) rankings (presented generally without multi-causal analysis). Still, remember the feedback loop: some bohemians move to lower-income zips.

Table 3 adds minorities to income to create further measures of disadvantage. The main results are unchanged using alternative disadvantage definitions. The differences are difficult to interpret as they may be driven by subgroups within each quartile acting in ways better studied with models more targeted on such distinct patterns.

Table 3. Adding minorities to Table 2 shows similar effects.

Though the analysis here considers over 20,000 zip codes, some small zip codes are excluded as the data are not disclosed by the US Census due to confidentiality concerns.

Bohemian scenes

One of the most dramatic findings is that bohemian effects do not just reinforce arts effects. They are generally opposed, in our data and time period. More bohemian scenes have less income, with the exception of the most disadvantaged neighbourhoods using an income-only disadvantage definition in Tables 2 and 3. Comparing impacts on income, the Narrow Arts Index explains 82% and the Bohemian Scene 17% of the effects generated by considering just the sum of these two variables in Table 2 Model 2 column 2. Measuring the arts and bohemia as two opposed effects should encourage others to look for potentially disparate factors driving these processes despite past historical accounts. More important, as we add more subgroups in terms of income, Bohemia, African-American, Hispanic and more, the patterns are often stronger and clearer than with the simpler bigger categories. This diversity illustrates the importance of context and multiple causal pathways. For instance, ironic hipster arts activities may appeal in more bohemian neighbourhoods, while gospel-inspired music is more in harmony with activist churches in other equally disadvantaged areas. This is generally consistent with Silver and Miller (2013) who find that the type and strength of a particular scene can weaken or strengthen the relationship between arts activity and income. How might bohemianism suppress income? Consider the case detailed by Lloyd (2010) of aspiring artists, many of whom working as bartenders in a Bohemian Chicago neighbourhood. They went to other bars on their days off and gave away much of their incomes to other bartenders as generous tips. Drinking undermined their arts work too. How census-defined disadvantage is locally ignored or proudly celebrated hugely matters. Stuart (in progress) shows that gang members make tough drill rap videos, whose YouTube ratings are their new bottom lines. Bohemian scenes can aid or inhibit leveraging buzz, depending on how these are combined. These examples illustrate patterns that demand subtlety to clarify. Our new findings of significant income effects, positive for the arts, negative for bohemia, should not be overgeneralized but spur more sensitive work that explicitly joins aesthetic style with socio-economic and ethnic factors.

Number of arts establishments

The more arts establishments in a zip code, the higher the income. On average, a 10% increase in a neighbourhood’s arts index is associated with a $2,111 increase in median household income. This positive relationship holds across wide and narrow arts types and of disadvantaged neighbourhoods but varies in magnitude. This result enhances past studies (Noonan, 2013; Schuetz, 2014; Stern & Seifert, 2010; Woronkowicz, 2016) by adding many controls, larger Ns, and explicit contrasts of more and less advantaged neighbourhoods.

Type and range of arts establishments

Different types of arts vary in their relationship to neighbourhood income. Differences shift with the type of consumer and number and types of staffing, material, and infrastructure. Our estimates suggest that the narrow art establishments measure is slightly more predictive of higher median household income than the wide measure, consistent with past studies considering multiple arts definitions (Grodach, Foster, et al., 2014; Kushner, 2013; Murdoch et al., 2016; Silver & Miller, 2013).

Level of disadvantage

Table 2 results show a weaker relationship between the arts and income in moderately disadvantaged neighbourhoods, relative to the most and least disadvantaged. The relative middle-class homogeneity has attracted less research and policy intervention than for the highest and lowest income groups. As groups like the National Endowment for the Arts add more types of art in more recent surveys (like knitting), specifics become more visible.

Conclusions and implications

These results show that the arts are positively linked to income in some 25,000 odd US zip codes within four to seven distinct income and ethnic groups. These patterns shift by scene context, illustrated by bohemianism. The most striking contrast with past work is how separate bohemianism is from the arts, specifically that bohemianism suppresses income.

While local scenes shift impacts, a striking result is that most neighbourhoods with more arts activity have more income. This holds within the wealthiest and the most disadvantaged of neighbourhoods. These results challenge the view that the arts simply follow, not drive, wealth, and suggest that the arts can add value (e.g. by generating buzz via better texts, posters, websites and more) and effectively foster neighbourhood revitalization.

Nevertheless, even if the arts help income in all sorts of neighbourhoods, there is no one-size-fits-all arts strategy for effective neighbourhood revitalization. Key to success is sensitivity to the local context by arts activists and policymakers, as illustrated in the diversity of local arts, lifestyle, and social background connections detailed in Silver, Lee, and Childress (2016) and Brown-Saracino (2018).

From a policy perspective, the largest US national arts programme is Our Town, supported by the National Endowment for the Arts. Unlike national programmes in more centralized countries like China and France, each of several hundred Our Town programmes is jointly created and implemented by local artists, civic groups, and a local government. The increasing global recognition that the arts are critical foundations for education, aesthetics, and creative neighbourhoods should encourage more detailed inquiries. We need to join the case studies of specifics with the larger comparative analyses to inform future local projects as well as national arts and culture policies around the world. To better understand context and thus improve the likelihood of success and equity, decision makers and planners can use the two approaches employed in the present study – analysis of distinct scenes and income groups – to better inform strategy and policy.

Notes

See National Assembly of State Arts Agencies (2015) for a brief on state level policies; the US National Endowment for the Arts’ “Our Town” grant programme, https://www.arts.gov/grants-organizations/our-town/grant-program-description; the EU “Capitals of Culture” initiative, https://ec.europa.eu/programmes/creative-europe/actions/capitals-culture_en; ArtPlace America, https://www.artplaceamerica.org; and Artspace, https://www.artspace.org.

See https://scenescapes.weebly.com.

References

Becker, H. S. (2008). Art Worlds (revised edition). Berkeley: University of California Press.

Benjamin, W. (2002). The Arcades Project. (R. Tiedemann, Ed.). New York: Belknap Press.

Brooks, D. (2000). Bobos in Paradise: The New Upper Class and How They Got There. New York: Simon and Schuster.

Brown-Saracino, J. (2018). How Places Make Us: Novel LBQ Identities in Four Small Cities. Chicago, IL: University of Chicago Press.

Clark, T. N. (2004). Gays and urban development: How are they linked? In T. N. Clark (Ed.), The City as an Entertainment Machine (pp. 179–192). Lanham, MD: Lexington Books, 2011 paperback; originally Amsterdam: Elsevier/JAI, 2004.

Clark, T. N. (Ed.). (2014). Can Tocqueville Karaoke? Global Contrasts of Citizen Participation, the Arts, and Development. Bingley, UK: Emerald.

Clark, T. N., Lloyd, R., Wong, K. K., & Jain, P. (2002). Amenities drive urban growth. Journal of Urban Affairs, 24(5), 493–515.

Florida, R. (2002). Bohemia and economic geography. Journal of Economic Geography, 2(1), 55–71.

Glaeser, E. (2000). The future of urban research: Non-market interactions. Brookings-Wharton Papers on Urban Affairs, 2000, 101–138.

Glaeser, E. (2008). Cities, Agglomeration and Spatial Equilibrium. Oxford: Oxford UP.

Grodach, C., Currid-Halkett, E., Foster, N., & Murdoch, J., III. (2014). The location patterns of artistic clusters: A metro- and neighborhood-level analysis. Urban Studies, 51(13), 2822–2843.

Grodach, C., Foster, N., & Murdoch, J. (2014). Gentrification and the artistic Dividend: The role of the arts in Neighborhood change. Journal of the American Planning Association, 80(1), 21–35.

Grodach, C., & Silver, D. (Eds.). (2012). The Politics of Urban Cultural Policy: Global Perspectives. Routledge.

Harvey, D. (2003). Paris, Capital of Modernity. New York: Routledge.

Hunter, M. A., Pattillo, M., Robinson, A. F., & Taylor, K.-Y. (2016). Black placemaking: Celebration, play, and poetry. Theory, Culture & Society, 33(7–8), 31–56.

Inglehart, R., & Welzel, C. (2005). Modernization, Cultural Change, and Democracy. New York: Cambridge University Press.

Jacobs, J. (1961/1992). The Death and Life of Great American Cities. New York: Random House.

Kushner, R. (2013). Cultural enterprise formation and cultural participation in America’s counties. In M. Rushton (Ed.), Creative Communities: Art Works in Economic Development (pp. 144–165). Washington, DC: Brookings Institution Press.

Lloyd, R. (2010). Neo-bohemia: Art and Commerce in the Postindustrial City. New York: Routledge.

Murdoch, J., Grodach, C., & Foster, N. (2016). The importance of neighborhood context in arts-led development: Community anchor or creative class magnet? Journal of Planning Education and Research, 36(1), 32–48.

National Assembly of State Arts Agencies. (2015). Cultural district policy brief. Retrieved from http://www.nasaa-arts.org/Research/Key-Topics/Creative-Economic-Development/StateCulturalDistrictsPolicyBrief.pdf

Noonan, D. S. (2013). How US cultural districts reshape neighbourhoods. Cultural Trends, 22(3–4), 203–212.

Patterson, M., & Silver, D. (2015). The place of art: Local area characteristics and arts growth in Canada, 2001–2011. Poetics, 51, 69–87. Redmond, S. (2008, January 10). Bohemian rhapsody. New City, 4–5.

Schuetz, J. (2014). Do art galleries stimulate redevelopment? Journal of Urban Economics, 83, 59–72.

Schumpeter, J. A. (1942). Capitalism, Socialism, and Democracy. New York: Harper.

Silver, D. A., & Clark, T. N. (2016). Scenescapes: How Qualities of Place Shape Social Life. Chicago, IL: University of Chicago Press.

Silver, D., Clark, T., & Graziul, C. (2011). Scenes, innovation and urban development. In D. E. Andersson, E. Andersson, & C. Mellander (Eds.), Handbook of Creative Cities (pp. 229–258). Cheltanham, UK: Edward Elgar Publishing.

Silver, D., Lee, M., & Childress, C. C. (2016). Genre complexes in popular music. PloS one, 11(5), e0155471.

Silver, D., & Miller, D. (2013). Contextualizing the artistic dividend. Journal of Urban Affairs, 35(5), 591– 606.

Spirou, C., & Judd, D. (2016). Building the City of Spectacle. Ithaca: Cornell UP.

Stern, M. J., & Seifert, S. C. (2010). Cultural clusters: The implications of cultural assets agglomeration for neighborhood revitalization. Journal of Planning Education and Research, 29(3), 262–279.

Stuart, F. (in progress). The battle of the bullet. (Draft MS. Sociology). University of Chicago.

Wherry, F. F. (2011). The Philadelphia Barrio: The Arts, Branding, and Neighborhood Transformation. Chicago, IL: University of Chicago Press.

Wilson, E. (2000). Bohemians. New Brunswick, NJ: Rutgers University Press.