Resilience has become a ubiquitous concept among both academics and practitioners of urban and regional studies. Yet for all its potential as a framework for examining how communities can protect against and respond to adversity, resilience risks becoming another economic development buzzword if not employed in a meaningful way. This article examines how the concept has been applied to cities and regions, and what approaches researchers are taking to measuring regional resilience.
The Emerging Framework
The term “resilience” was first used in physics and mathematics to describe a material’s ability to regain equilibrium following displacement.1 In the 1970s, C.S. Holling applied the resilience metaphor to ecological systems and their capacity to adapt to adverse conditions, which often entails multiple “new normal” points of stability rather than a return to the previous, single equilibrium.2 The concept has been advanced and expanded by many researchers, who have drawn from both the narrower and broader visions of resilience to model conditions in disciplines ranging from economics and psychology to sociology and urban planning. Because regions are complex systems of overlapping economies and social and political networks, it is unsurprising that resilience has become a key framework for analyzing regional capacity.
Although resilience is widely used as a framework, researchers have criticized its use as “fuzzy” and trendy.3The sheer scale and breadth of research on resilience makes it a rapidly evolving topic. Swanstrom finds, for example, that “the number of references to the term ‘resilience’ as a topic in the Social Science Citation Index…increased by more than 400 percent” from 1997 to 2007.4 As research proliferates, so do definitions. Norris et al. list more than 20 representative definitions of “resilience” — each of which shares features yet is distinct — that focus primarily on community resilience to disasters.5
For resilience to be a helpful metric for community leaders, it needs consistent definitions that maintain the interdisciplinary nature of the concept. As Christopherson et al. explain, “an interdisciplinary discussion helps clarify the assumptions underlying different perspectives on regional change and how to measure it.”6
Perspectives on Resilience Analysis
When the resilience framework is applied to cities and regions, a fundamental issue is the type of stress or disturbance affecting the area. Some stresses take the form of acute shocks, often natural or manmade disasters. In other cases, regions face chronic, long-term strains, such as the decades of declining employment and population afflicting many older American industrial areas. The measures and frames for evaluating resilience vary depending on the type of stress.7 And because the capacities needed to respond to each form of stress can differ, regions may be more resilient to one type of disturbance than another.
The variations in types of stress as well as the array of lenses through which a researcher might consider regional resilience result in studies that employ a spectrum of approaches. Many use a form of equilibrium analysis drawn from the physics and engineering perspective, concentrating on a region’s or community’s capacity to “bounce back” or return to normal. Pendall et al. note that this framework “tends to dominate in the fields of psychology and disaster studies, both of which seek to understand why people, infrastructure and places recover from disturbances or intense stress.”Metrics such as growth in population, income, and economic product and declines in poverty and unemployment rates are often used to measure a region’s return to equilibrium.8
Because of the complexity of regions, which are composed of many interacting governments, economies, and networks, using a single equilibrium as a baseline can sometimes be limiting or unrealistic. The “multiple equilibrium” model of resilience posits that system stress may permanently alter what could be considered “normal” regional conditions and that numerous possible new growth paths emerge following the disturbance. If dominant social and political institutions hinder restructuring and adaptation, a region may become locked in to a suboptimal equilibrium.Still, Pendall et al. believe that “the multiple-equilibrium perspective on regional resilience is arguably an optimistic one,” because it presumes that “reinvention is possible given the right mixture of foresight, hard work, endowment and compromise.”9
At the broader end of resilience studies, some research emphasizes the need for an evolutionary or complex adaptive systems framework that shows how resilience levels continually change as regional systems, and their many subsystems, evolve. For example, because a region’s needs may shift depending on whether it is in a period of growth, stability, or reorganization, a high level of connectedness between key actors may be steadying in one context but stifling in another.10 Using the evolutionary framework, Norris et al. define resilience as “a process linking a set of adaptive capacities to a positive trajectory and adaptation after a disturbance.”11 Models based on this vision of resilience consistently emphasize that resilience is a process rather than an outcome. The evolutionary framework perhaps better reflects the complexity of regions, but it is less amenable to measurement than equilibrium models.
Measures of Regional Resilience in Community and Economic Development
Because of the breadth of frameworks and research fields applying them, a targeted approach that measures resilience will enable a fuller understanding of how regions can better position themselves to withstand shocks and recover more effectively.
Also mentioned in this issue’s lead article, Hill et al.’s “Economic Shocks and Regional Economic Resilience” comprehensively evaluates regional resilience through a combination of quantitative analysis and qualitative case studies. Using employment and gross metropolitan product (GMP) data from 1978 through 2007, the study classifies shocks as either national economic downturns, national industry shocks to key regional industries, or local industry shocks. The researchers’ use of employment and GMP to measure resistance and resilience focuses on a region’s productivity — only one aspect of a region’s health, albeit an important one. In an equilibrium-style model, regions (defined as metropolitan areas) are placed in three categories based on how they respond to shocks: shock-resistant regions avoid significant declines in growth rates, resilient regions return to prior growth rates within four years, and the remaining regions are not resilient. Regions were less likely to be resistant to national economic downturns and national industry shocks than to local industry shocks, and affected regions (those not shock resistant) were less likely to be resilient to national economic downturns than industry shocks.12
To evaluate what factors made some regions more or less shock resistant or resilient, Hill et al. drew from regional economic development literature and tested a wide set of measures, including variables representing industrial diversification, employment by industry, prior growth rates, labor force skills, demographic characteristics, area population distribution, income inequality, state right-to-work status, and region of the country. Among many results, the researchers find that:
- regions with greater industrial diversity are less likely to experience shocks and more likely to be shock resistant,
- regions with a high percentage of employment in manufacturing are more susceptible to shocks but also more resilient in terms of employment due to demand cycles. By contrast, regions with a high concentration of employment in health care and social assistance tend to be more shock resistant but less resilient,
- regions in right-to-work states are less likely to experience downturns in GMP and appear to be more resilient, and
- income inequality increases the likelihood of employment downturns and reduces regional employment resilience but increases regional GMP resilience.13
To complement the quantitative analysis, Hill et al. performed case studies in six regions, each of which had experienced different types of shocks and levels of resilience: Detroit, Michigan; Cleveland, Ohio; Charlotte, North Carolina; Grand Forks, North Dakota; Seattle, Washington; and Hartford, Connecticut. Conclusions about what made regions more or less resilient vary by region, but common themes emerge. In terms of employment, resilience was closely linked not only to national and local industry conditions discussed above but also to “the strategic decisions of individual firms and their leaders, as well as decisions by entrepreneurs in the area.…” Regional shocks tended to prompt new partnerships to promote regional economic growth, but no one interviewed believed that such activities were key to resilience. Finally, the researchers find little evidence that regional policymakers had spent substantial time in taking precautions against shocks and note that some of the regions that would have benefited most from advance planning “may be ones in which regional actors are least equipped to carry it out effectively,” because they do not perceive the need, are unable to develop plans to sufficiently restructure the regional economy, or lack the social organization within the business and government communities.14
“Economic Shocks and Regional Economic Resilience” captures the complexity of regional economic resilience. Other research applies the resilience framework to more specific types of shocks and responses. For example, “Regional Resilience in the Face of Foreclosures,” by Swanstrom et al. examines regional resilience by looking at metropolitan areas’ responses to the foreclosure crisis, both prevention and recovery, using six paired cases based on strength of the local housing market. Focusing more on institutional processes than economic measures, they define resilience as a region’s ability to devise and implement a response, which involves effective governance and organizational relationships. The authors draw primarily from a multiple-equilibrium model in examining the region’s capacity to stabilize neighborhoods and minimize involuntary displacement, even when the region cannot “bounce back” to the status quo.15
Cleveland, the Inland Empire, and Chicago were more resilient to the foreclosure crisis than their paired cities, St. Louis, the East Bay, and Atlanta, respectively. Factors contributing to this greater resilience include higher levels of public attention to the issue, in the forms of press coverage and widely available data, which the authors believe reflected better-organized housing nonprofits and political leadership. Moreover, metropolitan areas “with a history of collaboration between housing nonprofits and the public sector were able to generate more resources to address foreclosures than metropolitan areas that had not established relations of trust over time.” Community Development Block Grant program entitlement status also played a role; entitlement communities typically had more capacity to respond to the crisis than did nonentitlement areas. The most significant finding, as discussed at length in our lead article, was that places where horizontal, cross-sector ties were supported by vertical ties in the form of state and federal policies performed better than places without such vertical connections.16
“Vulnerable People, Precarious Housing, and Regional Resilience,” by Pendall et al. recognizes that a region’s resilience depends in part on the resilience of its residents and their communities.17 A resilient region, the authors explain, is one that can identify and anticipate shocks, avoid them when possible, and mitigate the effects when avoidance is not possible. The study looks at how individual vulnerability is linked to precarious housing conditions and can affect resilience. The authors posit that various traits can be considered “vulnerabilities” that can hinder life opportunities: being a member of a minority race or ethnicity, elderly, a recent immigrant, an adult without a high school diploma, a post-1990 veteran, or a minor; having a disability; and living below the poverty line or in a single-parent household. People with multiple vulnerabilities are less likely to both be resistant to shocks and be resilient — that is, to recover when they experience shocks. Likewise, people in precarious living situations, such as those experiencing high cost burdens, overcrowding, or renter occupancy as well as those living in old buildings, multifamily housing,18 and trailers or mobile homes, are less likely to be resistant or resilient.19
Many conditions likely to cause individual vulnerability correlate with housing conditions that are considered precarious — income levels are most closely linked, but minorities and immigrants are also substantially more likely than others to live in precarious conditions. The authors recommend that regions undertake efforts to address these conditions: “[R]egions that anticipate the many challenges of protecting and improving this housing stock will do much to guard against stresses that will affect our most vulnerable residents and thereby exhibit greater resilience.”20
Measures of resilience often include general productivity measures, such as educational attainment or age of the region’s workforce, but also rely on measures of agglomeration: specifically, the number and type of industries located in a region. For this reason, the agglomeration literature provides useful insight into the resilience of a region. For example, Hollar shows that an industrially strong central city is relatively more important to regional growth than a vibrant economy in the surrounding suburbs.21 More generally, regions that remain internally fragmented and competitive — for example, localities competing for the relocation of jobs within the area — will be more adversely affected by negative shocks than regions that are less internally competitive.
The Resilience Capacity Index
To better tease out differences in local conditions and response, many studies on regional resilience to date have focused on a small set of metropolitan areas. One attempt to more systematically compare resilience across U.S. regions is the Resilience Capacity Index (RCI), a project spearheaded by University at Buffalo Regional Institute senior fellow Kathryn Foster. As mentioned in Growing Toward the Future: Building Capacity for Local Economic Development, the RCI ranked 361 metro areas using 12 indicators in 3 capacity categories: regional economic, sociodemographic, and community connectivity.22 Two other key categories — environment and infrastructure, and governance and leadership — are not included because of the difficulty in obtaining comparable data sets for the former and quantifying the latter.23
Because not all regions face similar shocks in similar timeframes, the RCI frames resilience as a capacity to confront future unknown challenges. It serves as “a generalized index of the kind of factors that have been hypothesized to matter for responding well to a crisis.”24 The RCI standardizes and combines very different types of indicators by reporting all values as z-scores (which show how many standard deviations above or below the average an indicator performs), and then averaging the 12 z-scores to create a composite value. To make higher indicator scores consistently correspond to more resilient outcomes, the RCI inverts some values: hence “out of poverty” and “without disability.”25
The RCI proves somewhat surprising; the five metropolitan areas it ranks as most resilient are Rochester, Minnesota; Bismarck, North Dakota; Twin Cities Metropolitan area; Barnstable Town, Massachusetts; and Dubuque, Iowa. Foster notes that Midwest and Northeast metropolitan areas tend to rank highly because “slower growth regions actually have more capacity to withstand the shock. It’s counter-intuitive, but they tend to be stable. They’re often more affordable. There are higher rates of homeownership and they tend to have greater income equality.”26 The 5 regions with the lowest RCI rankings are in Texas and California, and the other 35 regions with the lowest rankings are also in the South and West. The rankings could vary using different indicators or data from a different time period, suggesting a role for future research in exploring some of these measurement issues. The RCI’s developers stress that, although some regions are better poised to recover from stress than others according to the index, any number of factors might cause an area to under- or overperform.27
The degree to which regions embrace the RCI remains to be seen, but the index points toward a future for resilience studies in which regions can better compare themselves to similar areas and craft policies that draw from the best practices of their peers. The RCI also may be useful as researchers continue to explore what factors enable regions to better respond or withstand stresses to their economies, their communities, and their residents.
As the field of regional resilience research develops, research efforts will continue to confront several critical issues inherent to long-term studies of large, complex systems. Most evident and important, perhaps, is the need to set appropriate timeframes and geographic boundaries. Since slow-burn stresses may take decades to be fully felt, researchers must carefully consider whether a region has had enough time to prove resilient or not — especially since regions can be affected by overlapping combinations of jolts and longer-term challenges.28 Likewise, regions combine numerous political, economic, and social systems at many different levels. As Katz recently noted, for instance, “the Chicago metropolis alone crosses 14 counties in three states and is chopped up into 347 municipalities, 365 school districts, and 137 library districts.”29 The difficulty of defining a region’s borders requiresresearchers to be mindful of what is being omitted.
A close look at the RCI reveals additional challenges faced in measuring regions’ resiliency. The economy, governance, and organizational structure of a small metropolitan area like Barnstable Town is very different from those of a large metropolitan area like Rochester. It seems counterintuitive that a region experiencing strong economic growth, and the res-ources such growth generates, would be less resilient than a slow-growth one. For example, greater affordability, which is used as an indicator of resilience, may reflect a region’s inability to attract in-migration, keeping housing prices low and encouraging homeownership. As research into regional resilience continues, researchers will need to develop a strong theoretical model to address these challenges.
Regional resilience studies also face the challenges common to other social science research. Studies looking across many regions often must rely on national data sources, which can be old or insufficiently detailed because local data may not be comparable.30 Meanwhile, studies that focus on a small number of cases may offer clearer details on the local mechanisms of resilience at the expense of broader applicability.
Regions face numerous challenges of varied types; increasing their resilience may enable them to better withstand or adapt to the shocks and disturbances they will inevitably experience. Research plays an important role in better understanding how regions can increase their resistance and improve their resilience, but such research must be sensitive to selecting the appropriate framework for the situation.
1 Fran H. Norris, Susan P. Stevens, Betty Pfefferbaum, Karen F. Wyche, and Rose L. Pfefferbaum. 2008. “Community Resilience as a Metaphor, Theory, Set of Capacities, and Strategy for Disaster Readiness,” American Journal of Community Psychology 41:1–2, 127.
2 Todd Swanstrom. 2008. “Regional Resilience: A Critical Examination of the Ecological Framework,” 4.
3 Susan Christopherson, Jonathan Michie, and Peter Tyler. 2010. “Regional resilience: theoretical and empirical perspectives,” Cambridge Journal of Regions, Economy and Society, 3:1, 4.
4 Swanstrom, 3.
5 Norris et al., 129. While many researchers examine communities’ resilience to disasters, it is important to note that disasters often bring additional resources to the impacted jurisdictions, which can be a critical component in recovery and future growth.
6 Christopherson et al., 4.
7 Rolf Pendall, Kathryn A. Foster, and Margaret Cowell. 2009. “Resilience and regions: building understanding of the metaphor,” Cambridge Journal of Regions, Economy and Society, 3:1, 10–11.
8 Ibid., 2–3.
9 Pendall et al., 5–6. “Lock-in often is a consequence or manifestation of path dependence…. As one technological or political regime comes to the forefront, human systems of all sorts begin to take shape that reflect and respond to that dominant regime. Soon a complex social, physical, economic and cultural infrastructure develops that makes it seem logical, and perhaps even natural, to continue on the development path of that regime.”
10 Swanstrom, 8–9.
11 Norris et al., 130.
12 Hill et al., 8–10.
13 Ibid., 12–8.
14 Ibid., 62–3, 66.
15 Todd Swanstrom, Karen Chapple, and Dan Immergluck. 2009. “Regional Resilience in the Face of Foreclosures: Evidence from Six Metropolitan Areas,” 3–4.
16 Ibid., 46–8.
17 Rolf Pendall, Brett Theodos, and Kaitlin Franks. 2011. “Vulnerable People, Precarious Housing, and Regional Resilience: An Exploratory Analysis,” MacArthur Foundation Research Network on Building Resilient Regions at the University of California, Berkeley, 3–6.
18 Pendall et al. (6) explain their rationale for including multifamily housing as a separate criterion from renter occupancy as follows: “While much of the vulnerability of multi-family housing is a direct consequence of its rental tenure, combinations of structure type and tenure may also combine in complex ways to condition the vulnerability of units. Rented single-family homes and two- to four-unit multiples may be quite vulnerable to degradation because their landlords lack experience and capital…. Large rental complexes, by contrast, are often professionally managed and command higher rents than small multiples, possibly reducing their precariousness compared with smaller structures during downturns but more likely to experience rent increases during upswings.”
19 Ibid., 3–6.
20 Ibid., 15–6.
21 Michael K. Hollar. 2011. “Central Cities and Suburbs: Economic Rivals or Allies?” Journal of Regional Science 51:2, 231–52.
22 “Sources and Notes.” Building Resilient Regions Network (brr.berkeley.edu/rci/site/sources). Accessed 14 November 2011.
23 Christina Hernandez Sherwood. 2011. “Ranking the ‘resilience’ of hundreds of U.S. cities.” Smart Planet (www.smartplanet.com/blog/pure-genius/ranking-the-8216resilience-of-hundreds-of-us-cities/6778). Accessed 14 November 2011.
25 “Sources and Notes.”
27 For an effort to measure disaster resilience that indexes similar indicators applied to counties in the Southeast, see Susan L. Cutter, Christopher G. Burton, and Christopher T. Emrich. 2010. “Disaster Resilience Indicators for Benchmarking Baseline Conditions,” Journal of Homeland Security and Emergency Management, 7(1), Article 51.
28 Pendall et al. 2009, 10.
29 Bruce Katz. 2011. “Why the U.S. Government Should Embrace Smart Cities.” The Brookings Institution (www.brookings.edu/opinions/2011/0726_cities_katz.aspx). Accessed 14 November 2011.
30 See, for example, Cutter, et al., 17.