Rethinking the Theory and Practice of Local
Richard C. Schraggerâ€
Scholars of urban law and policy tend to assume that local officials can exert
some influence over city well-being. More specifically, the literature assumes that government policiesâ€”either at the federal, state, or local levelâ€”can influence local economic growth and decline, the most important determinants of a cityâ€™s health. This connection between policy and local economic outcomes implies a theory of how cities form
and grow, however, and legal scholars have not adequately articulated such a theory.
This Article argues that we need a better (and more self-conscious) account of city formation and local economic growth. The Article then tests our intuitions about the relationship between policy and economic development by considering a number of explanations for why cities have resurged over the last fifteen to twenty years. Finally, the
Article contrasts two economic development policies that have been adopted in New
York Cityâ€”one that preceded the recent financial crisis and one that followed it. The
Article concludes that we do not know enough to be able to predict how one policy or
another will affect city growth and decline.
Both the law and economics and participatory literature on local
government law share the assumption that policymakers can exert
some influence over city well-being. More specifically, the literature
tends to assume that cities can influence local economic growth and
decline, the most important determinants of a cityâ€™s health. The law
and economics literature most forcefully adopts this assumption, for it
generally holds that competition among localities will induce them to
enact policies that will attract mobile residents and firms.1
â€ Professor of Law, Class of 1948 Professor in Scholarly Research in Law, University of
Virginia School of Law.
Many thanks to Risa Goluboff for extensive comments on this draft and to the organizers of
(and participants in) the Symposium, Reassessing the State and Local Government Toolkit at The
University of Chicago Law School.
1 At the heart of this literature is an emphasis on Tieboutian sorting. See, for example,
Vicki Been, â€œExitâ€ as a Constraint on Land Use Exactions: Rethinking the Unconstitutional Conditions Doctrine, 91 Colum L Rev 473, 511â€“17 (1991) (using Tieboutâ€™s theory to argue that local
governments will be constrained in their ability to engage in predatory land use practices). See
generally Paul Peterson, City Limits (Chicago 1981) (analyzing urban public policy as a product
of the cityâ€™s limited ability to constrain the flow of resources into and out of the jurisdiction);
William Fischel, The Homevoter Hypothesis 58â€“61 (Harvard 2001) (arguing that mobile homebuyers are knowledgeable about the amenities available in different locations and can â€œshop for
a communityâ€ that best fits their preferences).
312 The University of Chicago Law Review [77:311
petition would be relatively useless if local government policy had
little effect on local economic outcomes. By the same token, participationist scholars have claimed that given enough and the right kind of
power, cities can do a great deal to improve the lives of their citizens.2
Implicit in this argument is that law is significantly to blame for poor
urban outcomes. Fixing law would go a long way, on this account, to
generating prosperous cities.
What is often missing in these accounts is a theory of how urban
economic development happens. Without such a theory, we are left to
assume that generally good policies will induce growth in the economy and generally bad policies will induce decline. But this is an assumption. If cities do not have much control over their economies,
then neither competition induced by exit (on the law and economics
account) nor innovation induced by voice (on the participatory account) is particularly relevant to urban outcomes.
This Article claims that local government law scholarship often
rests on a set of unstated assumptions about how cities form and how
local economic growth and decline occurs.3
Part I describes these assumptions, observing that the law and economics scholarship in particular adopts a relatively incomplete notion of how cities form and
Tieboutian public choice scholarship is not alone, however.
The literature on local government law generally assumes that economic development follows, at least in part, from government policy,
whether scholars are advocating increased decentralization, regionalism, or some other set of policy prescriptions. The wider fiscal federalism literature also shares this assumption.5
One way to test our intuitions about the relationship between local policies and economic development is to consider the urban resurgence of the last fifteen to twenty years. Part II reviews a number of
reasons that have been offered to explain the renewed vitality of ci-
2 See, for example, Gerald E. Frug and David J. Barron, City Bound: How States Stifle
Urban Innovation 231â€“33 (Cornell 2009).
3 In two prior articles, I touch on the question of how cities form, grow, and decline, but
only indirectly. See Richard C. Schragger, Mobile Capital, Local Economic Regulation, and the
Democratic City, 123 Harv L Rev 482, 520â€“26 (2009) (observing that local economic growth may
not be undermined by local redistributionist policies); Richard Schragger, Cities, Economic Development, and the Free Trade Constitution, 94 Va L Rev 1091, 1100â€“07 (2008) (noting that the
diversity and specialization of industry in cities plays an important role in driving growth).
4 See David Schleicher, The City as a Law and Economic Subject, 2010 U Ill L Rev *3â€“6
(forthcoming), online at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1471555 (visited
Jan 4, 2009). See also Schragger, 123 Harv L Rev at 492â€“97 (cited in note 3).
5 See, for example, Barry Weingast, The Economic Role of Political Institutions: MarketPreserving Federalism and Economic Development, 11 J L, Econ, & Org 1, 24â€“25 (1995).
2010] Theory and Practice of Local Economic Development 313
ties. Scholars and policymakers have offered numerous policy prescriptions based on these reasons, all of which are aimed at cities seeking to ride the wave of urban popularity. Those policies, however, seem
somewhat beside the point when considered in light of large-scale
economic trends that affect city prosperity. Indeed, the recent financial
crisis might make us hesitant to attribute a strong link between local
economic policy and economic outcomes. City and regional policies
might matter in terms of urban outcomes, but we should have a
healthy skepticism about how they matter.
This is important, for local economic development policy assumes
a theory of local economic growth. In Part III, I compare two economic development policies that have been adopted in New York City. The
first policy involves the granting of location subsidies to large-scale
business. This approach is relatively unremarkable. Attracting and retaining significant employers by offering them subsidies has been a
central tool of local government economic development offices for
some time. The second policy is somewhat different. It still entails a
distribution of public monies to private enterprise, but, in this case, the
city retrains laid-off workers and subsidizes new startup ventures. This
policy represents a shift away from attraction and retention to an emphasis on small-scale, homegrown innovation.
Both the attraction and retention policy and the homegrown innovation policy provide a window into what local industrial policy
looks like in a post-industrial era; the efficacy of either approach turns
on whether oneâ€™s theory of economic development is correct. Unfortunately, we do not have a consensus account of how cities form, develop, and grow or decline. Thus, any claim that one policy or another
will generate local economic growth should be made with a great deal
I. IS THE CITY A BYPRODUCT, A PRODUCT, OR A PROCESS?
To what extent we believe that local policy can affect local economic outcomes depends in part on how we describe what a city is
and our account of how it grows and declines. Some stories of city development leave little room for policy. The conventional historical
story of the economic rise and fall of the great industrial cities, for
example, seems mostly devoid of local agency. These stories often begin with agricultural surpluses.6
In explaining the great industrial cities,
6 But see Jane Jacobs, The Economy of Cities ch 1 (Random House 1970) (arguing that
cities were a necessary precondition for increases in agricultural productivity).
314 The University of Chicago Law Review [77:311
they start with trading and transportation nodes. Cities came into existence at the confluence of goods and services, the terminuses of ports
or railheads, or along stagecoach lines, where raw materials could be
transformed into goods to be shipped, or provided to a large local
market. There, too, the legal, financial, and governance expertise necessary to coordinate these large entities and provide them with capital developed. With the advent of the automobile, the subsequent extraordinary decline in transport costs, and the lessening importance of
coal-fired plants, the industrial city lost its technological reason for
On this account, government policy has comparatively little to
do with urban economic growth and decline; cities are essentially byproducts of distance-changing technologies.
This relatively deterministic narrative sits somewhat uneasily alongside a different conception of what the city is and how it grows and declines: the notion that cities are a product that one can understand in
terms of the supply and demand for urban space (and for particular urban spaces).8
On this account, individuals and firms choose city space
(versus other kinds of space) and choose among cities based on various
costs and benefitsâ€”say, the costs of transport and congestion balanced
against local amenities, or balanced against the gains of propinquity to
customers, other firms, labor markets, and raw materials.9
The supply and demand story does not have to be at odds with
the technological storyâ€”in fact, those stories are often told in tandem.
But sometimes the supply and demand story implies that cities control
the factors that make them desirable in the spatial supply and demand
market. It is not a difficult conceptual leap from positing a market in
city space to the idea that cities â€œcompeteâ€ with one another to attract
persons, goods, and capital. Indeed, the rhetoric of â€œcompetitionâ€ follows somewhat easily.
That rhetoric has been most fully embraced by law and economics scholars who start with Charles Tieboutâ€™s construct of a market in
local government.10 Tieboutian governments compete to provide services to residents. This competition must be for something; if the demand for city space (and for a particular cityâ€™s space) was outside the
control of any particular city, then the language of competition would
7 See Douglas Rae, City: Urbanism and Its End 215â€“16 (Yale 2003).
8 See Edward Glaeser, The Death and Life of Cities, in Robert Inman, ed, Making Cities
Work: Prospects and Policies for Urban America 22, 25â€“26 (Princeton 2009).
9 See id at 26â€“27.
10 See generally Charles M. Tiebout, A Pure Theory of Local Expenditures, 64 J Polit Econ
2010] Theory and Practice of Local Economic Development 315
make little sense. Competition implies some level of agency, an ability
to improve the product being offered. So we arrive at the notion that
cities compete to provide amenities to firms and residents, that this
competition disciplines local governments so that they provide efficient public services, and that this competition further encourages salutary innovation.11
The competition model supports the idea that cities can do something about economic growth and decline if they have good management. Moreover, good management means adopting a business model
that entails keeping costs down while keeping quality up. Thus, a conventional view is that cities can do well by keeping their fiscal houses
in order, keeping taxes relatively low, not redistributing monies from
rich people to poor people, keeping crime down, and by otherwise
providing an environment conducive to capital attraction and retention, either with policies that attract skilled labor or with policies that
attract firms that will employ skilled labor.12 The idea that good management can make a difference to the city economy is what animated
supporters of Mayor Michael Bloombergâ€™s recent successful campaign
to repeal term limits in New York City. Supporters argued that
Bloomberg had the skills to guide the city through tough economic
times.13 The folk wisdom seems to be that the occupant of the mayorâ€™s
office matters in terms of urban outcomes.
Of course, many local government scholars argue that local managerial competence is not sufficient. A popular view among local
government scholars (including myself) is that cities are not well positioned to help themselves because they do not have the legal and
structural tools to do so.14 The prescriptions that follow are geared toward giving cities the capacity to govern effectively. Thus, policymak-
11 See id at 418â€“23; Fischel, Homevoter Hypothesis at 58 (cited in note 1). See also Weingast,
11 J L, Econ, & Org at 5 (cited in note 5) (noting that the mobility of capital gives rise to a diverse
array of public goods packages); Wallace Oates, Fiscal Federalism 49â€“50 (Harcourt, Brace 1972).
But see Hongbin Cai and Daniel Treisman, Does Competition for Capital Discipline Governments?
Decentralization, Globalization, and Public Policy, 95 Am Econ Rev 817, 818 (2005) (arguing that
competition has a polarizing effect, causing cities that are unable to attract mobile capital to abandon business-friendly policies in favor of predation or the interests of existing citizens).
12 See Peterson, City Limits at 25â€“27 (cited in note 1).
13 See Michael Barbaro and David Chen, Term Limits Get Notice in Crisis, NY Times B1
(Sept 17, 2008).
14 See, for example, Richard Schragger, Can Strong Mayors Empower Weak Cities: On the
Power of Local Executives in a Federal System, 115 Yale L J 2542, 2564 (2006). For the seminal
work along these lines, see generally Gerald Frug, The City as a Legal Concept, 93 Harv L Rev
1059 (1980) (detailing â€œcity powerlessnessâ€ within the government structure that places both
federal and state restrictions on a cityâ€™s ability to self-govern, tracing this development through
history, and arguing for the increased power of cities).
316 The University of Chicago Law Review [77:311
ers argue that cities should have more legal authority so that they can
innovate,15 that they should be able to annex suburban land so that
they can capture economic growth,16 or that there should be legal rules
that make it more difficult for suburban jurisdictions to avoid their
own responsibilities for the regional poor.17 The notion here is that
certain legal and political structures make cities more or less able to
affect their own outcomes, but that these structures can be reformed,
thus placing cities in a position to pursue prosperity unconstrained.
What these accounts share is an emphasis on city agency. Whether one believes that local governments currently have too little,
enough, or (in some cases) too much power, the assumption is that
city, suburban, andâ€”more generallyâ€”regional policies have consequences for local economic growth and decline. One can assume that
there is a fairly full-fledged market in location because of the mobility
of residents and firms and that this market is, on the whole, a positive
one (for the same reasons that all markets are good). This is the stance
of many Tiebout-inspired public choice scholars. One can assume,
more cautiously, that there is something of a market in location, in the
sense that some residents and firms are mobile, but that it is shaped to
a significant extent by government policy. This is the stance of many
reformist local government scholars. In either case, jurisdictional competition remains. Either cities themselves are to blame for their decline or the policies that circumscribe cities are to blame.
As I have noted, these kinds of claims sit uneasily alongside technological accounts of the city. Nevertheless, one does see the argument
that cities will more likely develop and prosper in relatively open societies, where trade is encouraged and predation by despots is limited,
and where there is a relatively stable legal regime.18 In these stories,
the quality of governmentâ€”an amenity over which policymakers exert a great deal of controlâ€”matters a lot.
But we should be cautious. First, there is a causal problem: it is far
from clear whether urban life requires stable political and legal insti-
15 See, for example, Frug and Barron, City Bound at 50â€“51 (cited in note 2).
16 See, for example, David Rusk, Cities without Suburbs 85 (Woodrow Wilson Center 1993). 17 See, for example, Southern Burlington County NAACP v Township of Mount Laurel, 336
A2d 713, 724â€“25 (NJ 1975) (holding that zoning restrictions may not be employed to eliminate
low- and moderate-income housing). See generally Richard Briffault, Our Localismâ€”Part II:
Localism and Legal Theory, 90 Colum L Rev 346 (1990) (criticizing local autonomy as insulating
suburbs from the problems of urban poverty).
18 See Robert Inman, Financing City Services, in Inman, ed, Making Cities Work 328, 328
(cited in note 8).
2010] Theory and Practice of Local Economic Development 317
tutions or that it generates those institutions.19 Cities do not arise after
the establishment of cultural, legal, and political institutions. The city
may in fact be a necessary precondition for them.20 Second, urbanization is occurring in the absence of good institutions. The mega-cities of
the developing world are growing dramatically, often despite the absence of stable political institutions or formal property rights, let alone
the provision of basic municipal services.21 To say that the worldwide
process of urbanization represents intercity competition for providing
urban amenities seems absurd. Urbanization is occurring, but it is
simply not plausible to describe it as the outcome of a market in local
governments or to claim that some cities are winning in a kind of good
I do not mean that institutions do not matter in the growth or decline of cities. They might, though in ways that we cannot necessarily
anticipate. My main point is that â€œcompetitionâ€ is an unconvincing
story of how cities come to be, grow, and decline. Indeed, as David
Schleicher has recently observed, Tiebout-inspired competitive accounts of local government cannot tell us why cities form in the first
place.22 The competition story treats the city as a product (to be sold in
the â€œlocation servicesâ€ market), while the technological story views
the city as a byproduct (of distance-changing technology). Neither
account gives us an adequate insight into why cities exist and why in
A different way to look at the city is as a process. Consider Jane
Jacobsâ€™s famous The Death and Life of Great American Cities.
approached the urban environmentâ€”the city, the neighborhood, the
streetâ€”ecologically.24 She showed us how a safe street emerges from
tiny interactions among hundreds of independent and noncoordinated
19 This is the causal problem that bedevils the international law and development literature and particularly the â€œnew institutional economics.â€ See Adam Przeworski, The Last Instance:
Are Institutions the Primary Cause of Economic Development?, 45 Eur J Sociol 165, 165â€“68, 185
(2004) (arguing that it is incoherent to argue that institutions cause economic growth because
institutions and economic growth have a reciprocal relationship).
20 See Jacobs, The Economy of Cities at ch 1 (cited in note 6) (asserting that agriculture
may have originated in cities and pointing out the fallacy of mistaking the results of city economic development for preconditions of this development).
21 See Barney Cohen, Urban Growth in Developing Countries: A Review of Current Trends
and a Caution Regarding Existing Forecasts, 32 World Dev 23, 48â€“49 (2003); National Research
Council, Cities Transformed: Demographic Change and Its Implications in the Developing World
362â€“84 (National Academies 2003).
22 See Schleicher, 2010 U Ill L Rev at *4â€“5 (cited in note 4).
23 Jane Jacobs, The Death and Life of Great American Cities (Random House 1961).
24 See generally id. See also Jacobs, The Economy of Cities at 125â€“26, 129 (cited in note 6).
318 The University of Chicago Law Review [77:311
actors, how an urban economy emerges from small-scale production
and invention, and how small changes at the street or neighborhood
level can cause catastrophic changes in the urban landscape.25 In short,
Death and Life shows us how order arises out of the seeming disorder
of urban life. It is a caution to those who would attempt to create a prepackaged and attractive urban landscapeâ€”a productâ€”out of a process.
Similarly, the â€œnew economic geographyâ€â€”for which Paul Krugman recently won a Nobel Prizeâ€”approaches cities as â€œselforganizingâ€ systems: â€œsystems that, even when they start from an almost homogeneous or almost random state, spontaneously form largescale patterns.â€26 The new economic geography holds that given particular parameters, urban systems will exhibit certain spatial regularitiesâ€”seemingly natural development patterns that emerge from a set
of basic assumptions. A growing city, on this account, is more â€œlike a
developing embryoâ€ than it is like a widget that is produced and sold
in the marketplace.27
Thinking about cities as processes rather than products highlights
the role of contingency and path dependence in city growth and decline. And it does so without relying on technological determinism. It
thus steers a middle path between predestination and agency. What
the new economic geography seeks to show is why agglomeration
happens, that is, how cities, business districts, and regional and local
industrial clusters occur. Krugman starts with some basic assumptionsâ€”namely, that there are forces encouraging people and businesses to clump together, and that there are forces encouraging people and
businesses to disperse.28 He then models how a city might form under
these circumstances even if individuals and firms begin by being relatively uniformly distributed throughout a particular geographical
25 See generally Jacobs, The Death and Life of Great American Cities (cited in note 23);
Jacobs, The Economy of Cities (cited in note 6).
26 Paul Krugman, The Self-Organizing Economy 3 (Blackwell 1996). See also Masahisa
Fujita and Paul Krugman, The New Economic Geography: Past, Present and the Future, 83 Papers Reg Sci 139, 140â€“41 (2004); Masahisa Fujita, Paul Krugman, and Anthony J. Venables, The
Spatial Economy: Cities, Regions and International Trade 1â€“4 (MIT 1999) (describing urban economic landscapes as â€œthe result not of inherent differences among locations but of some set of
cumulative processes, necessarily involving some form of increasing returns, whereby geographic
concentration can be self-reinforcingâ€). David Schleicher is the first local government scholar to
address this work directly. See generally Schleicher, 2010 U Ill L Rev (cited in note 4). See also
Schragger, 123 Harv L Rev at 491â€“93 (cited in note 3) (discussing economic geography); Schragger,
94 Va L Rev at 1100â€“06, 1156 (cited in note 3) (discussing Jacobs and economic geography).
27 See Krugman, Self-Organizing at 1, 49 (cited in note 26). Compare this description with
Jacobs, The Economy of Cities at 129 (cited in note 6) (analogizing the development of a cityâ€™s
economy to the differentiation of cells in a developing embryo).
28 Krugman, Self-Organizing at 9, 22â€“30 (cited in note 26).
2010] Theory and Practice of Local Economic Development 319
space.29 He also shows how an initially almost uniform distribution of
business within a metropolitan region â€œevolves spontaneouslyâ€ into a
highly structured metropolis with two concentrated business districts.30
Change the parameters a little bit, and one will generate a metropolis
with four business districtsâ€”the so-called â€œEdge Citiesâ€ that Joel Garreau describes in his book of the same name.31
A key insight in all of these models is that small differences in initial conditions or small perturbations in an otherwise stable equilibrium can lead to dramatically different outcomes; that once growth
or decline starts, it does so explosively or catastrophically; and that the
spatial order that emerges may have little to do with individualsâ€™ or
firmsâ€™ preferences understood in isolation. There was nothing particularly unique about the geographical place that became Silicon Valleyâ€”given slightly different initial conditions, such an agglomeration
of high-tech firms could have arisen elsewhere. Nevertheless, once one
sets certain initial conditions, one can make some predictions about
where such agglomerations will arise.32 Similarly, the development of a
regional economic core and a less economically developed periphery
in a particular nation is not a function of the innate characteristics of
those geographic places, but rather the result of historical accident.
Nevertheless, that the industrial belt developed where it did in the
northeast United States at the turn of the century was predictable
considering the initial starting points and the self-reinforcing effects of
economic development.33 Citiesâ€”spatial agglomerationsâ€”are both
contingent and path dependent.
The idea that order arises from chance is not unfamiliar to economists or urban theorists. As Krugman points out, his account of city
formation is inspired by Thomas Schellingâ€™s famous model of how a
segregated metropolis can emerge even given individual preferences
that would be consistent with full integration.34 As Schelling showed,
the integrated distribution of persons in space is unstable; add an almost imperceptible perturbation in the initial distribution of persons
and the equilibrium completely unravels, resulting in a spatial order
that is almost entirely segregated.35 Such cascades were familiar to
29 Id at 24â€“25.
30 Id at 22â€“30.
31 Id at 26â€“27. See Joel Garreau, Edge City: Life on the New Frontier ch 1 (Doubleday 1991).
32 Krugman, Self-Organizing at 37 (cited in note 26).
33 See Fujita and Krugman, The New Economic Geography at 147 (cited in note 26).
34 Krugman, Self-Organizing at 15â€“22 (cited in note 26). See Thomas Schelling, Micromotives and Macrobehavior ch 4 (W.W. Norton 1978).
35 Krugman, Self-Organizing at 18â€“19 (cited in note 26).
320 The University of Chicago Law Review [77:311
Jane Jacobs. She argued that making small changes in a neighborhoodâ€”replacing a store with a parking lot, building a road, closing
down a local groceryâ€”could lead to a neighborhoodâ€™s rapid failure.36
She also noted how small changes in the other directionâ€”the opening
of a store, the influx of a few new residentsâ€”could result in cascades
in the other direction.37 Both decline and growth happen in bursts.
The emergence of regularity from randomness seems to be at
work at the most basic level of economic geography. Consider Zipfâ€™s
law of city size. As Krugman points out, cities in a particular urban
system often exhibit a striking regularity: the population of a given
city is inversely proportional to its rank.38 Thus, in any country, we tend
to see a distribution of city sizes in which the number 2 city has half
the population of the largest city, the number 3 city has one-third the
population of the largest city, and so on. This â€œrank-sizeâ€ rule has remained fairly consistent in the United States since at least 1890; a universal law of city size seems to be at work.39
One can see immediately how Zipfâ€™s law undermines competitive
accounts of city growth and decline. One would predict a much different distribution of population if in fact cities were competing for firms
and residents by providing particular services or goods, or amenities.
Unless individualsâ€™ and firmsâ€™ preferences for location exactly tracked
the rank-size rule, cities would arguably fall within relatively similar
size ranges. Zipfâ€™s law implies that cities exist in an urban system: population increases and decreases operate across the urban system, not
solely at the level of a particular city or metropolitan area. Something
about the ways in which cities form, become populated, and develop
or decline is geographicalâ€”that is to say, is a reflection of an emergent
Does all this mean that cities cannot do anything about their economic development, as the technological account implies? Not necessarily. City self-organization grants a large role both to historical acci-
36 See Jacobs, The Death and Life of Great American Cities at 275â€“76 (cited in note 23).
37 See id at 279 (discussing the potential for â€œunslummingâ€ in lively, diverse neighborhoods).
38 Krugman, Self-Organizing at 39â€“43 (cited in note 26) (discussing the â€œspookyâ€ consistency of the rank-size rule in US cities). See also Xavier Gabaix, Zipfâ€™s Law for Cities: An Explanation, 114 Q J Econ 739, 741â€“43 (1999).
39 See Krugman, Self-Organizing at 43â€“46 (cited in note 26). But see Voker Nitsch, Zipf
Zipped, 57 J Urban Econ 86 (2005) (reviewing quantitatively the empirical literature on Zipfâ€™s
Law and suggesting that â€œcities are on average more evenly distributed than suggested by (a
strict interpretation) of Zipfâ€™s lawâ€).
40 See generally Luis Bettencourt, et al, Growth, Innovation, Scaling, and the Pace of Life in
Cities, 104 Proc Natl Acad Sci 7301 (2007) (describing additional regular, scalar relationships
across urban systems).
2010] Theory and Practice of Local Economic Development 321
dent and the self-reinforcing effects of economic development. Luck
matters because small random events (David Packard and William
Hewlett begin their business in Silicon Valley) can produce large consequences (once HP is there, other firms want to be there). Thus, cities
might be able to intervene in the local economy in such a way as to
nudge the process forward. Clustering of people and firms is the chief
characteristic of city space and the central economic benefit for which
some firms and residents are willing to pay higher rents. Thus, if firms
believe that other firms are going to settle in (for example) Chicago,
then the city becomes a self-fulfilling prophecy.41 Chicago is able to
beat St. Louis at the start of the urban race for no reason other than
firmsâ€™ beliefs that other firms will want to be there. Indeed, the original city boosters believed that the city was really the promise of the
city. They might have been right.
Nevertheless, it is very hard to tell what the effects of any particular policy intervention will be over the long term. Certainly, Chicago
would do well to set itself up as the gatekeeper to the West by building
railroads; in an era in which railroads drive the economy, a city would
do well to have as many lines running through it as possible.42 Nevertheless, the reason that Chicago beat St. Louis, Cincinnati, and Milwaukee to become the leading metropolis of the West had as much to
do with luck as with any set of policies. During the first third of the
nineteenth century, each of those cities (and many with names less
well known) were aggressively pursuing infrastructure development,
gathering capital, and positioning themselves to control regional trade.
Many of these cities started out with similar geographical, social, infrastructural, and economic attributesâ€”or at least attributes that offset one another.43 Indeed, in the antebellum period, St. Louis seemed
to be better positioned than Chicago, having a larger population and a
seemingly firm grip on the Mississippi trade.44 The rapid emergence of
an urban hierarchy with Chicago at the top is a result of the dramatic
41 Paul Krugman, History versus Expectations, 106 Q J Econ 651, 666 (1991).
42 But see Robert Fogel, Railroads and American Economic Growth: Essays in Econometric History 10â€“16 (Johns Hopkins 1964) (arguing that railroads were not essential prerequisites
for economic growth).
43 For a discussion of the Chicago-St. Louis rivalry, see Jeffrey S. Adler, Capital and Entrepreneurship in the Great West, 25 J Interdiscipl Hist 189, 189â€“93 (1994) (arguing that Chicago
emerged as the leading city of the West because it had more extensive connections to the Northeast marketplaces than St. Louis); William Cronon, Natureâ€™s Metropolis: Chicago and the Great
West 295â€“309 (W.W. Norton 1991).
44 See Cronon, Natureâ€™s Metropolis at 296 (cited in note 43).
322 The University of Chicago Law Review [77:311
instability of the urban system during the mid-nineteenth century.45 It
could have been otherwise.
Instability is endemic. Indeed, it is not at all clear that cities can
resist the ups and downs of the urban cycle. That cycle might be independent of city policy, or might be caused by the very processes unleashed by an urban boom in the first place. The new economic geography treats spatial economic phenomena the same as temporal economic phenomena.46 Thus, in the same way that temporal economic
busts follow booms, so do spatial economic busts follow booms. Urban
systems are simply spatial economic systems, and thus will experience
the spatial equivalent of the business cycle.47
What policy interventions might work in such a system? The view
that cities are location providers tends to support policies like fiscal
conservatism, low taxes, minimal economic regulation, and provision
of amenities to skilled workers.48 But if cities are not merely location
providers, if they are systems within a larger urban system, then it is
not entirely clear that adopting any of these policies (or their opposites) would be bad or good.
It is certainly difficult to say what cities ought not to do. For example, a popular assumption among urban economists is that cities
cannot and should not redistribute from rich to poor.49 If they do, the
rich will flee to other jurisdictions, thus bringing down the city economy. But this prediction depends very strongly on the notion that cities
are products and that mobile residents and firms pick and choose
among them. Now, it is true that firms and residents are mobile; the
45 Id at 265, 297â€“309.
46 See Krugman, Self-Organizing at 53, 61â€“73 (cited in note 26).
47 Id at 99â€“100.
48 For the best statement of this view, see Robert Inman, City Prospects, City Policies, in
Inman, ed, Making Cities Work 1, 1â€“19 (cited in note 8). A similar view holds in the international
development arena. For a critique, see Alfred C. Aman, Jr, Law, Markets and Democracy: A Role for
Law in the Neo-liberal State, 51 NY L Sch L Rev 801, 814â€“15 (2006â€“2007) (criticizing the assumption that markets, not law, will solve global problems and advancing a notion of law as a tool to
ensure that individuals participate in the democratic process as citizens rather than mere consumers
of services); Curtis Milhaupt and Katharina Pistor, Law and Capitalism: What Corporate Crises
Reveal about Legal Systems and Economic Development around the World 21 (Chicago 2009).
49 See, for example, Peterson, City Limits at 210 (cited in note 1). But see Schragger, 123
Harv L Rev at 518 (cited in note 3) (noting that cities can accomplish redistribution when capital
is differently mobile). The effect of redistribution on mobile taxpayers is at the center of the
debate over the imposition of a â€œmillionaire taxâ€ in New York. Compare Nicholas Confessore,
Taxing Rich Wonâ€™t Necessarily Make Them Flee, Experts Say, NY Times A26 (Mar 19, 2009) with
Edward L. Glaeser, Will a â€œMillionaire Taxâ€ Cause an Exodus of Talent?, NY Times Economix
Blog (Apr 27, 2009), online at http://economix.blogs.nytimes.com/2009/04/27/will-a-millionairetax-cause-an-exodus-of-talent (visited Nov 3, 2009).
2010] Theory and Practice of Local Economic Development 323
economic models that Krugman deploys depend on firms and residents changing their location. The problem is that mobility across geographic spaceâ€”toward or away from the cityâ€”is not the same kind of
mobility that characterizes purchases of consumer goods. Mobility in
geographical space is â€œlumpyâ€ and prone to all the characteristics of
self-organizing systems: interdependence, cascades up or down, and
Cities might be competing in particular arenas. No doubt some residents and firms make decisions about where to locate based on the existence of quality urban infrastructure, like a good airport, for example. But
cities also emerge without any infrastructure whatsoeverâ€”the infrastructure comes later, after people and firms have flocked there.
II. THE URBAN RESURGENCE (AND THE CRASH OF 2008)
So why are people and firms flocking to particular places? One
way to test our intuitions about the role of policy in city growth and
decline is by considering the renewed popularity of the central city
over the last twenty years or soâ€”what has been dubbed the â€œurban
resurgence.â€51 This resurgence has defied the conventional wisdom
that â€œsun and sprawlâ€ beats â€œold and cold.â€52 During the second half of
the twentieth century, we were told, residents and firms were acting on
their preferences for warmer weather and a car-driven lifestyle, thus
leading to the decline of the industrial cities of the Northeast in favor of
the sprawling cities of the Sunbelt. And even if people were not migrating south, they were migrating out of the cities to the suburbs. These
preferences were made possible by technological innovationsâ€”
relatively cheap transportation abetted by road-building, inexpensive
communications, and the end of an era of industrial production, permitting firms to locate at a distance from suppliers and consumers.53
The puzzle is that while sun and sprawl cities have undoubtedly
been gaining, northern industrial cities have also surged in recent
50 Krugman, Self-Organizing at 99â€“100 (cited in note 26).
51 See generally Michael Storper and Michael Manville, Behaviour, Preferences, and Cities:
Urban Theory and Urban Resurgence, 43 Urban Stud 1247 (2006). See also Edward L. Glaeser
and Joshua D. Gottlieb, Urban Resurgence and the Consumer City, 43 Urban Stud 1275, 1275
(2006). Compare Ingrid Gould Ellen and Katherine Oâ€™Regan, Reversal of Fortunes? LowerIncome Urban Neighborhoods in the US in the 1990s, 45 Urban Stud 845, 866 (2008) (finding
strong evidence that â€œan urban resurgence of sorts did take place in the 1990s and that it extended to the lowest income neighborhoods,â€ but that higher-income neighborhoods did not
52 Storper and Manville, 43 Urban Stud at 1249, 1251 (cited in note 51).
53 See Rae, City: Urbanism and Its End at 361â€“63 (cited in note 7).
324 The University of Chicago Law Review [77:311
years. New York, Boston, Chicago, London, Paris, and to a lesser extent, places like Pittsburgh and St. Louis, have all seen their fortunes
rise dramatically. It is difficult to find a single explanation or set of
explanations for why both these trendsâ€”continued growth in sprawling, sunny cities and resurgent growth in old, cold citiesâ€”are happening at the same time.54 It is even more difficult to point to specific government policies that have led to the rise of certain cities, the decline
of others, and the simultaneous growth of suburban and some central
Consider a number of explanations. One possibility is that certain
cities became more efficient in their provision of essential public services. If we saw significant improvements in urban schools or a notable reduction in city taxes with no noticeable decrease in city services,
we might conclude that resurgent cities were simply performing better. But we do not see dramatic improvements along those lines, and
certainly not consistently across cities.55 Though it is commonly asserted that cities can only attract and keep the middle class if they
improve their education systems, education gains do not seem to have
preceded the urban boom in places like New York, Chicago, and Boston.56 And dramatic improvements in education or increases in education spending do not seem to explain the boom in Las Vegas or other
54 See Storper and Manville, 43 Urban Stud at 1251, 1269 (cited in note 51) (noting that none of
the major explanations adequately explain the increasing popularity of both types of cities).
55 Consider local tax rates. While a number of booming cities have low local tax rates,
others have relatively higher local tax rates. It is true that struggling northeast cities like Bridgeport, Detroit, and Baltimore are at or near the top in terms of local tax burden. It seems unlikely
that the higher tax burden has led to the decline of those cities, however. Rather, the decline of
those cities has led to the higher tax burden. For an analysis, see Government of the District of
Columbia, Tax Rates and Tax Burdens in the District of Columbiaâ€”a Nationwide Comparison
56 At best, student achievement during the 1990s shows mixed results. See G. Alfred Hess,
Jr, Understanding Achievement (and Other) Changes under Chicago School Reform, 21 Educ
Eval & Pol Analysis 67, 79â€“80 (1999); Diane Ravitch and Joseph P. Viteritti, Introduction, in
Diane Ravitch and Joseph P. Viteritti, eds, City Schools: Lessons from New York 1, 3â€“5 (Johns
Hopkins 2000); S. Paul Reville and Celine Coggins, eds, A Decade of Urban School
Reform: Persistence and Progress in the Boston Public Schools 133â€“34 (Harvard Education 2007).
As for New York City, the data arguably show some gains starting in 2002. See generally National Assessment of Educational Progress, 2007 Trial Urban District Assessment: New York City
Highlights (2007), online at http://schools.nyc.gov/daa/reports/2007_NAEP_TUDA_Results.pdf
(visited Oct 27, 2009) (detailing the standardized test results of fourth and eighth graders in New
York City and comparing the results to 2002â€“2003 levels). But see Diane Ravitch, Mayor
Bloombergâ€™s Crib Sheet, NY Times A23 (Apr 10, 2009) (arguing that any gains in the New York
City schools have been overstated). City-suburban gaps across all major citiesâ€”including Sunbelt citiesâ€”are still dramatic. See Sam Dillon, Large Urban-Suburban Gap Seen in Graduation
Rates, NY Times A14 (Apr 22, 2009).
2010] Theory and Practice of Local Economic Development 325
Sunbelt cities.57 Dramatic improvements in city provision of other
kinds of servicesâ€”trash pick-up, social services, infrastructureâ€”also
do not seem to be forerunners of an urban resurgence, though perhaps
they have accompanied it.58
Crime reduction might be an exception, but its role in the urban
boom is likely overstated.59 No doubt, many cities became safer in the
1990s. But it is not at all clear that city policies were directly responsible for that outcome.60 Though community and broken windows policing are credited with the turnaround, especially in New York City, it
appears that all cities throughout the United States experienced the
same trend.61 Crime was simply going down in the late 1980s and into
the 1990s. The reasons are complex and controversial, but arguably
not attributable to any particular cityâ€™s policing policies.62 Another
more significant problem with the crime explanation is that it does not
explain the resurgence of European old and cold cities63â€”which did
not have a crime epidemic in the first place.64 Nor does it necessarily
explain the booming of Sunbelt cities. For example, in Las Vegas crime
stayed relatively constant through the 1990s but began to spike quite
dramatically in 2002.65
57 For a summary of Las Vegasâ€™s progress, see Roberta Furger, Full House: The Las Vegas
Building Boom Has Stretched the Creativity and Resources of the Fastest-Growing School District
in the Nation, Edutopia 31, 33â€“34 (Sept/Oct 2004). But see Dillon, Urban-Suburban Gap (cited in
58 See Richard Deitz and Jaison R. Abel, Have Amenities Become Relatively More Important than Firm Productivity Advantages in Metropolitan Areas? 17 (Federal Reserve Bank NY
Staff Reports No 344, Sept 2008), online at
http://www.newyorkfed.org/research/staff_reports/sr344.pdf (visited Nov 3, 2009) (observing that
it is difficult for any region to change its relative amenity position over a decade).
59 See Steven D. Levitt, Understanding Why Crime Fell in the 1990s: Four Factors That
Explain the Decline and Six That Do Not, 18 J Econ Persp 163, 163â€“67 (2004).
60 See Philip Cook, Crime in the City, in Inman, ed, Making Cities Work 297, 301 (cited in
note 8) (noting that â€œno expert predicted [the decline in crime] and it remains something of a
mysteryâ€ and noting that the decline of the 1990s made any policy intervention â€œlook goodâ€).
61 See id at 301.
62 See Levitt, 18 J Econ Persp at 172â€“73 (cited in note 59).
63 See Ivan Turok and Vlad Mykhnenko, Resurgent European Cities?, 1 Urban Rsrch &
Prac 54, 66â€“72 (2008).
64 See Martin Killias and Marcelo F. Aebi, Crime Trends in Europe from 1990 to 1996: How
Europe Illustrates the Limits of the American Experience, 8 Eur J Crim Pol & Rsrch 43, 45â€“54 (2000).
65 See University of Nevada Center for the Analysis of Crime Statistics, Uniform Crime Report
Database, online at http://www.unlv.edu/centers/crimestats (visited Nov 4, 2009) (showing dramatic
increase starting in 2002). See also Police Executive Research Forum, Chief Concerns: A Gathering
Stormâ€”Violent Crime in America (Oct 2006), online at http://www.policeforum.org/upload/GatheringStorm-PRINT-Final_110473745_1027200610304.pdf (visited Nov 4, 2009) (describing the recent dramatic rise in violent crime across the country).
326 The University of Chicago Law Review [77:311
If cities have not gotten better at providing goods and services,
then something else might explain the urban resurgence. For some
theorists, the preferred theory is that firms have chosen to congregate
in cities because of the economic gains of agglomeration.66 The â€œdeath
of distanceâ€ was supposed to have spelled the end of the city altogether. As that story was told, firms in the new service and knowledge
economy could locate anywhere and so would avoid the congestion
and higher rents of cities, especially old, cold cities.67 But the opposite
seems to have happened, at least in certain industries and in certain
cities. While manufacturing has tended to decentralize, â€œknowledgeâ€
and â€œinnovation-basedâ€ industries have tended to congregate, either
in places like Silicon Valley (technology) or in cities like New York
(finance, advertising, art). Having a deep labor pool has always been a
defining feature of the city; it is there that firms can find the specialized workers necessary for producing their products and where workers can find multiple firms that might hire them.68 But in a knowledge
economy, the benefits of proximity may be more in the transfer of information and in the cross-industry fertilization necessary for innovation.69 Cities, far from being dead, may be more important than ever to
knowledge- and innovation-specific industries.
This story is a popular one, and it helps explain why cities continue to exist, but it does not tell us why certain cities become the hosts
of certain agglomerations.70 Why did financial services agglomerations
end up in London and New York? Why has Charlotte, North Carolina
not overtaken both of them? Charlotte has sun, sprawl, and an increasing array of financial firms. One possibility is simply historyâ€”
New York has been the financial capital of the country since the
founding of the market at Wall Street; for that reason, New York retains its preeminence.
66 See, for example, Storper and Manville, 43 Urban Stud at 1250 (cited in note 51).
67 Id at 1248â€“49.
68 See Alfred Marshall, Principles of Economics 271â€“72 (Macmillan 8th ed 1940). See also
Storper and Manville, 43 Urban Stud at 1249â€“51 (cited in note 51).
69 Storper and Manville, 43 Urban Stud at 1249â€“51 (cited in note 51). See also Edward Glaeser, Are Cities Dying?, 12 J Econ Persp 139, 147â€“50 (1998); Masahisa Fujita, Towards the New Economic Geography in the Brain Power Society, 37 Regional Sci & Urban Econ 482, 484â€“85 (2007)
(incorporating knowledge as a force of agglomeration in the field of economic geography).
70 See Storper and Manville, 43 Urban Stud at 1254 (cited in note 51):
The disadvantage of emphasizing agglomeration economies is the great weakness we discussed before: the inability to explain the where question, and therefore the inability to
draw policy-relevant conclusions. The firms may attract (or create) the labour and a virtuous circle may begin from there, but why do the firms end up where they do?
2010] Theory and Practice of Local Economic Development 327
Certainly path dependence plays a large role in urban economic
growth and decline. But how path dependence will be experienced is
often hard to predict. For example, Detroitâ€™s long history of dominating the automobile industry could have led it to its current deteriorated state. Because it was undiversified, Detroit could not weather
the decline of the American car industry. A different story could be
told, however. In this one, Detroitâ€™s long engineering history is an asset. Having a deep bench of engineers, manufacturing plants, and specialized labor, Detroit could have been well placed to compete in the
post-industrial knowledge economy, and it could now be a leader in
spin-off industries. Of course, Detroit is not a leader in practically anything. Detroitâ€™s history works against it. New Yorkâ€™s history works for
it. But how would we know except in retrospect?
One possibility as to why particular cities are thriving is that firms
go where the labor is and that labor likes certain amenities. This explanation for the urban resurgence argues that rising cities have been
able to attract skilled labor or individuals with high levels of human
capital.71 Certainly, cities with higher growth seem to have higher
numbers of college-educated residents. Attracting such residents, famously labeled the â€œcreative classâ€ by Richard Florida,72 seems an obviously smart strategy. Thus, much effort has been made to create the
kinds of amenities that highly skilled people favor. Those amenities
might include particular kinds of shops, cultural offerings, a vibrant
urban street life, or a generally â€œbohemian, tolerant atmosphere.â€73 The
more young, college-educated people there are in a city, the more likely it will be successful. The idea is that if you build it, they will come,
and the jobs will follow them.
One has to wonder why they come, however. Between 1995 and
2000, many of the old, cold cities saw a net in-migration of young college-educated people, thus supporting the theory that dense, urban
cores were attracting this highly mobile group.74 But cities like Charlotte and Atlanta grew faster along this demographic, and Las Vegas
experienced the greatest increase of all.75 To be fair, it is possible that
preferences for amenities are still at work and that different cities
simply offer different packages, depending on their circumstances.
71 See Glaeser, The Death and Life of Cities at 22, 50 (cited in note 8).
72 Richard Florida, The Rise of the Creative Class: And How Itâ€™s Transforming Work, Leisure, Community and Everyday Life (Basic Books 2002).
73 Storper and Manville, 43 Urban Stud at 1252 (cited in note 51).
74 Id at 1253.
328 The University of Chicago Law Review [77:311
Cities are â€œbundledâ€ goods: you have to take the weather in New York
to get access to the Metropolitan Museum of Art; you can have an
exciting nightlife in Las Vegas if you are willing to put up with the
desert. Making sure that those amenities are in place to the extent that
a city can control them seems like a good bet. Indeed, one urban
economist argues that the â€œbest economic development strategy is to
provide amenities that will attract smart people and then get out of
The problem with this claim is that it is not at all clear whether
these high human capital individuals are migrating to Boston and Las
Vegas because of the amenities or whether the amenities are there
because of the high human capital individuals. Indeed, the amenity
story might have the causation exactly backwards. As Michael Storper
and Michael Manville point out, this is certainly true of places like
Silicon Valley, which had no preexisting amenities to offer before it
became a technological center; Las Vegas, which was similarly devoid
of the kinds of urban amenities that serve a growing permanent population; and Hollywood, which developed its amenities simultaneously
with the development of the motion picture industry.77 Places like Atlanta and Charlotte have all seen their consumer amenities grow with
the influx of a more skilled population. The old, cold cities may be
able to attract some of the mobile educated with their urban charm,
their prewar architecture, and their restaurants, but if there are no
jobs, it is unlikely amenities alone will help them. It may be that
people generally locate where they can maximize their access to
jobsâ€”jobs precede people, not the other way around.78
Nevertheless, policy prescriptions for cities have long emphasized
attracting the â€œright kindâ€ of people. The old recommendations were:
tear down slum housing, provide festival marketplaces, attract suburbanites by building roads into the downtown, and subsidize manufacturing by providing location incentives. The new policies are: go wireless or green, attract the creative class, subsidize universities and hospitals (â€œeds and medsâ€), provide walkable downtowns, and have prewar housing. Of course, some of these prescriptions are not in the
76 Glaeser, The Death and Life of Cities at 50 (cited in note 8). See also Glaeser and Gottlieb, 43 Urban Stud at 1297 (cited in note 51).
77 See Storper and Manville, 43 Urban Stud at 1254 (cited in note 51).
78 See id. See also Deitz and Abel, Have Amenities Become Relatively More Important? at
16â€“18 (cited in note 58) (estimating that the share of wage- and rent-compensating differentials
attributable to amenities rose slightly between 1990 and 2000, but observing that firm productivity advantages remain dominant and that it is difficult to change a regionâ€™s relative amenity position over a ten-year period).
2010] Theory and Practice of Local Economic Development 329
cityâ€™s control. A city has as much prewar housing as it will have. But
there are other policies over which the city does have some control,
such as keeping taxes low or adopting policies that reduce labor or
These policies sound plausible, but what if they are not actually
responsive? The amenity explanation generally leads to policy recommendations designed to attract highly mobile, skilled residents and
prevent them from fleeing.80 But if the industrial agglomeration story
is more accurate than the amenity story, then individual highly skilled
amenity users are not particularly mobile.81 They follow the firms that
employ them, and those firms need particular locales where other
firms are. And where other firms are located might be an accident of
history, or a quirk, or a function of a spatial equilibrium that is difficult
to predict. It seems unlikely that a city is going to attract those firms
by building a municipal Wi-Fi network, by developing a â€œcoolâ€ city
persona, or even by keeping development costs down.
Consider New York City. In the 1970s, few would have predicted
the kind of real estate appreciation that practically all of New Yorkâ€™s
boroughs have now experienced. But New York still has very high
construction costs and very high development costs.82 It also has relatively high taxes.83 While crime declined during the cityâ€™s ascendency,
the schools did not improve dramaticallyâ€”whatever education gains
have been made are relatively recent.84 And while the city has attracted skilled workers, its advantage may actually be in its ready
supply of low-wage workers, especially recent immigrants. New York
79 See, for example, Robert P. Inman, Financing City Services, in Inman, ed, Making Cities
Work 328, 349â€“52 (cited in note 8); Joseph Gyourko, Urban Housing Markets, in Inman, ed,
Making Cities Work 123, 147â€“51 (cited in note 8).
80 For the clearest statement of this view, see Glaeser, Exodus of Talent (cited in note 49).
81 Schragger, 123 Harv L Rev at 514â€“17 (cited in note 3).
82 See New York Building Congress and New York Building Foundation, New Yorkâ€™s
Rising Construction Costs: Issues and Solutions 1 (2008) (reporting that nonresidential construction costs in New York City average 60 percent more than in Dallas, 50 percent more than in
Atlanta, and that total construction costs for high-rise office towers can exceed $400 per square
foot in New York, compared to $180 in Chicago).
83 New Yorkâ€™s overall tax burden places it in the top quarter of big cities. See Government
of the District of Columbia, Tax Rates and Tax Burdens at 14 (cited in note 55). New York City
residents pay the highest income tax rate in the country. See Josh Barro, NYC Income Taxes
Going from Ridiculous to Ridiculouser, Tax Foundation Tax Policy Blog (Dec 5, 2008), online at
http://www.taxfoundation.org/blog/show/24013.html (visited Nov 4, 2009).
84 The data arguably shows some gains starting in 2002. See National Assessment of Educational Progress, 2007 Trial Urban District Assessment at 21, 24â€“31, 37, 43 (cited in note 56). But
see Ravitch, Mayor Bloombergâ€™s Crib Sheet, NY Times at A23 (cited in note 56) (arguing that
any gains in the New York City schools have been overstated).
330 The University of Chicago Law Review [77:311
illustrates the limits of generalizable explanations for the urban resurgence. Those explanations often seem to be a lot of â€œjust soâ€ stories,
not really capable of explaining the renewed popularity of New York,
Boston, London, and San Franciscoâ€”or Phoenixâ€™s or Las Vegasâ€™s rise
(and potential fall?) and Detroitâ€™s continued fall.
Moreover, what if a cityâ€™s particular attraction policies are insignificant compared with larger economic trends? In the same way that
the urban resurgence tests our theories of urban growth, the crash of
2008 tests our theories of urban resiliency, raising questions about
what cities can do in the context of large-scale economic restructuring.
New York is again the example. It is telling that Edward Glaeser, explaining New Yorkâ€™s resurgence, has argued that the financial sector
has been an â€œinnovation engine.â€85 In the 1960s, Glaeser writes, â€œthe
groundwork was being laid for New Yorkâ€™s finance-based resurgence,â€
after which â€œidea built on idea, as people in older, denser areas
learned from each otherâ€ to create â€œongoing improvements in the
ability to assess the mispricing of assets.â€86 Of course, we now know
that what we got was the opposite, a system of learning that reinforced
mistake after mistakeâ€”a much different story about the actual operation of financial industry agglomerations. The story of New Yorkâ€™s success is thus also the story of its undoing.
But there is a larger lesson to be learned, other than that agglomeration economies do not always produce the best ideas. We now
know that the urban boom of the last two decades has been driven by
what all urban booms have been driven byâ€”land speculation. Land
has always been the engine of spatial boomsâ€”land drives the urban
political economy and is the initial basis for city wealth.87 Ultimately,
the city has to be based on real productive activity, howeverâ€”and that
holds for the cities that profit from underwriting the creation of new
cities. In the 1800s, New York money made Chicago; the New York
financiers bet on that city and won.88 More recently, New York money
bet on Las Vegas, Miami, and thousands of other places. The financiers
lost those bets, however, and the costs to the New York financial sector (and the New York-area economy) are significant.
More important, for our purposes, are the policy implications.
Prior to 2008, many (though not all) urban theorists would have told a
relatively positive story about New Yorkâ€™s financial sector. And very
85 Glaeser, The Death and Life of Cities at 49 (cited in note 8).
87 Schragger, 94 Va L Rev at 1156 (cited in note 3).
88 See Cronon, Natureâ€™s Metropolis at 297â€“307 (cited in note 43).
2010] Theory and Practice of Local Economic Development 331
few would have argued that the cityâ€”were it empowered to do soâ€”
should more heavily regulate or tax that sector. New York was blessed
with a money-making and innovation-making financial sector agglomeration that was the envy of every other city in the world; to regulate or tax it would be to precipitate its flight. But, of course, it turns
out the worst thing for New York was to be so heavily dependent on
the financial sector. The best thing the city could have done five or ten
years ago was to limit its exposure to that sector and deemphasize it.
But how would one have known?
III. OLD STRATEGIES AND NEW ONES
What should New York do in the face of the general economic
collapse that followed the collapse of the cityâ€™s financial industry? It is
worth observing that the city exercises little to no regulatory control
over the financial industryâ€”that is the responsibility of federal and, to
a lesser extent, state regulators. Even if city leaders had wanted to,
they could have done very little to prevent the irresponsible lending
and securitization practices that led to the current economic crisis.89
Despite the limits of city regulatory authority, New York, like
most cities, engages in ongoing economic development activities, and
has done so with special regard to the financial industry. In 2004, the
city provided $100 million worth of incentives to Goldman Sachs in
response to its threat to relocate to New Jersey.90 The incentives may
have worked; Goldman stayed in Manhattan. In 2009, following the
collapse of the financial sector, the city proposed investing $45 million
to retrain finance professionals and provide them with seed capital
and office space for startup firms.91 The cityâ€™s economic development
corporation hopes to make small investments in new ventures with
the aim of attracting additional private capital and keeping finance
professionals in the city.
89 The collapse of the US financial industry nicely illustrates the mismatch between regulatory scale and effectsâ€”national regulation seems both too large and too small considering both
the disproportionate local effects of the collapse and its global reach. The gap between local and
national effects is endemic. Consider the recent argument about financial industry bonuses: for
the city, the bonuses would be a tax boon and any decent mayor would want them to be paid. For
the nation, however, the bonuses look like the redistribution of federal tax dollars to the undeserving rich. Mayor Bloomberg can weigh in on the bonus issue, but he has almost no tools to put
his preferences into policy. See David W. Chen, Much Vilified, Financial Titans Find a Friend in
Bloomberg, NY Times A17 (Apr 14, 2009); David W. Chen, Economistâ€™s Forecast: Chance of
Change 100%, NY Times A17 (Feb 16, 2009).
90 See Patrick McGeehan, After Reversal of Fortunes, City Takes a New Look at Wall Street,
NY Times A19 (Feb 23, 2009).
91 See id.
332 The University of Chicago Law Review [77:311
How should we evaluate these two approaches to economic development? The first policyâ€”location subsidies to large employersâ€”
has been and continues to be a standard tool of city economic development offices. Their efficacy, however, has been strongly disputed.92
Cities appear not to gain back what they put in, either in the short
term or the long term, and there is some evidence that subsidies do
not ultimately alter the location decisions of firms.93 And even if location subsidies do enhance local welfare, they do not improve overall
welfareâ€”one city loses what another city gains.94 This is especially true
if the firm relocates in the same economic region, as Goldman Sachs
would have had it moved to New Jersey. From an economic perspective, Goldman Sachs was never going anywhere.
Finally, if firms are ultimately influenced by subsidies, those subsidies distort firm location decisions. A move by Goldman Sachs to New
Jersey is really a move from high rents to low rents. But there is no reason that owners of commercial real estate in Manhattan should be subsidized; all location subsidies do is keep rents artificially high, which is
not good for anyone. Giving money to Goldman Sachs is the postindustrial version of smokestack chasing: firms play one city or region
against another, generating a subsidy race with dubious welfare effects.95
The cityâ€™s more recent response to the decline of the financial
services industry might be more promising. Instead of giving money to
a large, established firm to encourage it to stay, the city is using taxpayer money to retrain finance professionals and subsidize startup
firms. The city might have no other choice; it cannot effectively subsid-
92 See Yoonsoo Lee, Geographical Redistribution of US Manufacturing and the Role of
State Development Policy, 64 J Urban Econ 436, 436â€“37 (2008).
93 See id at 445. See also Terry F. Buss, The Effect of State Tax Incentives on Economic
Growth and Firm Location Decisions: An Overview of the Literature, 15 Econ Dev Q 90, 97â€“99
(2001); Carlos F. Liard-Muriente, US and EU Experiences of Tax Incentives, 39 Area 186, 189â€“90
(2007) (reviewing literature).
94 See William Thomas Bogart, The Economics of Cities and Suburbs 237â€“39 (Prentice Hall
1998) (comparing competition among cities to the â€œprisonerâ€™s dilemmaâ€ problem and arguing
that while the overall situation would be improved if cities did not invest resources competing
with one another, any given city has good reasons to offer location subsidies); Peter D. Enrich,
Saving the States from Themselves: Commerce Clause Constraints on State Tax Incentives for
Business, 110 Harv L Rev 377, 397â€“98 (1996) (arguing that state tax incentives may provide a
benefit to the states if these states are able to attract more businesses, but that incentives do not
create a net benefit to the country as a whole because businesses choose one state to the detriment of others).
95 See Enrich, 110 Harv L Rev at 395 (cited in note 94). See also Schragger, 94 Va L Rev at
1139 (cited in note 3). Jacobs thought that â€œtransfer economiesâ€â€”economies that grow by relocating assets from elsewhereâ€”were a weak basis for ongoing economic development. See Jane
Jacobs, Cities and the Wealth of Nations 208â€“10 (Random House 1984).
2010] Theory and Practice of Local Economic Development 333
ize the location choices of an industry that is declining precipitously.
But even so, the new focus on small startups might be better over the
long term. By investing in human capital, the city keeps and improves
its labor force, which may be the key determinant for attracting new
firms. And by backing small, entrepreneurial ventures the city reduces
its vulnerability to one employer or industry, which could leave or decline. Moreover, smaller ventures have the capacity to grow and may
be more attached and committed to their original locations. Indeed, by
encouraging small-scale innovation, the city might help seed a homegrown agglomeration at low cost. Growing your own is better than
stealing from others or preventing others from stealing from you. It
enlarges the economic pie rather than just shifting it around.
At least that might be the theory. Jane Jacobs, for one, argued that
a growing city needs to generate new ideas, processes, and servicesâ€”
to â€œadd new work to old.â€96 She favored a small business strategy,
arguing that the city should encourage entrepreneurial firms and the
learning that goes on in those firms. In an unavoidably volatile economy, the cityâ€™s most important â€œskillâ€ is its ability to reset its product
life cycle by generating new startup firms (Jacobs called them â€œbreakawaysâ€).97 In her estimation, large, vertically integrated multinationals have less ability to drive such a process; a more fluid and open local economy does better than one dependent on large firms.98
Jacobsâ€™s emphasis on startup firms is consistent with a theory Ronald Gilson has offered for why Silicon Valley leads the Route 128
corridor outside of Boston in economic growth.99 Gilson argues that a
quirk in competition lawâ€”adopted in the 1800s when California became a stateâ€”helps explain the difference.100 For reasons mostly unrelated to competition policy, California does not enforce covenants not
to compete, while Massachusetts does. The result, according to Gilson,
is that Silicon Valley employees experience much higher rates of mobility across firms and to new startup ventures, and that this fluidity
has given Silicon Valley an innovation edge.101
The conclusions Gilson draws are pure new economic geography.
He argues that (1) happenstance plays an outsized role in develop-
96 Jacobs, Economy of Cities at 59 (cited in note 6). See also Schragger, 94 Va L Rev at
1101â€“02 (cited in note 3).
97 Jacobs, Economy of Cities at 67, 97â€“98 (cited in note 6).
98 Id at 71, 79.
99 See Ronald Gilson, The Legal Infrastructure of High Technology Districts: Silicon Valley,
Route 128, and Covenants Not to Compete, 74 NYU L Rev 575, 577â€“79 (1999).
100 See id at 613â€“19.
101 See id at 576â€“80.
334 The University of Chicago Law Review [77:311
ment; (2) small policy interventions might make a huge difference in
outcomes; but (3) it is very difficult to predict what those interventions should look like.102 In the case of technological firms, a more fluid
employment environment seems to encourage knowledge spillovers,
but that is not necessarily the case with other kinds of firms or industries. Moreover, Gilson might be wrong about the long-term effects of
a serendipitous legal rule on the local economy. As of todayâ€”ten
years after Gilsonâ€™s articleâ€”Route 128 is considered a highly successful technological agglomeration, admittedly second to Silicon Valley,
but still robust despite the continued enforcement of noncompete
clauses in Massachusetts.103 It turns out to be difficult to predict how
policy will affect spatial economies over the long term.
So should cities be in the business of subsidizing existing firms or
growing new ones? How one answers this question implicates a longrunning debate in urban economic development concerning the degree to which cities should look outward or inward for economic
growth. The outward argument is economic base theory, which holds
that cities can only grow economically by increasing exports to other
places.104 Fewer exports means less money for local spending and less
growth overall. This means that economic development strategy is
unlikely to favor small businesses that service the local community, or
businesses that are intended to substitute locally made goods for those
that the city formerly imported. The export-based theory of local economic development tends to favor large transnational corporations or
those corporations that can exploit local resources for cross-border
sale. And it favors specializationâ€”for example, producing goods and
services that cannot be imported.
Economic base theory has been the dominant approach to local
economic development for some time, but it has not been uncritically
102 See id at 619â€“28.
103 See Boulanger v Dunkinâ€™ Donuts, 815 NE2d 572, 582 (Mass 2004) (enforcing a former
franchise ownerâ€™s agreement not to operate or work for a competing business within five miles
of any Dunkinâ€™ Donuts establishment for two years). Moreover, Silicon Valley hiring practices
might be more constrained than Gilson describes. See Miguel Helft, U.S. Inquiry into Hiring at
High-Tech Companies, NY Times B2 (June 3, 2009) (reporting that the Justice Department is
investigating allegations that Silicon Valley technology firms have been agreeing not to actively
recruit employees from one another).
104 See Douglass North, Location Theory and Regional Economic Growth, 63 J Polit Econ
243, 251 (1955). For an overview, see generally Ted Rutland and Sean Oâ€™Hagan, The Growing
Localness of the Canadian City, or, On the Continued (Ir)relevance of Economic Base Theory, 22
Local Econ 163 (2007). See also John Adams, Editorial, 29 Urban Geography 741, 742 (2008)
(providing a synopsis of economic base analysis and asserting that the approach does not account for large and increasingly self-sufficient metropolitan economies).
2010] Theory and Practice of Local Economic Development 335
adopted. In the 1950s, Charles Tiebout questioned whether exports
were the only mechanism of city growth.105 Tiebout claimed that as a
local economy grows â€œits market becomes large enough to efficiently
produce some goods and services that it had previously imported.â€106
Jacobs also argued, even more strongly, that import substitution could
drive city growth, as locals make for themselves what others had formerly made for them.107 A city can grow by providing more goods and
services for itself, and by preventing money and resources from flowing outside the local economy. The implication is that self-sufficiency
and leakage prevention should be a large part of economic development efforts.108
Current-day import-substitution advocates do not argue that making your own is sufficientâ€”as Jacobs and Tiebout recognized, export
growth is still needed. But current advocates do point to trends in the
latter half of the twentieth century to support their argument that economies are becoming more local and that those economies are also growing.109 As US manufacturing declined during the second half of the twentieth century, the portion of the metropolitan-area economy that was
produced and consumed locally rose.110 This increasing localness has been
attributed to the rise of the service sector economy in general.111 In larger
cities, localness has increased as a bigger share of the local economy
moved towards professional services. Producers of accounting, legal,
marketing, and management consulting services tend to locate near their
clients, within the same city or metropolitan area. It is this clustering that
105 See Charles M. Tiebout, Exports and Regional Economic Growth, 64 J Polit Econ 160,
161 (1956) (â€œThere is no reason to assume that exports are the sole or even the most important
autonomous variable determining regional income.â€); Charles M. Tiebout, Exports and Regional
Economic Growth: Rejoinder, 64 J Polit Econ 169, 169 (1956). See also Charles M. Tiebout, The
Urban Economic Base Reconsidered, 32 Land Econ 95, 97 (1956); Rutland and Oâ€™Hagan, 22
Local Econ at 165â€“66 (cited in note 104).
106 Rutland and Oâ€™Hagan, 22 Local Econ at 165 (cited in note 104).
107 See Jacobs, The Economy of Cities at ch 5 (cited in note 6) (discussing, for example, how
Tokyoâ€™s imported bicycle trade gave rise to a market for locally made bicycle parts which in turn
led to the domestic manufacture of complete bicycles).
108 See id at 169. For advocacy of this strategy, see Michael Shuman, Going Local: Creating
Self-Reliant Communities in a Global Age 52â€“58 (Free Press 2000). For a review of the importsubstitution debate in the context of international economic development, see Henry J. Bruton,
A Reconsideration of Import Substitution, 36 J Econ Lit 903, 904 (1998).
109 See Rutland and Oâ€™Hagan, 22 Local Econ at 179â€“83 (cited in note 104).
110 See Joseph J. Persky, Marc Doussard, and Wim Wiewel, Export Orientation and the
Limits of Local Sovereignty, 46 Urban Stud 519, 522â€“23 (2009).
111 See id; Michael Thomas Power, Lost Landscapes and Failed Economies 37 (Island 1996);
Paul Krugman, Pop Internationalism 211 (MIT 1996).
336 The University of Chicago Law Review [77:311
appears to account for the dramatic economic growth of particular
â€œglobal cities,â€ like New York, London, or Tokyo.112
Whether the local share of productive activity will continue to increase is an open question. Some scholars have indicated that the
trend is starting to move in the opposite direction, as service firms
begin to export more of what they do.113 This is not surprising. Indeed,
it seems consistent with Jacobsâ€™s argument that once cities begin to
replace exports with homegrown goods and services, the city is likely
to become proficient to the point of creating a new export market for
those services.114 In an industrial era, it made sense to think of local
economies as starting with exports, which would in turn generate a
demand for local services as residents and workers flowed into the
jurisdiction. But in a service economy, that story looks less plausible,
and the causation might be reversed.115
New York Cityâ€™s two interventions in the financial industry can
be understood in the context of this wider debate. There is a significant shift in emphasis that is reflected in the decision either to support
a large-scale multinational investment bank or to fund local startups.
Local economic development strategies can seek to enhance â€œglobal
competitivenessâ€â€”an oft-used phrase in city development circles. Or,
a local economic development strategy might focus on supporting
those firms and industries that mainly provide goods and services to
locals. The startups that New Yorkâ€™s business incubator will fund will
most likely begin by providing services to local clients and by producing goods for the local market. And the retraining of existing workers
seems like a leakage-prevention strategy rather than an exportproducing one.
Of course, growing your own is not inconsistent with exportingâ€”
Jacobs describes import substitution and export growth in cyclical
terms. And we certainly know that having particular industrial agglomerations can be beneficial to the city as long as the city is not overly
reliant on them. Incentivizing Goldman Sachs to stay in Manhattan
may have been a means of preventing the unraveling of New Yorkâ€™s
financial services agglomeration. The city surely feared that Goldmanâ€™s move might trigger other firms to follow, though in retrospect
112 Persky, Doussard, and Wiewel, 46 Urban Stud at 522â€“23 (cited in note 110). See also
Saskia Sassen, The Global City 20â€“22 (Princeton 1991).
113 Persky, Doussard, and Wiewel, 46 Urban Stud at 523â€“24 (cited in note 110).
114 See Jacobs, Economy of Cities at ch 5 (cited in note 6).
115 See id at 245â€“46. See also Persky, Doussard, and Wiewel, 46 Urban Stud at 519â€“20 (cited
in note 110).
2010] Theory and Practice of Local Economic Development 337
that fear looks a little foolish considering the current fortunes of the
industry as a whole. Relying too heavily on exports is risky because it
makes a city vulnerable to global demand crashes. The less selfsufficient oneâ€™s economy, the more global trends will affect it.116
As always, however, there is the question of agency. An exportdriven or import-substitution orientation implies different approaches
to local economic development efforts. But it does not give us much
purchase on the merits of specific, time-bound government decisions.
That is because contingency is so hard to plan. New Yorkâ€™s location incentives turned out to be a bad bet, but the cityâ€™s investment in startups
might also turn out badly. Innovation is a characteristic of a flourishing
urban economy, but it may be very difficult for policymakers to generate or sustain it through specific economic development policies.
Indeed, it may be that cities experience booms and busts over
which they have little control. Douglas Raeâ€™s story of post-industrial
New Haven certainly seems to conclude with this lesson, though he is
much more of a technological determinist than he is a believer in
business cycles.117 So does Guian McKeeâ€™s recent book, The Problem
of Jobs, which recounts Philadelphia policymakersâ€™ attempts to deal
with a looming post-industrial future.118 City officials understood the
problem of de-industrialization much earlier than one might have
thought. They took steps to address the problem, and even experienced limited success, but ultimately could do little about larger economic trends.119 One possibility is that they adopted the wrong strategy,
concentrating on manufacturing when they should have been looking
at something else. In the 1950s, however, it would have been truly visionary for the cityâ€™s managers to adopt a different approach.
As a practical matter, any economic development strategy that
shifts money from taxpayers to private firms has to be measured
against some other use of taxpayer moneyâ€”say, building better
schools, providing more policing, or producing better health care. And
this gets us back to the question of what makes a city do better or
worse economically. Providing good municipal services and creating
healthy, smart people is something that any city should aspire to. But
doing so does not ensure economic success. There are lots of reasons
for this: any given city policy will be mismatched to the metropolitan-
116 Persky, Doussard, and Wiewel, 46 Urban Stud at 525 (cited in note 110).
117 Rae, City: Urbanism and Its End at 363 (cited in note 7).
118 See generally Guian A. McKee, The Problem of Jobs: Liberalism, Race, and Deindustrialization in Philadelphia (Chicago 2009).
119 Id at 15â€“16.
338 The University of Chicago Law Review [77:311
wide scale of economic development; much is driven by chance or
path dependency; larger economic and technological events will overshadow local interventions. Some of these problems can be overcome.
But the biggest problem is that we do not really know enough about
what works and what does not.
All of which may be beside the point. One might argue that government should provide services to make people smarter, healthier,
and safer anywayâ€”whether those policies attract and retain economyproducing residents and firms or not. Indeed, there are lots of good
reasons to enact policies that reduce inequality, foster participation, or
limit urban sprawl, even if we are unsure what effects those policies
will have on economic growth. Certainly, much of reformist local government scholarship is driven by this view.
If one adopts a competitive paradigm of city growth, howeverâ€”
the assumption that a cityâ€™s economic development is really a competition for mobile taxpayersâ€”then the city cannot (and should not)
engage in policies that attend solely to the well-being of current residents. Those policies will fail, undermining local welfare by inhibiting
economic growth. Cities must compete, and one way to do so is to reduce taxes to the point of subsidizing highly mobile residents and
highly mobile industry.
Thinking of cities as spatial economic processes undermines this
simple assumption. There may be a form of â€œcompetitionâ€ between
cities but it is not as straightforward as providing good municipal services while keeping taxes low.120 In short, the provision of public services in a hypothetical local government marketplace is not an explanation for local economic growth and decline. Tiebout knew thisâ€”his
marketplace model was offered as a solution to the problem of public
goods provision, not as an account of how cities generate economic
growth. His other scholarshipâ€”which legal scholars of local government ignoreâ€”makes this plain.
I am not arguing that law and policy are irrelevant to urban outcomesâ€”only that they are not relevant in the ways we have sometimes
120 See Richard E. Baldwin and Paul Krugman, Agglomeration, Integration and Tax Harmonisation, 48 Eur Econ Rev 1, 1â€“2 (2004) (arguing that the existing paradigm misunderstands
how localities â€œcompeteâ€ in a spatial economy characterized by the uneven distribution of economic activity). See also Richard Shearmur, Of Urban Competitiveness and Business Homelessness, 29 Urban Geography 613, 613â€“15 (2008) (dismissing the idea of city competitiveness as
2010] Theory and Practice of Local Economic Development 339
assumed. A city is neither a product to be sold in the local government
marketplace nor a byproduct of distance-changing technologies. The
city is a spatial economic process, a part of a larger urban system. That
urban systemâ€”like all economic systemsâ€”is beset by booms and busts.
Any set of local policies must take this fact into account.
Nor am I arguing that the insights of economic geography will
provide a definitive answer to the central question of whether cities
are capable of controlling their long-term economic fates. Nevertheless, how policymakers address urban problemsâ€”and centrally the
problem of economic developmentâ€”turns importantly on how we
conceive of the city and the metaphors we use to describe it. Treating
the city as an economic process provides for a more nuanced account
of the relationship between government policy and economic development. The literature on local government and urban policy has often taken that relationship for granted.