“Superstar Cities” & the Generation of Durable Inequality

The striking economic agglomerations emerging in affluent democracies are generating, reproducing, and expanding inequalities. A major mechanism for this is housing, which is both a repository for wealth and, under these conditions, a magnifier of wealth. Access to urban areas – the site of educational, labor, and marriage market advantages – is contingent upon access to housing. We use comparative analysis of cases in Europe (London and Paris) and the United States (New York and San Francisco) to consider the capacities of different societies to limit or ameliorate these new sources of diverging opportunity. These seemingly local issues remain shaped by distinct national political contexts, which vary dramatically in their capacity to support local affordable housing and reduce the collective action problems confronting major metropolitan areas.

The economic processes of creative destruction, long ago defined (and celebrated) by political economist Joseph Schumpeter, have a pronounced spatial dimension. The Marxist geographer David Harvey described "the spatial fix" of capitalism and the increasing use of land and property as financial assets to be traded like any commodity. 2 Some cities and regions-their institutions, culture, economy, and political organizations-are made obsolete and marginalized. Mobile capital is reinvested in new places, cities, or states offering the highest rates of return. These spatial transformations are  central to the creation of new inequalities, as evidenced by the decline of older, once prosperous industrial centers, major cities and small towns alike, from Detroit and Milwaukee in the American Rust Belt, to Northern England, Coventry, Liege, Lille Roubaix Tourcoing, Leipzig, large chunks of Southern Italy, and the nonmetropolitan parts of Eastern Europe. Working-class manufacturing cities are dying, along with their hinterlands. Those living there-especially young, white, working-class men-are facing social decline and a sense of loss. 3 Simultaneously, modern political economies are producing extraordinary agglomerations of wealth in key urban centers well placed to benefit from the rise of new technologies, services, and finance. While a number of forces drive this trend, including the search for safe havens among global economic elites, the primary factor is the increasing value of density in a knowledge economy. 4 Ideas emerge, spread, and can be exploited more easily in dense urban settings. As a result, favorably situated cities become vital centers for both attracting and generating human capital, radically reshaping the spatial structure of advanced capitalism. 5 The shifting spatial distribution of economic activity has generated growing social and political tensions in both North America and Europe. Some of these are within these agglomerations, as their rapid development disrupts established social patterns and communities. Others are between these agglomerations and less-favored areas outside these economic cores. With the increasing centrality of place in distributing economic and cultural rewards and opportunities, these urban concentrations constitute important emerging structures of inequality production and reproduction. We begin this essay by briefly describing these socioeconomic shifts. We then outline the ways in which they both generate and, increasingly, perpetuate inequalities. It is that perpetuation-the process of deepening inequality in which advantage builds upon advantage-that we wish to emphasize and explore.
Our empirical focus is housing. Central to this structural shift in opportunity is a sharp rise in property values as a result of these agglomeration processes, reinforced by the rapid growth of mobile capital at a time of cheap credit. The combination of low growth and low interest rates tends to create asset bubbles, and the largest and most consequential of these assets is housing in the most desirable and booming cities. 6 On their own, rapidly rising housing prices are a generator, repository, and transmitter of inequality. 7 This is most apparent in the distribution of wealth, which is probably the most important indicator of durable distributions of advantage and disadvantage. Rising housing wealth is a straightforward mechanism for the intergenerational transmission of assets, and is the most important one for the overwhelming share of households who hold little or no financial wealth. Indeed, some economists argue that Thomas Piketty's story of rising inequality in twenty-first-century capitalism (at least in the United States, which he sees as at the vanguard of a broader transformation) is mainly a story of the growth of highly unequal housing wealth. 8 In many countries, as we shall see, the erosion of inheritance taxes has further enhanced these intergenerational wealth transfers.
Yet rising housing prices have profound indirect effects on the transmission of inequality as well. Increasingly linked to the sites of greatest economic opportunity, changes in housing prices have a powerful impact on the distribution of life chances. For those without access to this increasingly valuable property, the new urban economy is becoming a landscape of exclusion. Economic advantage is increasingly concentrated and reproduced through property, which carries with it the cumulative advantages of social proximity. If you don't have it (or stand to inherit it) you are pushed down, into increasingly precarious conditions, or out, forced to relocate or remain at a distance. Exclusion from these economic centers has consequences for life chances. Structures of social advantage-access to elite education, social networks, employment, and the acquisition of cultural capitalare geographically concentrated. Distance from these structures, fueled by housing costs, comes with an increasing penalty.
Akin to the changes in educational structures analyzed in the contribution of David Grusky, Peter Hall, and Hazel Rose Markus to this issue of Daedalus, the rise of these superstar cities contributes to the commodification of opportunity-highly unequal opportunity-through property ownership. To see how this works, and why it plays out differently in distinct settings, requires an approach that pays attention to processes unfolding over time and involving the interplay of actions at the micro-, meso-, and macrolevels. At the microlevel, individuals choose locations for status and access to services and labor markets, with effects that shift the opportunity structures for both the privileged and disadvantaged. At the mesolevel, interest groups representing competing claimants and organizations like political parties struggle over how to respond. At the macrolevel, the available options and pressures these actors face are shaped by changes in the global economy, including the transformation of financial flows and the growing metropolitanization of capitalism.
We examine these interlinked processes in four "superstar cities," the high-income, globally connected metropolitan areas diverging from the rest of their respective countries: New York, the San Francisco Bay Area, Paris, and London. The comparisons reveal important commonalities in both the processes generating these agglomerations and the profound social challenges associated with them. At the same time, we emphasize important differences. Public authorities in these cities often are the ones directly responsible for facing these problems. Their responses, however, have varied in both content and effectiveness. A principal reason, we argue, is that these authorities operate within quite different national contexts. National institutional arrangements remain essential determinants of potential reactions to emerging place-based inequalities, creating different opportunity structures for local actors, as well as very different options and capabilities associated with existing policy structures.
Cities have always been cauldrons of social change. In the late nineteenth century, the rise of globalized large cities (Vienna, Berlin, London, Paris, New York, Buenos Aires) led sociologist Georg Simmel to highlight their distinctiveness: the emergence of different mentalities, the constant excitation of these spaces, their increased social diversity, and the accompanying fears of the unknown and unfamiliar. In the last third of the twentieth century, analyses typically (especially in the United States and the United Kingdom) dealt with urban decline: New York's bankruptcy, massive deindustrialization, or low quality of life (boring and gray London or museified Paris). 9 In many Western countries, the population of the national capital was stable or in decline. Los Angeles, seen by many as the "new" urban model, yielded a new school of urban studies, preoccupied with the

Patrick Le Galès & Paul Pierson
dystopian challenges of brutal policing, social unrest, and gangs.
Today, the allure of cities has returned. Urban prosperity, while accompanied by astonishing inequality and exclusion, is a defining feature of advanced capitalism. A generation of political economists, economic geographers, economists, and urbanists has documented the urban revival. From the mid-1980s onwards, European cities began enjoying economic and demographic growth along with greater political capacities. 10 In the United States, gloom about cities persisted longer, but the trend was the same. Leading U.S. cities have enjoyed considerable economic and demographic growth. 11 New York and Los Angeles contain 10 percent of the U.S. population, and account for 13 percent of its gdp. The gap is even more pronounced in Europe. London and Paris account for about 20 percent of their national populations and close to 30 percent of their respective gdps. Relative rates of growth in income, employment, and the formation of new enterprises have all shifted in favor of these dynamic urban areas. 12 High-income globalizing cities are becoming more productive, attracting an increasing share of the skilled labor force, and pulling in tourists along with new inhabitants, firms, and capital. For those with economic means, "urban hubs" have become desirable places to visit and to live, with unparalleled concentrations of amenities: job opportunities, social networks, transport, elite educational institutions, sophisticated hospitals, and media and communications networks. They are also places of conspicuous consumption, with a prestigious cultural infrastructure from opera to jazz clubs, museums to theaters, and elite and specialized shops, star restaurants, and trendy cocktail bars.
Those on the outside are affected too. More and more indicators point toward greater differences between those living in cities and those living far away, particularly in rural areas or declining industrial towns. From life expectancy to access to health services, from income to social mobility, from education to voting behavior, social life is becoming more spatially polarized. 13 The divide between core and periphery is intensifying. Clearly the vote for Hillary Clinton, for Remain in the United Kingdom, and for Emmanuel Macron and Jean-Luc Melenchon in France has been a distinct "urban vote." By contrast, Donald Trump, Marine Le Pen, and Brexit have done particularly poorly there. As an example, Marine Le Pen's share of the presidential vote in the core of Paris was only 5 percent.
The remarkable transformation of these urban areas has had a powerful impact not only on immediate wealth and income distribution, but also on the generational distribution of social and economic opportunity. Economic dynamism in these geographically compact spaces operates as a kind of escalator. Those who can get on have improved opportunities for upward mobility in comparison with the rest of the country. 14 Those who stay (or arrive) have a good chance to find high-paying jobs, to build human capital, and to access social and cultural resources that enhance the likelihood of passing their improved social prospects, and the opportunity for more, on to their children. As a result, these urban hubs attract large numbers of newcomers from their own countries and abroad.
These "escalator" qualities are part and parcel of the concept of agglomeration. Large cities absorb the high end of the labor market, attracting skilled people (the young in particular) and investment. New York, San Francisco, London, and Paris all have per capita gdps substantially higher than national average incomes. New York's per capita gdp is a full one-third higher than national averages, while San Francisco's is almost 40 percent higher. High-income, larger European cities (such as Paris, London, Munich, Milan, and Stockholm) are becoming increasingly different from the rest of Europe's cities. 15 In the emerging knowledge-based economy, spatial proximity confers multiple advantages. 16 First, residents in these locations have access to high-wage labor markets, providing future economic opportunities. Wage dispersal among U.S. cities has grown. 17 Some of this economic edge reflects the substantial benefits to being near clusters of high-skilled workers and firms. Your work is more valuable (and your wages higher) when it can be easily combined with that of others with high skills. The people and organizations around you allow you to make more of the human capital you have.
Second, with density comes a "thickness" of labor markets that offers substantial protection against individual risks associated with the knowledge economy. An increasing share of high-end economic activity is in highly volatile sectors. Constant innovation means that new firms or even industries emerge quickly. Most firms do not survive, while a few succeed spectacularly. Since any single position carries considerable risk, skilled workers gravitate to locations where there are many firms seeking their talents. In short, they want to be in an agglomeration.
Third, there is growing evidence that part of what makes these agglomerations function is that they accelerate the accumulation of human capital. Initial opportunities in these economic clusters generate potential for learning more, which in turn creates additional opportunities. Proximity also allows you to gain access to better and more extensive job networks.
Fourth, access to high-wage urban areas can mean access to economically favorable marriage markets, particularly for upper-middle-class women. 18 Indeed, the rising participation of women in highskilled labor markets increases the appeal of agglomeration cities. Whether already part of a couple or anticipating the future, the highly educated wish to be in "thick" high-skilled labor markets where both partners will have good employment prospects. Thus, restricted access to the economic opportunity structures in these high-wage cities may encourage homophily in marriage, increasing durable inequality. The concentration of young, high-wage individuals increases the prospects for widening inequalities between two-earner households able to access these dynamic economies and those who cannot. There is good recent evidence of the inverse pattern in deindustrializing areas of the United States. 19 Loss of manufacturing jobs leads to declining marriage prospects for men. With this decline comes an increase in single-parent households and child poverty.
Finally, the clustering of high-income households in particular areas makes it easier to provide a range of high-quality public goods as well as cultural amenities. This too can produce a self-reinforcing cycle of advantage. Differentiation across space in the quality of public goods (schools, parks, public safety) intensifies incentives for the affluent to cluster, and to prevent these advantages from leaking out to other groups. The poor, of course, face the reverse situation: a limited capacity to finance public goods discourages all but those with no economic alternative from living in those areas. In the United States, the increasing spatial segregation of income has accompanied growing income segregation in schools. 20 The escalator qualities in these economically dynamic cities may have made

Patrick Le Galès & Paul Pierson
them sites of considerable upward social mobility. A remarkable paper by economist Raj Chetty and colleagues reveals large differences in social mobility across the United States depending on where you are born and raised. 21 Strikingly, the "neighborhood effects" identified seem to operate before an individual enters the labor market. The prospects of moving from the lowest to the highest quintile of the income distribution were strongest where racial and class segregation and rates of single parenthood were relatively low. San Jose (San Francisco's southern neighbor), San Francisco, and New York are all among the top ten metro areas (among the fifty largest) for social mobility.
Similar escalator qualities have been identified in France. In the Paris Île-de-France region, for instance, the per capita gdp is 60 percent above the French average (with household income 26 percent above the national average after taxes and transfers). Paris Île-de-France has a greater concentration of middle-class residents and managers (41 percent compared with 27 percent nationally) and a smaller share of working-class residents (9 percent compared with 15 percent nationally). 22 Recent research following the work of Chetty and colleagues has shed some light on the geography of social mobility, defined as the percentage of children from working-class families that become middle class. 23 In deindustrialized regions with low levels of education (such as Northern France), the rate is less than 30 percent; it is the highest in the Paris Île-de-France region (47 percent). Quite strikingly, working-class children born in the Paris Île-de-France region who leave at an early age for another region have a rate of social mobility that is significantly lower than those born in a different region who move to Île-de-France as children.
There are reasons, however, to worry that in agglomeration cities, the social context that these researchers describe is fading. Because Chetty's research focuses on cohorts born in 1980, it represents, inevitably, a view of the recent past rather than the present. And conditions in these agglomeration economies are changing rapidly. Those at the bottom are pushed toward increasingly precarious conditions, or out of these locations entirely. Perhaps even more important, it is becoming increasingly difficult for those not already economically advantaged to move to these escalator regions. A crucial factor in both of these developments is the transformation of housing and exploding housing prices in superstar cities, in particular.
The comparison of house prices faces many obstacles. First and foremost, there is a lack of systematic, comparable, longitudinal data. A second classic obstacle in comparative urban studies concerns scale. Is the proper unit the core city (in London, "inner London" with fourteen boroughs and 3.5 million inhabitants), the extended city (the Greater London Authority, with thirty-two boroughs and 8.6 million inhabitants), or the economic region (London South East, population eleven million)? For our purposes, all of these units are relevant. Despite the challenges, we provide evidence of the patterns of house price increases in London, Paris, New York, and San Francisco.
London (here understood as the Greater London Authority) is emblematic. Its remarkable property boom is linked to an equally remarkable demographic turnaround. Between 1939 and 1991, London lost 2.2 million inhabitants, about 25 percent of its population. In the past quartercentury, it has grown by two million inhabitants. In just the first decade of the twenty-first century, it grew by one In the United Kingdom, house prices have increased by almost 300 percent in real terms between 1960 and 2010. Even as housing prices have risen faster than income across the United Kingdom, London has, especially since the late 1990s, pulled away from the rest of the country. 24 The bulk of the increase started in the late 1990s, with an acceleration after 2000. By 2015, the average home price in London (£536,000) was almost twice the average for England as a whole (£301,000). The average rent is about £15,000 a year. Wages and incomes have failed to keep up. Unsurprisingly, there has been a decline in home ownership in the United Kingdom since its peak in the early 2000s (from 70 percent to 64 percent). Many predict a sharp fall of new homebuyers for the next decade, as Londoners are increasingly priced out of the housing market. 25 Paris consists of three concentric circles: the Paris City Council with twenty arrondissements and 2.2 million inhabitants; the Métropole du Grand Paris, which comprises 131 communes (basic local authorities) and twelve territoires (groups of communes) and 7.5 million inhabitants; and the Île-de-France region, with twelve million inhabitants. The population of the Paris City Council has been relatively stable, while that of the Métropole du Grand Paris has begun to grow, with an increase of one million people in the past decade.
The Paris urban region has, like London, faced massive deindustrialization. Unlike London, however, its main economic engines are still rooted within the French economy in close connection to large firms and French investors. It is a classic escalator region, with an everincreasing proportion of professionals and the upper-middle class and fewer manual laborers. Foreign investors have only become significant since the mid-1990s.
Because of long-standing investments in social housing and the rental sector, the rate of home ownership is relatively low in France. In the Île-de-France urban region, the rate of home ownership is lower still, only 48 percent (against 58 percent nationally). The rate is 35 percent within Paris, the inner core, but 61 percent in the periphery. The rental sector is extensive in France. Forty percent of the population rents, with both a large private rental sector (23 percent of the population) and a substantial public sector (17 percent) of mostly affordable housing supported by government subsidies. 26 Housing prices show a trend similar to London but with less intensity. The price increase was greatest between 1998 and 2008, when prices rose 185 percent in Paris and about 150 percent in the Métropole du Grand Paris, while disposable income increased by only 43 percent. If one measures the cost of housing in the Paris urban region as a percentage of household disposable income (that is, for those who rent or pay a mortgage), the figure was 14.4 percent in 1988 and 20.7 percent in 2013. 27 The average masks an important variation: the growing burden is more evident for the worse off. For those with low incomes, the cost was 21.6 percent in 1988 and 36.2 percent in 2013. However, they have benefited more from public subsidies.
San Francisco and New York, despite important differences explored below, have followed broadly similar trajectories. New York City consists of its five boroughs, with a population of 8.5 million, within a metropolitan area of about Like London, these cities have witnessed a sharp rise in the cost of housing, dramatically escalating already significant issues of affordability. The same changes that make agglomeration cities increasingly attractive to those with high skills have also driven up housing prices. In turn, the sharp rise in housing prices has fed into inequality by triggering processes of increasing spatial segregation.
It is striking that San Jose, identified by Chetty and colleagues as the place offering the greatest prospects for upward mobility for those born thirty years ago, has become the least affordable place in the country to buy a home. 28 In nearby San Francisco, only 14 percent of new homes are affordable to a family with a median income. Home ownership in the city is increasingly limited to those who already own, have a very high income or substantial wealth, or who can draw on the wealth of their parents. Thus, professional/managerial groups are increasingly concentrated in these dynamic urban areas, which have also seen a sharp growth of income in the past few decades. In San Francisco, for instance, 44 percent of those over twenty-five had at least a bachelor's degree, compared with an average of 28 percent in American cities as a whole. 29 In 1999, during the last tech boom, about one-third of San Franciscans spent more than half of their gross income on housing (policy analysts generally regard anything over 30 percent of spending on housing as burdensome). By 2015, that proportion was 47 percent of the population. 30 In New York, the share of households spending 30 percent or more on rent rose from 41 percent in 2000 to 52 percent in 2014. Unsurprisingly, those in the lowest income brackets are most likely to face these intense burdens. By 2014, three-quarters of low-income residents in New York were rent-burdened. 31 In turn, these pressures create the kind of precarious position that can easily spill over into eviction and homelessness. 32 While our main focus is on the impact of rising housing prices on access to mobility opportunities, we wish to return briefly to a direct and powerful effect of these rising asset values on durable inequality, one that highlights the need to consider changes in housing markets and public policy simultaneously. A small number of cities-including New York, Moscow, Hong Kong, San Francisco, and Los Angeles-are home to a large share of the world's booming class of billionaires. 33 London in particular stands out, having refashioned itself as a locale where the superwealthy can invest heavily in property as a useful tool for money laundering, wealth diversification, and hoarding beyond the reach of their homestate's public authorities. 34 This is a distinctive dynamic, with particular causes and effects, which calls out for separate analysis. It does, however, have important ripple effects on broader property markets, which are especially evident in London.
Yet the wealth impact of the housing boom is not just, or even primarily, a question of oligarchs parking (or laundering) their assets. Rising property values in high-income cities are driving large shifts in wealth inheritance. In a precursor of the argument we develop more extensively in the next section, policy choices of national governments figure prominently in this development. In Thus, on one side of the housing divide, one finds existing homeowners and their children. Not only can they sustain access to these sites of economic opportunity and benefit from the forces pushing up the value of their property, now they can also pass more of this wealth across generations without tax. On the other side are those within younger generations who face the specter of massive house price increases without the advantage of wealthy parents. Often, they must cope simultaneously with stagnating incomes and declining prospects for pensions. These young people cannot buy. Locked into renting in increasingly expensive markets, their housing conditions may deteriorate, or their tenancy may become precarious. They may have to leave the high-cost region altogether. All these effects restrict social mobility.
In the United Kingdom, those on the short end of this transformation have been dubbed "Generation Rent." Indeed, a wealth of evidence shows a stark generational contrast. Those born in the 1960s and 1970s had good access to property ownership. The Institute for Fiscal Studies' report on inheritance shows the scope of wealth accumulated by the most favored within the older cohort of property owners. According to the report, due to rising housing prices, "the average wealth of elderly households (where all members are 80 or older) increased by 45%" between 2002-2003 and 2012-2013. 35 The richest 10 percent of the elderly own 40 percent of the wealth to be inherited. By contrast, within the following generation, prices are too high, wages too low, and rents too burdensome to allow an easy path to home purchase. For many, the rise of rents has generated precarious housing situations if not outright homelessness (there are now eight thousand "rough sleepers" in London). At a minimum, it prevents most young people from saving money for the future, with potentially massive consequences for long-term inequality.
In Paris, again, those impacts exist but are more muted. Economist Clément Dherbécourt has mapped the strong increase of inheritance, which has grown from the equivalent of 8 percent of household income in 1980 (€3.5 trillion) to about 19 percent of household income today (€10.6 trillion), and is projected to reach about 30 percent by 2050. 36 Rising house prices explain most of the change. Indeed, since the early 1990s, the net assets of households increased much more rapidly than incomes, as house prices doubled between 1998 and 2008. From 1980 to 2015, the level of inheritance went from €60 to €250 billion (constant prices). As in the United Kingdom, this goes together with an increase of inequality both between and within generations. Assets are far more concentrated than incomes and therefore the increasing importance of inheritance is a powerful mechanism to create long-term inequalities within generations. Without

Patrick Le Galès & Paul Pierson
reaching the levels of inequality seen in London, Dherbécourt nonetheless shows that the average incomes of households who inherit more than €100,000 are more than 30 percent higher than other households.
So far we have examined general social processes associated with high-income metropoles. These processes generate higher housing prices and, in turn, cycles of concentrated advantage and disadvantage. We now contrast the policy experiences of our four cities. All have been part of the broad spatial revolution concentrating increased economic advantage in urban areas. In key respects, however, divergence among these cities is as evident as commonality. Housing prices have not increased equally in all these places. More important, the translation of high housing prices into durable mobility advantages is not fully automatic. Paris stands out as a site where both the run-up in housing prices and its impact on the distribution of economic and social opportunity have been more limited.
The situation of housing today is partly the result of long-term policies and collective choices about what is public and private, and about the priority of housing for the poor and those with modest incomes. Public policy scholar Allen Scott and urban geographer Michael Storper have stressed the urban-land nexus related to density. 37 How do authorities resolve the competition among individuals and groups who want to live in these high-income, dynamic metropoles? What mechanisms should be used to decide land-use patterns that usually reflect a mix of power relations and efficiency concerns? In all of our cases, the increased competition for housing has benefited the upper-middle classes and tended to limit access for lower-income groups, or expel them. Policy responses to address the crisis and to limit inequalities have been diverse, and generally insufficient. Yet the degree to which this inequality has grown and become entrenched has varied. Below we identify some of the proximate sources of these differences and then proceed to explore some of the deeper structures that help account for the distinctive pathways that cities find themselves on.
At least on the surface, housing policy is led by the public and private actorsespecially mayors-within these large cities who have struggled to create coalitions and mobilize resources to develop affordable housing. But looking at cities themselves is not enough. The fragmented governance of these metropolitan areas requires a combined analysis of what is done at the level of the municipality (or the borough in London) and at the level of the metropolitan government, where there is such a thing. Intermediate authorities such as the regional council in Paris or the states of California and New York also possess financial and regulatory resources that may be mobilized.
In the end, we argue, national commitments are crucial. Cities are embedded within national societies. Nation-states have not disappeared, nor have welfare states. A major source of city robustness in the past has been the redistribution and services that national welfare states have provided for urban populations. Despite strong pressures for austerity in some settings, these features continue to shape the pathways of urban transformation. Cities-even "global" ones-remain situated within particular national contexts that continue to exert considerable influence on the development of structures of social and economic opportunity. National policies influence the capacities, options, and incentives of more geographically proximate actors. If, as we argue, Paris is different, it is primarily because France is different. A useful place to begin this exploration is to distinguish supply and demand pressures on housing prices. On the demand side, critical factors include the prosperity of a region, flows of foreign investment into property markets, and population trends. Increasing population, especially among those who can afford to pay more, will bid up housing prices. On the supply side, one can distinguish between public and private construction. To the extent that there is increasing supply from either source, upward pressure on prices will be diminished. By contrast, in the presence of heightened demand, supply constraints will generate what economist Fred Hirsch long ago labeled positional competition, in which housing markets begin to resemble an ever-faster game of musical chairs. 38 The main point is that even within this group of dynamic, global cities, all of which see rising housing prices as a huge challenge, there are important differences in outcomes. We briefly sketch those differences before exploring some of the differing policy structures that generate them-differences that help to illuminate some of the patterns of political and social responses to housing affordability challenges as well.
London has experienced a long-term decline of the role of the public sector, and the national government in particular, in providing affordable housing. The city was once characterized by social housing-its famous housing estatesbuilt before and after World War II. In 1980, the Thatcher government famously introduced the "right to buy" social housing, leading to the sale of 1.8 million council homes nationwide at considerable discount to their occupants. Sales were especially brisk in London, where rising home prices made the deal particularly attractive to tenants. On average, the number of new affordable housing units being built, typically by nonprofit housing associations, offset roughly 10 percent of the number of properties sold. And more recently, under Conservative Treasury Secretary George Osborne, the right to buy was extended to housing association properties as well. There is a general sense of a housing crisis in the United Kingdom and in London in particular. Never-ending fiscal pressure on local authorities (uk public expenditure is one of the three most centralized in Europe) has marginalized social housing. Report after report emphasizes the lack of new construction. The Conservative Cameron-Osborne government made only modest efforts to address this challenge. In 2010, the government developed its "First Buy" scheme, aimed at helping ten thousand households to buy new property by guaranteeing interest-free loans. The program has delivered but the scale is limited. The Cameron-Osborne government's "Help to Buy" scheme offered interest-free loans to first-time buyers and a guarantee to allow them to offer 95 percent of the mortgage with little risk. It was designed to help seventy-four thousand buyers over three years for a cost of £3.5 billion.
It is not just that these efforts were very modest; others cut in the direction of worsening housing affordability. In England, the number of new dwellings built as social housing for rent has declined from about twenty-five thousand in 2006 to six thousand in 2015. A major shift in policy was the decision to cap the housing benefit, in particular in London in 2013, a move that was strengthened in 2016. The importance of the housing benefit in England (and in London in particular) had grown because of increasing house prices. Capping the housing benefit provoked rent increases up to several hundreds of pounds a month in the worst cases. The likely result will be an

Patrick Le Galès & Paul Pierson
additional push on those at the low end of the market, leading either to worsening housing conditions or induced departure from London.
The housing charity Shelter issued a critical report on housing policy under Cameron. 39 It noted that the government's policies have done little to address the supply problem. In 2013, construction was at a postwar low. Conservative Mayor Boris Johnson worked closely with leading private developers and international partners to attract money, but the result was a limited amount of housing, mostly at the very top end of the market. At the end of Johnson's mandate, affordable housing in London represented just 25 percent of new construction, compared with 40 percent when he was first elected. According to government statistics, there are about seventy-five thousand households in temporary accommodation or homeless in the city. Shelter reports that between 2011 and 2016, the number of households in temporary accommodation increased by 60 percent. A National Audit Office study emphasized the curtailment of private-sector tenancies, together with a general decline of affordability of private rentals, as the major reasons for the rise of homelessness. 40 With the withdrawal of national government, efforts to combat the affordability crisis and homelessness have fallen to fiscally stressed local councils. Yet the Cameron-Osborne government also cut local budgets dramatically, up to 50 percent in some cases. As a result, councils in London faced intense pressure to raise capital by selling pieces of land or council estates to private developers. Given tight budgets, most of the new dwellings were built to sell or rent at market prices. And this comes on the heels of decades of decline for social housing. There is now a reduced social housing stock to house homeless people. Private landlords with more lucrative alternatives are less and less willing to house them. 41 Rather than address these long-term supply shortages, scarce resources have been redirected to face the overwhelming immediate crisis. As the National Audit Office report concludes, "local authorities have increased their spending on homelessness while reducing spending on preventing it." Since 2011, spending on overall housing services (mostly for the low-income groups) has decreased by 21 percent in real terms despite a spending increase on temporary accommodations.
Unsurprisingly, housing was the major issue of the 2016 mayoral election in London, figuring prominently in the manifesto of the winning Labour candidate Sadiq Khan. His long-term goal is both to increase the production of housing in London and ensure that 50 percent of new dwellings are affordable. Over £1 billion have been diverted to subsidize affordable housing construction. Yet, forced to rely on local fiscal resources, Khan faces the same challenge as New York and San Francisco. The major resource of local governments is their control over zoning, which gives them regulatory authority over increasingly valuable land. Inclusionary zoning has become the watchword for using this remaining form of government authority to address both the challenges of rising prices and rising exclusivity.
Khan has implemented a number of restrictions on luxury development. He is working closely with Transport for London (tfl), which has become a major developer, releasing land to build new housing around stations. Yet the resources generated in these projects are largely destined to finance transport, another urgent need in an increasingly congested London. Like other booming cities, London faces massive infrastructure needs that compete with subsidized housing for limited local funds. Still, the mayor The mayor is thus employing a carrotand-stick strategy, working with private developers to enroll them in his agenda while exercising coercion when they are recalcitrant. He has launched a number of new initiatives, including subsidies for a new generation of "naked homes"-that is, affordable housing stripped down to the bare essentials-that would sell for 40 percent under the existing market price. 42 In addition to working with private developers and finance companies, Khan is developing partnership deals with nonprofit housing associations to build thousands of new units. His team works on developing incentives and planning regulations at the borough level to deliver 35 percent "genuinely affordable" housing in any new development. The mayor also plans to introduce a "new private sector tenure" with rents set at one-third of average household incomes: that is, "genuinely affordable" for a couple earning around £4,000 per month. 43 Some new homes would be accessible for £1,000 a month to rent instead of the market rent of £1,450.
Mayor Khan has built a coalition to fuel affordable housing construction and fully use the regulatory power at his disposal. Whether these efforts are commensurate with the scale of need, given the backdrop of a pronounced long-term decline in support for housing from the national government, remains very questionable. At the end of the day, neither the Greater London Authority nor the boroughs possess the resources necessary to build the houses needed by lower-income groups. These same challenges and constraints are evident in major agglomeration cities in the United States.
The forces that have led to skyrocketing prices in San Francisco and New York City are similar to those in London. The influx of foreign investment has been more modest, although in New York it is still notable. In both New York and San Francisco, the critical driver is their enhanced status as economic magnets. This has created intensified demand, as highend wages skyrocket and draw affluent workers to these cities. Compared with London, however, population growth has been low, especially in San Francisco. Indeed, American economists have been struck by what previously would have been seen as an economic paradox: booming economies coexisting with limited population growth. 44 The population of both cities has grown, but not nearly as much as rising economic productivity might lead one to expect. Many highskilled workers have entered, but many with fewer skills have headed in the other direction. Many more have been deterred from entering by prohibitive housing costs. They have instead chosen to locate where housing prices are lower, even though economic opportunities are more limited as well. 45 That housing prices have skyrocketed despite modest population growth points to problems on the supply side, where both American cities have faced severe challenges. There is now widespread consensus that supply limitations are central to escalating prices in both New York and San Francisco. With an influx of highincome households and very limited new construction, housing markets quickly exhibit Hirschian-style positional competition. In this brutal game of musical chairs, seats go not to those who are fastest, but to those already in secure possession of a chair and those equipped to win a bidding war for the few available spots. Supply constraints in a context of increasing demand form the backdrop to both the escalating prices and massive resource transfers to property owners described at the beginning of this essay.

Patrick Le Galès & Paul Pierson
There is some dispute about the reasons for the restricted supply. Public resources to support affordable housing have always been relatively limited in the United States. Today, national resources for such efforts are scant by comparative standards (and even by historical standards in the United States). Moreover, zoning restrictions and local opposition have placed marked restrictions on new construction. Indeed, American housing economists and government officials at both the state and national levels now point to these supply restrictions as a chief culprit in the affordability crisis. 46 They document that other American cities lacking such restrictions have seen a more limited run-up in house prices. Despite its booming economy, San Francisco, in recent decades, has built new housing at a rate far below the norm for American metros, as well as the pace it set in prior decades. New York has done only a little better.
The two American cities we examine have had somewhat different responses to the challenges associated with rising housing prices. We begin with the more modest actions of city officials in San Francisco before turning to New York's more aggressive policy response. There are clear differences between the two cities, as well as over time (with each stepping up its efforts in recent years). But the commonalities largely swamp the differences, revealing some of the key features constraining effective policy responses in the United States.
In both New York and San Francisco, housing affordability (along with its most alarming manifestation, widespread homelessness on the same streets that display unprecedented affluence) has come to dominate local politics. San Francisco and the surrounding areas were among the first places to recover from the financial crisis, and even within the booming Bay Area, San Francisco gained ground economically. Between 2010 and 2015, San Francisco and San Jose added over one hundred thousand digital service jobs; their share of the nation's tech employment continued to grow to a remarkable 17 percent. 47 The tenure of San Francisco's Ed Lee, who became mayor in 2011 (and was reelected in 2015), coincided with striking changes in the city. Lee's initial focus was on consolidating San Francisco's economic appeal to the already extensive tech industry. His efforts reflected and reinforced a shift in the political economy of the Bay Area, with a growing share of start-up and venture-capital activity (along with an influx of high-skilled, high-wage workers) moving from Silicon Valley to the city itself. By 2010, new venture capital within the city's limits exceeded that in Silicon Valley. By 2015, there were almost twice as many "unicorns" (private companies with valuations of over $1 billion) in San Francisco as in the Valley; indeed, San Francisco is the headquarters for over 40 percent of all such companies in the world. 48 This economic boom, combined with the influx of high-skilled workers, coincided with a sharp spike in housing demand. Lee's tenure in office was accompanied by escalations in both housing prices and the intensity of conflicts over housing and development. Rents more than doubled between 2009 and 2015. Soaring housing costs in turn rapidly translated into growing signs of social and political strain: activist challenges to the tech industry's prominence, increasingly visible encampments despite growing city expenditures to combat homelessness, and dueling ballot measures seeking to finance affordable housing and either curtail or facilitate private development. Housing has not only become The city has responded to these challenges on a number of fronts. It has substantially expanded resources dedicated to combatting homelessness, both through an extension of social services and efforts to increase the supply of permanent housing for homeless families and individuals. In 2015, the city spent $240 million on homelessness, about 3 percent of its budget and a 25 percent increase in real terms since Lee first took office. 49 It has attempted to increase the supply of affordable housing by financing construction and rehabilitation. Notably, in 2015, the city passed a $310 million bond measure dedicated to affordable housing, with 73 percent voter support. It has encouraged the rehabilitation of public housing through partnerships with nonprofits. 50 And, most extensively, it has-like London and New York-expanded the use of inclusionary zoning both to increase new construction and leverage that construction to generate new affordable units.
Despite the importance of some of these initiatives, their overall impact has been modest. The scale is simply insufficient given the extent of the challenge. San Francisco's flagship bond initiative must be viewed against this background. It will, at current prices, yield, at most, 775 new units of housing-and probably less given escalating construction costs. 51 It is telling that a large share of the city's efforts (supported by modest state and federal funding) is required just to keep the current stock of affordable housing from disappearing. Budgets are far from sufficient for the maintenance of existing public housing. Nationwide, the federal budget for public housing repairs has fallen by over 50 percent in real terms since 2000. It now totals under $2 billion, while the system faces $26 billion in needed maintenance. 52 San Francisco's embrace of a pilot program to transfer management of its public housing stock to nonprofits reflects these painful realities. It may not only preserve, but also improve the quality of 3,500 units of affordable housing; it will not, however, increase the supply.
Meanwhile, extraordinary prices inexorably pull existing affordable housing into the open marketplace. Property owners face powerful incentives to cash in on the housing gold rush. The city's own Housing Balance report, which analyzes efforts over a decade, is telling. 53 It estimates that between 2004 and 2015, the city developed 6,559 new units of affordable housing. Over the same period, 5,470 units were "removed from protective status" with, for instance, the movement of owners into previously occupied apartments. In other words, despite its efforts, the city was barely holding even in absolute terms. Given the trends of sharply rising rents and growing population, it was almost certainly losing ground relative to need.
The California Housing Partnership estimates that state and federal funding for low-income housing in California dropped by 67 percent between 2009 and 2015. 54 All told, federal, state, and local financial assistance for affordable housing in California has subsidized the construction of about seven thousand rental units a year. This amounts to only about 5 percent of total construction in a state of almost forty million people, at a time when the total construction effort falls consistently and woefully short of estimated need. 55 With the national government in absentia and local authorities lacking resources and facing conflicting incentives, the most interesting policy activity

Patrick Le Galès & Paul Pierson
in California is in the state capitol. State policy-makers are increasingly focused on housing policy and, in 2017, finally brokered significant new legislation. The state has introduced new funding for affordable housing construction, but in the face of competing priorities, the impact of the modest new allocations is likely to be marginal. The bigger change is the introduction of zoning reforms that might foster private construction. That the state is now weighing in on these regulatory issues to encourage more housing construction represents an important change, but the new rules are likely to make only a modest difference against a backdrop of entrenched local resistance, in which both cities and suburbs are fiercely protective of their traditional authority.
New York has faced a broadly similar housing crisis, fueled by roughly similar forces. It has, by most accounts, been more aggressive than San Francisco in its response, and its aggressiveness has grown over time. The severity of the housing challenge was recognized under Mayor Michael Bloomberg, who presided over New York's booming economy from 2002 to 2014. The increasing burdens of housing affordability, combined with a spreading sense that the city was becoming the exclusive preserve of the affluent and superrich, was a growing blemish on the mayor's record. Bloomberg set a goal of creating or preserving 165,000 units of affordable housing, mostly through expanded use of inclusionary zoning. Yet despite the city's stepped-up efforts, it lost more affordable units than it gained during Bloomberg's tenure. 56 This disappointing record was one of the catalysts for the surprise election of the progressive Bill de Blasio, who won the mayor's office over Bloomberg's preferred candidate in late 2013. In his most prominent campaign speech, de Blasio promised a robust response to what he decried as a "tale of two cities" dividing rich and poor: Without a dramatic change of directionan economic policy that combats inequality and rebuilds our middle class-generations to come will see New York as little more than a playground for the rich . . . a gilded city where the privileged few prosper, and millions upon millions of New Yorkers struggle each and every day to keep their heads above water. . . . History has taught us that no economy-and no citycan thrive in the long-term under such circumstances. 57 After winning office, de Blasio immediately launched an unprecedented effort to expand city support to preserve and extend social housing. He set what was by historical standards an ambitious goal: two hundred thousand units of affordable housing over a decade. The primary tools for meeting de Blasio's aspirations were two sets of incentives for developers. First, extending Bloomberg's efforts, the city offered to relax zoning in some areas. "Inclusionary zoning" would allow for more and denser construction. In return, the city would get a commitment that a significant share of the resulting housing would be reserved for those with moderate incomes. Second, the city (and state of New York) proposed to offer more generous relief from property taxes for developments providing a designated share of affordable housing. In short, city policy-makers sought to exploit the main tool they controlled: regulatory authority over private construction in a context of exploding demand. The goal was to generate both an expanded housing supply and, within that expansion, a significant affordable component.
Like against a very strong current. Their efforts play out in an environment in which rapidly rising property values create powerful incentives for private actors to remove units from the affordable housing stock whenever possible. Each year, rent-stabilized units are allowed to increase rent modestly, but once they reach a specified threshold, these rents are deregulated entirely and can shift to market rates. Since 2007, a staggering 172,000 units have been deregulated. 58 This is one of several examples of policy drift that are rife in the urban housing crisis in the United States. 59 The interaction between changing circumstances and frozen policies produces fundamental change. Federal housing programs were designed at a time when rent burdens and prices were far lower. With each passing year of rising prices, they become less adequate. They cannot be adapted to meet current social demands without new laws that are impossible to pass in the current political climate.
On the contrary, the federal government is retreating ever further from its limited prior support for low-income housing. As federal agreements on previously subsidized units expire, some are not renewed, and formerly covered housing is converted to market rates. Local governments must devote their limited resources to offsetting the steep decline in affordable housing that would occur in the absence of concerted efforts. Like San Francisco, New York struggles against long odds to preserve its stock of public housing, constructed long ago with massive (but now long gone) federal support. The New York City Housing Authority faces an $18 billion backlog for repairs. Meanwhile, the already grossly inadequate federal spending that is supposed to finance those repairs continues to dwindle. Given the current political and budgetary climate, there is every reason to think federal housing programs may face steeper cuts in the near future.
Thus, despite the energy and resources devoted to meeting the growing housing crisis, many of the city's limited resources must be allocated to not slipping backwards. Bloomberg's administration failed to meet even this test. De Blasio's aggressive plan calls for two hundred thousand units of affordable housing in a decade, but fully 60 percent of the total refers to preserving existing units rather than producing genuinely new supply. As in San Francisco, the city simultaneously faces the additional challenge of dealing with the expensive fallout of the housing crisis, most palpably homelessness. The population in city homeless shelters has doubled, to over sixty thousand, since Bloomberg entered office in 2000. 60 Although declining affordability is not the only cause of homelessness, it is a fundamental one. Just as has happened in London and San Francisco, already scarce city funds must be shifted to deal with the immediate and severe humanitarian challenges.
New York City's housing efforts are on a scale unmatched in the recent American past or in any other present-day American city. In the words of Alicia Glen, de Blasio's deputy mayor for housing and economic development: "I can't even compare it to what other cities are doing. At the risk of sounding like we're the bee's knees . . . we are the bee's knees. New York has always . . . been the place where the most interesting housing policy and programs have happened." 61 And yet the effort is clearly incommensurate with the scale of the challenge. The city receives very limited support from the state of New York or the federal government. State officials are happy to see the city sacrifice its future property tax revenues (and indeed developed a more generous proposal that would involve mandated

Patrick Le Galès & Paul Pierson
high wages for union construction workers), but it offers little financial assistance. The federal government, always a marginal participant in housing, has reduced its efforts sharply over time. In the absence of such support, the needed public resources are certain to be lacking.
Paris, as we have already suggested, represents a distinctive case. It faces all the same pressures as the other three cities, but in more limited form. It remains more socioeconomically diverse, and the strains of housing affordability and precarious living are more muted. How has public policy contributed to this outcome?
France has had a long tradition of direct state and municipal intervention to build affordable and social housing. That impulse receded in the last part of the twentieth century. Since then, however, it has not only returned, but has been adapted and enhanced to meet contemporary challenges. New efforts at the local, regional, and national levels have limited the scale of the housing crisis, as well as the extent to which rising prices generate more durable forms of exclusion. The pressures remain severe. Yet concerted public efforts, both local and national, have made Paris a notably different type of global city.
A great deal of the massive development of the Paris urban region took place after 1945. The Ministry of Housing and the Caisse des Dépôts et Consignations (a publicly owned, large financial institution) started an ambitious program of social housing estates (les grands ensembles) in the suburbs of Paris. The centralist technocratic Gaullist regime accentuated the effort and introduced together an ambitious regional strategy for Paris in 1965, designed to transform urban planning and land use. It combined innovative policy instruments, granting resources to public authorities to impose zoning, to expropriate private owners, to build complementary infrastructure, and to develop social and affordable housing on a large scale. The effort included the construction of the regional transport system, five new towns around Paris, and La Défense (the new business district). The working-class industrial belt on the northern and eastern sides of Paris enthusiastically built up red bastions of municipal communism, concentrating social housing and workers alike.
In the 1970s and 1980s, the French state made a partial retreat. The liberal government of Prime Minister Raymond Barre curtailed the effort to build social housing, substituting large-scale subsidies toward the building of new houses for firsttime buyers. This encouraged a form of "white flight," as the departure of low-income white families led to a growing concentration of immigrants in social housing, in particular in the biggest estates far away from Paris. The priority given to affordable housing declined as many municipalities in the Paris region became afraid of attracting low-income populations or migrants. After 1982, housing policy further decentralized. Jacques Chirac, the conservative mayor of Paris, expanded reliance on market mechanisms and encouraged a shift to upper-class housing. Social housing was marginalized, falling to roughly 7 percent of housing in Paris. Municipalities gained control of land-use planning. Although left-wing municipalities were often keen to build more houses for low-income groups, elsewhere in the region nimbyism (not in my backyard) spread. Municipalities with middle-class constituencies and conservative officials worked to prevent the building of new housing.
In 2000, however, the political winds began to shift again, with renewed policy efforts to mitigate the impact of housing-price increases on the social A second policy change, led by centrist Minister Jean-Louis Borloo (after 2003), was the creation of the national agency for urban renovation (anru, Agence pour la Rénovation Urbaine). It sought to demolish and then restructure the massive housing estates, replacing them with affordable housing that would be more integrated into neighborhoods. anru undertook a massive public investmentabout €1 billion annually for twelve years -for four hundred neighborhoods in France (including 119 in the Paris Île-de-France region). Those projects supported the building of new social and affordable housing, to be integrated with economic development projects, new infrastructure, and expanded services.
Political and policy changes in Paris complemented these dramatic shifts in national policy. In 2001, the victory of a left-wing coalition led by Bertrand Delanoe ended twenty-four years of conservative municipal government. Delanoe was reelected in 2008 and succeeded by his former deputy Anne Hidalgo in 2014. Housing has been a top priority for the coalition. It introduced an ambitious program to build social and affordable housing, extending into more upper-middleclass neighborhoods. Paris is a very well-funded municipality, with a budget of nearly €10 billion for 2.2 million inhabitants and a powerful and capable administration. The Paris City Council also deploys powerful regulations to control land use and expropriate reluctant property owners where necessary.
The impact of these efforts has been notable, and the pace has accelerated. The Delanoe administration managed to build roughly five thousand new social or affordable dwellings per year; Mayor Hidalgo has increased the number to seven thousand a year (to which we can add three thousand in the private sector). As a reminder, this is equal to the number of social housing units being constructed annually in the entire state of California, a jurisdiction with almost twenty times as many inhabitants. All in all, the Paris City Council, backed by national policy, dramatically changed the housing situation. The percentage of social housing (with subsidized rents and strong rent-increase limitation) rose from 7 percent to 20 percent, and is scheduled to reach 25 percent by 2021. To meet this goal, all new housing projects are supposed to include a remarkable 60 percent share of social or affordable housing.
Finally, the encouraging developments within the core of traditional Paris were flanked by the creation of new institutional structures and initiatives for a "Grand Paris" encompassing the broader metro area of seven million people. While on the 2007 presidential campaign trail, the Conservative Party candidate Nicolas Sarkozy pointed to booming London and accused the city and region's transport, housing, and spatial planning policies of lacking a "grand vision" for the capital-city region. The ensuing political battle eventually led to three important developments: 1) a €40 billion investment in a large-scale regional metro system, Le Grand Paris Express; 2) the creation

Patrick Le Galès & Paul Pierson
of a new local authority, Métropole du Grand Paris, in 2017, designed to boost the supply of new housing and overcome the collective-action problems regarding housing and infrastructure that pervade metropolitan regions; and 3) the recruitment of financial institutions and private developers to design and cofinance largescale investment in housing within this urban core. The changing scale of Paris and the making of a Grand Paris are a major opportunity for new development. The massive expansion of transport infrastructure allows various actors to increase their investment beyond the city of Paris. 62 New policy instruments have been put forward to allow public-sector organizations to preemptively purchase land for future housing. The state and the municipalities have identified thirtythree strategic sites, many around new metro stations, on which to build two hundred thousand new dwellings in the coming decade.
Paris thus stands out as a distinctive case. French officials have marshaled national regulatory and financial resources, sought to coordinate local efforts, and, where necessary, reformed the division of labor among public authorities. All of these actions have been on a scale at least plausibly commensurate with the scope of the challenges facing agglomeration economies. To say this is not to ignore the obstacles or the weaknesses within the French response. Planning documents during the past two decades typically sought construction of seventy thousand units of new housing every year in the Paris Îlede-France region, but for many years, the realized totals remained under fortyfive thousand. However, the set of policies described above has led production to rise to about sixty thousand in 2015, and the tempo is increasing toward seventy thousand. All of the new schemes include substantial commitments to social and affordable housing.
But questions remain. The commitment to social and affordable housing is extensive, but still must be fully implemented. Risks for the most vulnerable remain evident. Fifteen percent of the population lives in poverty (less than €1,000 a month for one person). Almost 10 percent of all those living in the Paris region-about one million people-live in bad housing conditions. Rising prices have excluded increasing numbers from the private rental sector. New buyers are increasingly concentrated among the already privileged. They are less likely to have a mortgage (that is, often because they are drawing on inheritances). As in London, the rate of first-time homeowners has decreased by 20 percent between 1986 and 2016 (from 25 to 20 percent). 63 According to the National Institute of Statistics and Economic Studies, there are about thirty thousand homeless people in the Métropole du Grand Paris, and eight to ten thousand in Paris.
The new governance design for the Métropole du Grand Paris, intended to restrict nimbyism, remains an untested arrangement. The making of Le Grand Paris combines two objectives: to develop social, affordable, and inclusive housing, and to increase the attractiveness of Le Grand Paris. At a minimum, these goals stand in tension with each other. Public actors now have major resources and have developed important policies to combat housing-price increases, or attenuate some of the negative effects. Some groups, however, will be tempted to use those resources to increase the competitiveness of Paris, to get rid of poor populations, and to build for the very affluent, as in London.
Finally, the deeply ingrained policy commitments to house low-income groups and to maintain forms of social  64 France is not immune to the broad changes sweeping advanced capitalism. Power relations between private developers, private finance organizations, and local authorities are changing, with more expertise and financial capacity concentrated in the hands of major market actors. There are indeed uncertainties and reasons for concern that a shift in political coalitions and weakening of currently extensive policy commitments could lead to the dynamics on display elsewhere, promoting durable inequalities. Yet even noting these anxieties suggests the distinctiveness of the French experience. The fear, after all, is that possible events could eventually push Paris down a path that London, New York, and San Francisco have already traveled.
Housing is a crucial mechanism through which the dramatic changes associated with the rise of the knowledge economy, and its accompanying reconstruction of cities, generate new forms of durable inequality. Combined with changes in inheritance tax, exploding urban property values constitute a powerful form of inequality production and reproduction.
In the long run, the indirect impact of rising prices on inequality may be just as significant. The high and rising housing prices that characterize these settings are not just a generator and store of wealth. Increasingly, access to these assets becomes a prerequisite for opportunity. The concern is that only those who already have significant economic advantages can make effective use of what these urban agglomerations provide. It is getting harder and harder to get on the escalators present in high-income globalizing cities.
Thus, housing within these agglomerations provides a stark illustration of the broader theme of this volume: inequality is a process. Initial inequalities can trigger additional effects that intensify those original divisions. Housing costs become a divider between those who either already own or can bear the costs of ownership and everyone else. Sociologists have pointed to "the spatialization of class." 65 Today, there is growing evidence that these economic agglomerations can generate, absorb (through in-migration and gentrification), and then reproduce and expand enormous inequalities. Most dramatically, this operates in the astonishing concentration of income at the very top.
Housing unaffordability can come to dominate how people make critical choices about location. In the United States, dramatic differences in the cost of living are producing "movement to stagnation." 66 Geographic mobility in the United States is declining and mobility to the areas with the highest wages is increasingly difficult. Instead, populations flow to places with lower productivity and wages but much lower housing costs as well (the median home price in Houston, for example, is just one-fifth of the price in San Jose). Put differently, prohibitive housing costs cut off access to the escalator of high-productivity metro areas. 67 Two dynamics of exclusion are at work here: exclusion within and exclusion from. The radically new conditions in superstar cities operate not just as an escalator, but as a filter. Filtering doesn't just work by blocking in-migration by those with fewer resources; it operates within cities as well, as rising prices push individuals away from the escalator and into situations of insecure housing or homelessness. In many European cities, there are strong logics of differentiation and segregation at the extremes: geographer Sako Musterd and his colleagues have

Patrick Le Galès & Paul Pierson
provided evidence of increased segregation in European cities. Sociologists Douglas Massey and Jacob Rugh have found similar trends in the United States, with the affluent in American cities increasingly segregated, both racially and economically, from everyone else. 68 Those who find that the combination of very expensive housing and a weak market for their (low) skills renders life in these urban areas increasingly untenable are also pushed out. Indeed, absent immigration, San Francisco and New York would be losing population despite their booming job markets. Disadvantaged inhabitants exit to the declining industrial towns of New Jersey from New York, or to the Midlands and Northern England from London, or to the periphery of the Paris urban region, or far away from the coastal areas of California to less expensive but much poorer towns to the east or out of the state altogether.
The striking changes in superstar cities thus reveal shared patterns, highlighting the importance of global forces and technological shifts. Yet the contrast between Paris, on the one hand, and New York, San Francisco, and London, on the other, demonstrates the continuing significance of national institutions, political coalitions, and policies in shaping the new contours of inequality.
Both by comparative and historical standards, contemporary national programs to support affordable housing are extraordinarily weak in the United States. This structure of national weakness in urban policy reflects a national institutional framework of decentralized federalism that leaves an unusual amount of political authority in the hands of local officials. Much more than in most democracies, cities are forced to rely on their own resources.
In the current context, such self-reliance creates two massive problems. The first has long been well-understood. Localities are poorly placed to pursue aggressively redistributive policies. 69 Mobile companies and wealthy individuals find it both attractive and possible to evade such efforts if they become too burdensome. The dynamics of agglomeration, in which companies and skilled workers want to be in close proximity to other firms and workers, may mitigate this constraint to some extent. 70 In November 2018, San Francisco voters passed a controversial initiative, imposing significant taxes on large and profitable local firms in order to address the housing crisis. The initiative, if it survives legal challenges, will be an important test of this proposition. Redistributive initiatives that would promote affordable housing and combat homelessness must be primarily a regional or national responsibility.
Second, unconstrained by more overarching authorities, local politics creates powerful incentives for nimbyism. The politics are complex, and of course opponents of new construction rarely argue that they want to raise the value of their homes. In local politics, these incentives coexist with mobilization against developers stemming from concerns about neighborhood preservation and fear of gentrification, as well as environmental issues. Nonetheless, the massive contribution of zoning restrictions to rising property values, higher prices, and rising wealth inequality is very clear. 71 Existing property holders benefit from restrictive zoning rules that raise the value of their assets. Local political officials face incentives to curry favor with politically powerful blocs of existing homeowners, an incentive structure that has also been linked to the extraordinary levels of incarceration in the United States. 72 Regional or national actors are much better placed to limit such restrictive practices and encourage new construction. In the United States, however, such forces (especially national ones) have limited authority. Moreover, the strong rural bias built into American electoral institutions makes the urban housing crisis a very low priority for national officials. In short, one would be hard pressed to develop a structure of political authority less conducive to confronting the housing crisis than the one in place in the United States. New York and San Francisco show that some institutional arrangements and policy inheritances present formidable obstacles for confronting these new inequalities. By contrast, London's experience reminds us that choices remain to be made, and thus politics retains its vital importance. Acting on their own, London's public officials face many of the same constraints as their American counterparts. As the currents associated with the property boom push the disadvantaged down, or out, it is all city officials can do to slow the tide. But unlike the United States, British institutional arrangements concentrate relevant regulatory and fiscal resources with national authorities. The decision not to direct those resources toward coping with the housing affordability crisis has been a political one. In this respect, the contrast between Britain and France could hardly be starker. For decades, policy in the two countries has moved in opposite directions. The United Kingdom has largely abandoned its efforts to subsidize affordable housing and sustain broad access to its most thriving economy; France, by contrast, has redoubled its energy. In light of the highly distinctive social outcomes in London and Paris, the continuing centrality of these public decisions to the distribution of life chances is clear.