Resistance is not futile: rethinking ethnicity, tactics, and outcomes in civil conflicts

Under what circumstances do armed insurgents divert scarce resources to nonviolent tactics? Despite extensive research on the causes and consequences of civil wars and resistance campaigns, our understanding of when rebel organizations add nonviolent tactics to their repertoire is limited. Moving beyond traditional explanations of nonviolent resistance campaigns, we hypothesize that rebel organizations use nonviolent tactics–such as strikes and protest–to capitalize on successes on the battlefield. Rebels have many incentives to use nonviolent tactics, but their opportunity to do so typically depends on their ability to mobilize civilians. Military successes help pave the way for mobilization by shifting civilians’ expectations about the prospects for victory and repression. Using cross-national data on rebel nonviolent contention, including original data on nonviolent tactics in struggles for control over the central government, we find robust support for our argument. Probing potential scope conditions, we show that the association is strongest among rebel organizations with civilian wings and for rebels fighting for control over the government. Our findings provide new insight into the full array of tactics that insurgents deploy, beyond variations in armed activity. They contribute to our understanding of the many forms that contentious action can take, even amidst a backdrop of violence.

Given the appalling consequences of civil wars, why are the competing actors within a state unable to come to a settlement to avoid the costs of conflict? How might external parties affect the likelihood that a civil war begins? How do their actions affect the duration and outcome of civil conflicts that are already underway? This project draws on three main approaches—bargaining theory, signaling theory, and rational expectations—to examine how external actors might affect the onset, duration and outcome of civil wars. Signals from external actors are important because they represent a potential increase (or decrease) in fighting capabilities for the government or the opposition if a war were to begin. Costly signals should not affect the probability of civil war onset because they are readily observable ex ante, which allows the government and opposition to peacefully adjust their bargaining positions based on changes in relative capabilities. In contrast, cheap hostile signals make civil war more likely by increasing the risk that an opposition group overestimates its ability to stage a successful rebellion with external support. Cheap supportive signals work in the opposite manner because they represent increased fighting capabilities for the government. Furthermore, signals sent in the pre-war period have important implications for the duration and outcome of civil conflicts because competing intrastate actors develop expectations for future interventions prior to deciding to fight. Expected interventions should have little consequence for the duration and outcome of the conflict because they are endogenous to the pre-war bargaining positions. In contrast, unexpected

Freedom is not free: Examining health equity for racial and ethnic minoritized veterans.

Civil wars have occurred often in the post–World War II era. Their frequency of initiation decreased after the collapse of the Soviet Union in 1991, but the persistence of these conflicts meant that there was not a dramatic decline after the end of the Cold War. The causes of civil wars and their consequences for the stability of the international environment have, however, changed dramatically in the last two-and-a-half decades. During the Cold War, most civil wars were proxy battles between the Soviet Union and the United States; both superpowers were interested in maintaining regimes that were sympathetic to their side. The Soviet Union was never interested in the promotion of democratic regimes. The United States professed a commitment to democracy, but when faced with a choice between a Communist or even left-leaning democracy and an autocrat who aligned his state with the West, the United States chose the latter. The strongly positive statistical relationship between per capita income and democracy, which holds for most of the period between 1820 and 2000, disappears during the Cold War, when both superpowers were more interested in external alignment than in democracy.1The impact of civil wars on the stability of the international system has increased during the twenty-first century. September 11, 2001, marks a watershed because, for at least some observers in the advanced industrialized world, the ability of transnational terrorists to destroy two of the tallest buildings and kill thousands of people in the commercial center of the most powerful country in the world, as well as to fly a commercial airliner into the command center of the most powerful military (an event that one of us witnessed first-hand from inside the Pentagon and the other witnessed from the State Department across the Potomac River) represented a sea change in the extent to which developments in poor and remote countries could affect even the strongest and most powerful. September 11 created an urgency that was absent during the 1990s, when major powers believed that they could walk away from war-torn countries such as Somalia with limited consequences for their own polities.Greater urgency however, has not led to agreement, even in the academic world, on two critical issues: First, what are the potential threats to stability that might emanate from civil wars and weak governance in poor and remote areas of the world? Second, what policy instruments, if any, can be deployed to treat civil wars and reduce the downstream effects on other states and global order? There are no consensus answers to any of these fundamental issues.Rather than trying to identify some common ground, which we do not believe exists, we offer our own assessment of the consequences of civil wars, the nature of civil wars, and possible interventions that external actors might most effectively pursue. Our judgments have been informed by the essays in this issue of Dædalus and in the previous issue, but are not dictated by them.Civil wars can impact the wealthiest and most powerful countries in the world. The most consequential potential impacts are transnational terrorism and pandemic diseases, global crises that could be caused by intrastate conflict. Civil wars might also lead to large-scale migration, regional instability, and potential great-power conflict. And high levels of intrastate violence and loss of government control can often give rise to massive criminality, though this is most effectively addressed through domestic law enforcement rather than international initiatives.The nature of civil wars varies. The most important distinction is between civil strife that is caused by the material or political interests of the protagonists and civil strife that is caused by transnational ideological movements. The latter, if successful, might threaten regional stability and even the stability of the contemporary international system that is based on sovereign statehood. Transnational ideological movements, which in the contemporary world are almost all associated with particular versions of Islam, base legitimacy on the divine and reject both existing boundaries and secular authority. While transnational movements claiming divine authority are more threatening to the existing international order, it is very difficult for such movements to secure material resources. Institutions that control these resources, primarily states but also international organizations, NGOs, and multinational corporations, are manifestations of the extant global order. When combatants in civil wars are motivated by material incentives and accept the principles of the existing international order, then the “standard treatment” for addressing civil strife- UN peacekeeping plus some foreign assistance-is the most effective option if combatants believe that they are in a hurting stalemate, and if there is agreement among the major powers. If, however, combatants reject the existing order, then the standard treatment will not work.Finally, based on most, but not all of the essays in these two issues of Dædalus, the opportunities for external interveners are limited. Countries afflicted by civil strife cannot become Denmark or be placed on the road to Denmark; they cannot be transformed into prosperous democratic states. The best that external actors can hope for is adequate governance in which there is security, the provision of some services especially related to health and possibly education, and some limited economic growth. This is true whether the standard treatment is applied or if one side can win decisively. More ambitious projects aimed at consolidated democracy, sustained economic growth, and the elimination of corruption are mostly doomed to fail and can be counterproductive regardless of whether the combatants are interested in seizing control of an existing state or are motivated by some alternative, divine vision of how political life might be ordered. National political elites in countries afflicted with civil strife will be operating in limited-access, rent-seeking political orders in which staying in power is their primary objective. National elites will not accept accountability, legal-rational bureaucracies, or free and fair elections, all of which would threaten their power.The essays in these two issues of Dædalus and the literature more broadly identify six threats from civil strife that might directly impact the wealthy and more powerful polities of the world, or the nature of the postwar liberal international order. The first two-pandemic diseases and transnational terrorism-are potentially the most consequential, although neither poses the kind of existential threat presented by war among nuclear armed states.Pandemic diseases. As the essay by Paul Wise and Michele Barry points out, since 1940, some four hundred new diseases have emerged among human populations.2 Most of these diseases have been zoonoses: disease vectors that have jumped from animal populations, in which they may be benign, to human populations, in which they might cause serious illness. Most of these outbreaks have occurred in a belt near the equator, where human beings intermingle more closely with animals, such as bats and monkeys. The main impact of civil wars is, however, not in increasing the number of new diseases, but rather diminishing the capacities of health monitoring systems that could identify, isolate, and possibly treat new diseases. Effective detection requires constant monitoring, which is extremely difficult in areas that are afflicted by civil war. Epidemics, or at least disease outbreaks, are inevitable given the ways in which human beings impinge more and more on animal habitats, but allowing an epidemic to evolve into a pandemic is optional. If effective detection and monitoring are in place, a disease outbreak will not turn into a pandemic that could kill millions. So far, the world's population has been spared such an outbreak. If, however, a disease can be transmitted through the air, and if civil strife or something else prevents effective monitoring, the likelihood of a pandemic increases.Transnational terrorism. Terrorism, which in recent years has primarily, but not exclusively, been associated with Islamic jihadism, can arise in many different environments. At the time of the September 11 attacks, Al Qaeda and its leader Osama bin Laden were resident in Afghanistan, a very poor, land-locked country. Before that, Bin Laden had found refuge in Sudan. Most of the participants in the September 11 attack, however, were born in the heart of the Arab world, namely in Saudi Arabia, and had resided for a number of years in Germany. The perpetrators of the July 7 attacks on the mass transit system in London were Muslims of Somali and Eritrean origin, raised and schooled in the United Kingdom. The bomber, whose efforts to bring down an airliner headed for Detroit were frustrated by a courageous and alert passenger, was a Nigerian citizen who had spent time with jihadi ideologues in the Middle East. The attacks in Paris and Nice in 2015–2016 were carried out by individuals born in North Africa, but who had lived for many years in Western Europe. The murders of fourteen people in San Bernardino, California, were perpetrated by a U.S. citizen born in Chicago, whose parents were from Pakistan and who was educated at California State University, San Bernardino, and his wife, who was born in Pakistan but spent many years in Saudi Arabia. The massacre at the Orlando, Florida, night club in 2016 was carried out by the American-born son of a man who had emigrated from Afghanistan and had lived for many years in the United States.While terrorism associated with Islamic jihadism is hardly an exclusive product of safe havens in countries afflicted by civil strife or poor governance, the existence of such safe havens does, as Martha Crenshaw argues, exacerbate the problem.3 Safe havens are environments within which would-be terrorists can train over an extended period of time. A number of terrorists, even those raised in Western, industrialized countries, have taken advantage of such training. Transnational terrorist organizations might or might not secure weapons of mass destruction; they might or might not develop more effective training; their operatives might or might not be discovered by intelligence services in advanced industrialized democracies. Civil war and weak governance, however, increase the likelihood that transnational terrorist groups will find safe havens, and safe havens increase the likelihood of attacks that could kill large numbers of people.Global pandemics and transnational terrorism are the two most serious threats presented by civil wars. The probability that either will significantly undermine the security of materially well-off states is uncertain, but both are distinct sources of danger. Civil wars and weak governance increase the likelihood that large numbers of people could be killed by either threat. Neither is an existential threat, but both could have grave consequences for advanced industrialized democratic states. Hundreds of thousands or millions of people could die from a pandemic outbreak resulting from an easily transmissible disease vector or from a transnational terrorist attack that could involve dirty nuclear weapons, an actual nuclear weapon (still quite hard to obtain), or artificial biologics (increasingly easy to produce).Either a global pandemic or terrorist attack, possibly using weapons of mass destruction, would almost certainly lead to some constraints on the traditional freedoms that have been associated with liberal democratic societies.Migration, regional instability, and great-power conflict. Civil wars are also dangerous because they could lead to greater refugee flows, regional destabilization, and great-power conflict. Not every civil war has the potential for generating these global crises, but if generated, they would be a product not just of civil strife but also of policy choices that were made by advanced industrialized countries. In this regard, they should be contrasted with possible pandemics and transnational terrorism that, arguably, would occur regardless of the policies adopted by wealthy democratic states.As Sarah Lischer's essay shows, the number of migrants–especially people displaced by civil wars–has increased dramatically in recent years.4 Most of these migrants have been generated by three conflicts, those in Afghanistan, Syria, and Somalia. The wave of migrants entering Western Europe has destabilized traditional politics and contributed to the success of Brexit in the UK, the increased share of votes secured by right-wing parties in a number of Western European countries, and the electoral gains of a number of right-wing parties in Eastern Europe. Anxiety about immigration contributed to Donald Trump's victory in the United States. European countries, even those on the left like Sweden, have responded to rising numbers of refugees by tightening the rules for potential migrants. The European Union reached a deal with Turkey in 2016 to provide financial resources in exchange–among other things–for an increase in acceptance of refugees. At the same time, the sheer number of refugees in Jordan and Lebanon can potentially undermine government control in those countries.The impact of civil wars in one country can spread to surrounding areas. ISIL's ambitious campaigns have afflicted Syria and Iraq. Civil strife in Somalia has, as Seyoum Mesfin and Abdeta Beyene write, influenced the policies of Ethiopia.5 The FARC insurgency in Colombia impacted Venezuela and Ecuador. Conflict in the Democratic Republic of the Congo (DRC) drew in several neighboring states. Some regional conflicts have resulted in millions of deaths, most notably the war in the DRC, with limited impact on and attention from wealthy industrialized countries. Wars in the Middle East, however, have been more consequential because they have led to the involvement of Russia and the United States, they are closer to Europe and have therefore generated more refugees, and Middle Eastern oil is a global commodity on which much of the world depends. Regional destabilization in the Middle East does matter for the West; regional destabilization in Central Africa may only matter for those who live in the neighborhood.Direct confrontation between major powers has not occurred since the end of World War II. In well-governed areas, where civil wars are absent, the likelihood of great-power conflict is small. Territorial conquest has been delegitimized (though Russia's annexation of Crimea stands as a recent exception to this norm). The existence of nuclear weapons has removed uncertainty about the costs of a confrontation between nuclear-armed states with assured second-strike capability. Great-power confrontations are, however, more likely in areas that are afflicted by civil strife, because instability and appeals from local actors could draw in major state actors with vested interests. This is especially true for the Middle East. Moreover, in countries on the periphery of Russia that were formerly part of the Soviet Union, especially those with sizeable Russian ethnic populations, the government in Moscow has demonstrated that it can increase the level of internal unrest. There is no guarantee of stability, even in countries that might have been stable absent external support for dissident groups that would otherwise have remained quiescent.As Barry Posen suggests in his essay, multipolarity makes all aspects of external involvement in civil wars more fraught, including the possibility of a conflict among the major powers.6 In a multipolar world, no single pole is likely to be able to dictate outcomes to potential combatants. The possibility of a hurting stalemate declines because all sides hope that their fortunes could be resurrected by some outside power. Absent a hurting stalemate, which makes the standard treatment including UN Peacekeeping Operations (UN PKOs) and other forms of assistance attractive to major combatants, civil wars are more likely to continue. The contemporary international environment is more multipolar than was the case during the bipolarity of the Cold War or the unipolarity of the United States that lasted for a little over a decade after the Soviet Union collapsed. Managing civil wars will now be more difficult. The possibility of great-power conflict has increased. And because wars will prove harder to end, refugee flows will persist.Criminality. Criminality is a final area in which there may be some association between civil wars and weak governance, and the well-being of individuals in advanced industrialized countries. Because of the ease of transportation and communication, criminality is not limited to specific countries. Internet theft can originate from and impact many different countries. The loss of billions of dollars a year, drug smuggling, and human trafficking are familiar manifestations of transnational criminality. As Vanda Felbab-Brown writes, large-scale criminality can greatly exacerbate the challenges states face in defeating insurgencies and ending civil wars.7Addressing criminality associated with civil wars is fraught with difficulty. The association between criminal gangs and the state may be uncertain. National elites may protect criminal organizations. Some criminal organizations may generate revenues that help national elites stay in power. Yet while transnational criminality does affect individuals and institutions in the wealthier democracies, it is not a threat to their domestic political orders. The problem is best dealt with through national and international law enforcement.The most important conclusion that emerges from the discussions at the core of our project is that the policy options for addressing civil wars are limited. The essays in these two issues suggest that there are four factors that external actors must take into account when considering responses to intrastate warfare in weakly governed polities: the extent to which the interests of external and national political elites are complementary; the presence of irreconcilable groups in a civil conflict; the threat of great-power conflict; and the costs of intervention.Alignment of interests. Of these four factors, the greatest impediment to successful interventions is the misalignment of domestic and external elites' interests. Domestic elites governing an area afflicted by civil strife will be primarily interested in keeping themselves in power. The path to Denmark is paved with free and fair elections, rational-legal bureaucracies, and the rule of law, all of which are antithetical to the interests of those who hold power in closed-access or exclusive polities.The best that external actors can hope for is to bring some degree of security to areas that are afflicted with civil strife, which is easier to accomplish if none of the combatants are motivated by ideologies that cannot be reconciled, and if competing major or regional powers are not in proxy wars. even if irreconcilable and states are not part of a civil ambitious for and will fail because domestic elites are primarily interested in staying in not in and security assistance has been effective in a limited number of state institutions and the of civil but then only and only to some assistance might of but these are likely to or away when foreign assistance is the support of domestic external actors will fail to civil wars or effectively deal with from such of the world's especially polities by intrastate are rent-seeking states in which the political in power through foreign assistance and will not lead to into the night the number of votes the number of that political to stay in power. The in which the government is to but to within a is not the of almost all of human in almost all in the world, were and If individuals could the of the state they some external actors might be able to the incentives of national elites in the which this might are elites in countries by civil war are almost in what and political have an exclusive Their primary is to stay in power. This requires the and of of their support Most they must have command over of those who control the of violence that they cannot be in exclusive or rent-seeking orders are on the loss of and even life that would from a loss of will efforts for hold free and fair or to corruption as existential more like which are often sources for elites in exclusive might be as actors are only likely to have if domestic elites are on foreign as essay is often the and if external actors can threaten to which is often not the If domestic have sources of such as from or if the state is will not be able to threaten to assistance as government and have constraints were in Afghanistan, where the United States, billions of dollars in elections, and was to the of the rather than efforts to the fundamental of because such his The were because could not (though corruption on all to the of some from the of were by because the his and his in his essay, a for the that occur when the interests of external and internal actors are which will be the case when external actors to in rent-seeking on security that an effective national security at least effective in the of external is much harder than has been or As interests of domestic elites are often different from the interests of external The on power and domestic threats to their while the more on international or transnational threats that could their is, as a problem that cannot be the United States is most likely to provide security assistance to states that are governed if these polities were they would not external security In rent-seeking political will not the military as an the armed will be as as a potential that must be through some of and military in that to the of the A military of effectively in the is what in rent-seeking states do not As it would be almost for an external to such as with military or or which would be in the interests of national but not in the interests of external actors to an effective national military this the collapse of the in in one decade of U.S. military and billions of dollars of was hardly The United States that to effectively its ideological an that would not threaten and their on wars do not the that countries to stable polities and significantly the economic of large of the As with to the Middle East, the rent-seeking that were the conflict are likely to be during of civil is a that elites to those with they to stay in the of the external powers is the problem of to when the challenges of monitoring the of security assistance actors are not likely to be able to or even the interests and of actors in countries by civil may be may be power and their live in and for a while foreign and often for one at if the of the United States or other external is to help countries that have been afflicted with civil war consolidated democracy and there will be between the of domestic and foreign presence of and powers. If one or more of the major are or if two or more major powers have and interests conflict policy options to treat civil wars will be terrorism has been motivated primarily by ideological movements that reject the extant rules and of the global order. As the essays by and motivated have a that is antithetical to the almost of in the contemporary international the sovereign state The principles and associated with and international are to those that have been and by Islamic jihadi Islamic as points out, other authority is from not from some Islamic there is a fundamental distinction between the world of by Muslims and by Islamic law, and the of or where Islamic law is not to some of Islamic law, Islamic states can only with other Islamic with the world, are limited to the most contemporary of Islamic jihadi has that its is to a in the Middle East. a would state and the and rules of an secular who to a sovereign state the same to external powers that a on the of the contemporary international system and the of existing state cannot be with foreign assistance and they will not accept such the most policy option for those to the of the might be to war a As the armed were able to the while to the international order, from the of the may however, be to may have to be as the of Afghanistan and when with a a or parties are a more attractive as the misalignment of domestic and external interests has policy do the misalignment of major interests. The presence of major powers in a civil as can potentially threaten the security of as well as the international the

Disputes over self-determination (SD) have led to many civil wars, and a primary alternative, nonviolent campaign, is rarely successful in this context. Yet, while secession is rare, these movements often achieve more limited successes in the form of concessions from the state. This article provides a new assessment of the efficacy of different tactics – violent, nonviolent and conventional political action. It advances an argument that nonviolent tactics can help SD movements to generate indirect pressure on states that contributes to movement success, including greater autonomy. Nonviolence is used to garner attention from international actors using a human rights frame for their cause. International actors that are receptive to these human rights narratives then incentivize concessions for the SD movements and dis-incentivize repression. This happens through a variety of means, such as public shaming of the host state and withholding inter-governmental organization membership. Statistical analysis of violent and nonviolent tactics in SD movements shows that nonviolence can be effective in successfully obtaining concessions. Movements that use nonviolence are twice as likely to see concessions in any given year compared to those that just make demands but do not use nonviolence or violence. The effect of nonviolence is slightly larger than that of violence, which is also associated an increased chance of concession.

Do people from disadvantaged ethnic groups favor political violence over non-violent tactics? Studies of horizontal (between-group) inequality often concentrate on civil war. This article drills below the macro level and looks beyond civil war, to investigate individual participation in various types of conflict. Different types and combinations of ethnic disadvantage favor participation in different kinds of conflict, because of different opportunity structures. Political exclusion motivates leadership, which facilitates organized movements. Economic disadvantages restrict economic leverage, making non-violent tactics less likely to succeed. The article maps these components of groups’ opportunity structure onto different constellations of inequality. It uses Afrobarometer survey data ( N = 29,727) to show that economic disadvantages increase participation in political violence short of civil war. When they coincide with political exclusion, they also make people steer actively away from demonstrations. The evidence is less conclusive for political disadvantages alone, but points toward increased participation in demonstrations.

This paper reports the results of the first cross-national examination of the impact of the geography of nonviolent contention on regime transitions. Nonviolent tactics ‘work’ in part by signalling the preferences of non-participants through the symbolism of participants, unlike violent tactics. This opens the way for nonviolent campaigns to exploit variations in social-spatial meaning to enhance the informativeness of dissent. Capital cities are one such symbolic place and the main prediction of this study is a positive relationship between large protests and regime transitions in the capital, but not elsewhere. I also predict a strong direct relationship between the proximity to the capital of fighting in civil wars, and regime transitions; no relationship to the proximity of nonviolent contention; and that the intensity of violent conflict impacts regime transitions in a way that is largely independent of location. Results from an analysis of episodes of violent and nonviolent conflict from 1990 to 2014 generally support these contentions.

Civil wars are not only destructive: they can also generate new, long-lasting social, political, and economic structures and processes. To account for this productive potential and analyse post-conflict outcomes, I argue that we should analyse civil wars as critical junctures. Civil wars can relax structural constraints, opening opportunities for wartime processes to generate changes or to reinforce, rather than transform, the status quo. Changes or stasis may then be locked in by conflict outcomes, creating path dependencies. Studying civil wars as critical junctures allows for a clearer understanding of what variables mattered and interacted at different points in the conflict process, and the varying roles of structure and agency in producing institutional change or reinforcing pre-existing conditions. I explore the potential benefits of a critical juncture approach in the civil wars literature on different aspects of post-conflict politics and illustrate them in analysing the literature on women’s empowerment during and after civil wars. Applying the critical junctures framework to civil wars’ effects on institutions and socio-behavioural patterns can provide analytical clarity about complex processes and contexts, can facilitate comparison across cases and studies, and draws critical attention both to what civil wars change and to potential pathways not taken.

How does diffusion of civil war battles influence conflict termination? Recent advances in civil war literature have found that battle dynamics shape conflict termination by affecting the intra-conflict bargaining between disputants. This article extends the theoretical perspective and argues that how battles diffuse matters in determining conflict termination. While battlefield dynamics should in principle reveal previously unavailable private information, the relevance of information-revealing effect depends on the diffusion patterns of battles. The persistent, and possibly battle-exacerbated, commitment problem can also alter the prospects of conflict termination. We test the argument by distinguishing two distinct diffusion patterns of battles: distant and proximate. The empirical results reveal that distant diffusion, but not proximate diffusion, of battles makes civil conflicts less likely to terminate. The analysis also demonstrates that incorporating diffusion dynamics significantly improves our ability to predict conflict termination.

For several decades, the rate of civil war onset has outpaced the rate of conflict termination leading to a “relentless accumulation of ongoing civil wars” (Brandt). The deployment of commercial military actors (CMAs), i.e., private military and security companies and mercenaries, has been considered by some observers as an opportunity to put a halt to this trend, and end a war by writing a cheque. Others are more skeptical and consider CMAs to rather aggravate crises. The argument put forward in this investigation is that a cheque issued early in the conflict facilitates termination, while a later one does not. In the initial state of the conflict, contestants are unprepared to fight, lack capability, and are vulnerable to defeat. However, due to their own weakness, contestants are frequently unable to exploit the opportunity. A CMA intervention early in the conflict can remedy a client's weakness and enable exploitation of the opponent's weakness, and, in turn, conflict termination becomes more likely. The overall results show that CMA intervention within the first year of the conflict is associated with an increased risk of termination as opposed to later or no intervention. The investigation employs Cox proportional hazard models to test the influence of CMAs on conflict termination, and draws on our newly developed Commercial Military Actor Database, which includes PMSC and mercenary deployments in all civil wars between 1980 and 2016.

Improving access and quality for ethnic minority women— panel discussion

Introduction The residential neighbourhood is thought to affect residents because of presumed neighbourhood effects; the independent effects of a neighbourhood’s characteristics on the life chances of its residents. An enormous body of research has tried to measure neighbourhood effects, however, there are no clear conclusions on how much, if any, effect the neighbourhood has on its residents. There is non-random selection of people into neighbourhoods which causes a bias in the modelling of neighbourhood effects. Any correlation found between neighbourhood characteristics and individual outcomes might be explained by selection bias and can therefore not prove the existence of a causal neighbourhood effect. The question is; do poor neighbourhoods make people poor, or do poor people live in unattractive neighbourhoods because they cannot afford to live elsewhere (Cheshire, 2007). Therefore, insight in selection is important to gain more insight in neighbourhood effects (Van Ham and Manley, 2012). For neighbourhood effects research it is important to study selective mobility and neighbourhood choice and to combine neighbourhood effects research with neighbourhood selection research (Doff, 2010a; Van Ham and Manley, 2012; Van Ham et al., 2012; Galster, 2003; Hedman, 2011). The aim of this thesis therefore is to gain more insight in both the causes and the consequences of segregation and thus to study both individual residential mobility and neighbourhood selection and neighbourhood effects. Besides the neighbourhood effects literature, also the segregation literature will benefit from better insights in selective residential mobility because selective residential mobility is one of the main driving forces of segregation. There are two main research questions for this thesis. Firstly, I try to give insight in selective mobility and neighbourhood choice and thus to study where, when and why which people move. What is the effect of personal characteristics, neighbourhood characteristics and macro level housing market developments on individual neighbourhood satisfaction, moving wishes, moving behaviour and neighbourhood selection and on macro level selective mobility patterns and segregation? Secondly, I will test presumed neighbourhood effect mechanisms. Concentration areas of ethnic minorities are seen as undesirable, because their residents are thought to have less contact with the native majority which might hamper their integration and their life chances. It is, however, unclear to what extent social contact is affected by the residential neighbourhood. The second research question therefore asks whether ethnic minorities have less contact with the native majority if they live in minority concentration neighbourhoods. Segregation Segregation is defined as the population composition of neighbourhoods in relation to each other; that is, the concentration or underrepresentation of population groups in neighbourhoods compared to a city or national level average. Selective residential mobility is one of the main driving forces of segregation. Households move to a certain neighbourhood, either because they choose to live there, or because they are constrained in their choice options. Therefore segregation can be both voluntary and involuntary. This thesis focuses on selective residential mobility as cause of segregation, therefore it tries to understand why and where people move. According to residential mobility theory, personal characteristics determine residential preferences and if the residential situation is not in line with these preferences this will lead to dissatisfaction and a desire to move (Brown and Moore, 1970). Whether a dissatisfied household succeeds in moving to a dwelling and neighbourhood more in line with their preferences, depends on their personal resources and restrictions and macro level opportunities and constraints (Mulder and Hooimeijer, 1999). Successful households will move to a neighbourhood more in line with their preferences. There are, however, differences between households in which neighbourhoods are open to choice. Low income households will only be able to select neighbourhoods in which inexpensive dwellings are available. Similarly, households who depend on the social housing sector, or on the owner-occupied sector will only be able to select neighbourhoods where dwellings of this tenure are available. In addition, there are differences between households in which neighbourhood is (deemed) most attractive. People prefer to live among others who are similar to themselves and also facilities directed towards specific groups will make especially concentration neighbourhoods of the own ethnic or income group attractive. Besides residential preferences, also other factors will affect neighbourhood selection; population groups will differ in access to information on neighbourhood attractiveness or housing opportunities and discrimination, or fear of discrimination, can limit the opportunities of minority groups on the housing market. Neighbourhood effects It is typically assumed in European and American urban policy and academic research that spatial concentrations of low income households or ethnic minorities have negative effects on their inhabitants (Friedrichs et al., 2003). An enormous body of research has tried to measure neighbourhood effects; the independent effect of a neighbourhood on its residents when controlling for individual characteristics (see for a review Dietz, 2002; Ellen and Turner, 1997; Sharkey and Faber, 2014; Van Ham et al., 2012). The research attention for neighbourhood effects started with the seminal work of Wilson (1987). He argued that living in concentration areas of the jobless lowest class, isolated from role models, mainstream values and norms, and informal job networks and social contacts with employed, has a negative effect on your life chances. Neighbourhoods can affect their residents via a number of mechanisms (Ellen and Turner, 1997; Erbring and Young, 1979; Galster, 2012). Firstly, the geographical location determines job access and thereby labour market opportunities. Secondly, pollution, noise and disturbance affect health and (thereby) life chances via environmental mechanisms. In addition, neighbourhood stigmatisation can reduce life chances because others have prejudiced ideas and low expectations of the residents of stigmatised neighbourhoods. Also, the quality of institutions such as schools, museums, libraries and sport facilities will generally be lower in disadvantaged neighbourhoods, thereby reducing the life chances of residents. Finally, neighbourhood effects can transpire via social interactions with neighbours. Through collective socialisation and peer pressure people are thought to conform to local social norms (Jencks and Mayer, 1990). Positive role models and social network contacts with individuals with more social capital can help people advance in their work or educational career (Ellen and Turner, 1997). For ethnic minorities it can be important to have social interactions with the native majority to have the opportunity to learn the majority language, standards and values (Lazear, 1999) and to have bridging network ties that can provide access to valuable information not present within the own ethnic network (Buck, 2001). Many neighbourhood effect researchers believe that living in concentrated poverty has negative effects on individuals, and policymakers try to create mixed neighbourhoods to prevent these negative neighbourhood effects. However, although “it is perfectly plausible that poor people are made poorer by the characteristics of the neighbourhoods in which they live” (…) “a close examination of the best research available does not reveal any clear evidence to support it” (Cheshire, 2007: p. ix). Almost all neighbourhood effect studies struggle with selection bias (Cheshire, 2007). Any relation found between neighbourhood characteristics and individual outcomes might be a selection effect and therefore cannot prove the existence of a causal neighbourhood effect. Neighbourhood effects researchers have tried to reduce or eliminate selection bias. Firstly this is done by using quasi-experimental study designs, using households whose residential neighbourhood is determined by external factors (Sampson et al., 2002). Secondly, advanced statistical methods such as sibling studies, fixed effects studies, instrumental variables and propensity score matching are used to reduce selection bias or to control for selection (Harding, 2003). These advancements in methodology have improved our insight in selection bias and in neighbourhood effects, however, since there are no methods that can completely eliminate selection bias, there is still no clear evidence of causal neighbourhood effects. Generally, methods that apply more controls for selection bias find smaller neighbourhood effects. Instead of trying to eliminate selection bias, this thesis tries to provide insight in selection, in why and where which people move. In addition, this thesis tests presumed neighbourhood effect mechanisms. Social interactive mechanisms assume that neighbourhood effects transpire because the population composition of the residential neighbourhood affects with whom you interact (Ellen and Turner, 1997; Galster, 2012). Therefore I test whether the ethnic composition of the neighbourhood affects interethnic contact. Individual differences in determinants of residential satisfaction Residential satisfaction is a key variable in understanding individual residential mobility (Lu, 1999; Speare, 1974), as dissatisfaction leads to desires to move (Wolpert, 1965). Many researchers have studied the individual level and neighbourhood level determinants of residential satisfaction, however, very few have studied which neighbourhood characteristics are important to whom. People differ in which neighbourhood characteristics affect their residential satisfaction (Galster and Hesser, 1981). If certain neighbourhood characteristics lead to dissatisfaction and therefore to mobility desires for specific groups, this might lead to selective mobility and segregation. Therefore, in Chapter 2 I study individual differences in the determinants of residential satisfaction. I estimate ordered logit models explaining satisfaction on residents of urban areas within the Housing Research Netherlands 2012 survey. To test whether there are individual differences in the effects of neighbourhood characteristics on satisfaction I include interaction effects between individual characteristics and neighbourhood characteristics. These interaction effects test whether neighbourhood characteristics such as the neighbourhood ethnic composition, crime rates or dwelling values have similar effects on all individuals, or whether individual characteristics affect the size and direction of these effects. To my knowledge, previously only Greif (2015) and Parkes et al. (2002) have tested interaction effects between tenure and neighbourhood characteristics and there is no earlier research on ethnic or household differences in the determinants of neighbourhood satisfaction. Because there is almost no earlier research on individual differences in the determinants of satisfaction, I combine literature on residential satisfaction with literature on residential preferences, mobility desires and behaviour to create hypotheses about which neighbourhood characteristics are important to whom. In line with the literature, I find that the share of non-western minorities in the neighbourhood has a negative effect on neighbourhood satisfaction, an effect that is stronger for natives than for non-western minorities themselves. This can be explained by own group preferences; people are more satisfied in neighbourhoods with higher shares of their own ethnic group and when this is taken into account the differences between ethnic groups in the effect of the total neighbourhood share of non-western ethnic minorities on satisfaction disappear. Satisfaction is found to be more dependent on neighbourhood characteristics for owner-occupiers than for renters and more for households with children than for other households. However, while earlier research has found that owner-occupiers and households with children are especially sensitive to the neighbourhood ethnic composition (Ellen, 2000; Goyette et al., 2014; Greif, 2015; Xie and Zhou, 2012), I find that it is not the neighbourhood ethnic composition, but neighbourhood safety that is especially important for these groups. There are thus differences between ethnic groups, tenure groups and household types in the determinants of residential satisfaction. These differences might lead to selective mobility, segregation and high turnover rates. Policymakers in many countries try to create stable, attractive and mixed neighbourhoods (Bolt et al., 2010; Baum et al., 2009; Cheshire, 2007), also by attracting higher income households to deprived urban restructuring neighbourhoods (see Chapter 5). These insights in which neighbourhood characteristics are important to whom, are very important for effective policy design (Baum et al., 2009; Ellen et al., 2013; Pinkster et al., 2015). Ethnic differences in realising desires to leave the neighbourhood Residential dissatisfaction leads to mobility desires which could lead to residential mobility (Brown and Moore, 1970; Wolpert, 1965). Whether people realise their desire to move depends on their personal resources and restrictions (Mulder and Hooimeijer, 1999), there are thus individual differences in how successful people are in realising their desires to move. In Chapter 3, I focus on people who expressed a desire to leave their neighbourhood and study who realises this desire within two years and who manages to escape from poverty neighbourhoods or minority concentration neighbourhoods. To do this, I use a unique combination of survey data and register data. Cross-sectional survey data in which people are asked about their desire to leave the neighbourhood are merged with longitudinal register data on their subsequent residential mobility behaviour. This allows me to test if people with a desire to leave the neighbourhood actually do leave their neighbourhood within two years and which neighbourhoods they move to and from. Earlier research has found that ethnic minorities are less likely to leave ethnic minority concentration neighbourhoods (Bolt and Van Kempen, 2010; Pais et al., 2009; South and Crowder, 1998) and poverty neighbourhoods (Bolt and Van Kempen, 2003; Quillian, 2003; South et al., 2005; South and Crowder, 1997). It was, however, unclear whether this was explained by the fact that ethnic minorities less often want to leave these neighbourhoods, or whether they are less successful in leaving these neighbourhoods, also if they have a desire to leave. It is important to understand why there are ethnic differences in mobility patterns. If there are ethnic differences in mobility desires, this might lead to voluntary segregation. However, if certain (ethnic) groups are equally likely to want to leave certain neighbourhoods, but less successful than others in realising this desire, this indicates segregation is involuntary. I find that non-western ethnic minorities are less successful than natives in realising desires to leave their neighbourhood. In addition, they are found to be less likely than natives to escape from ethnic minority concentration neighbourhoods and poverty neighbourhoods, also if they have expressed a desire to leave their neighbourhood. Non-western ethnic minorities who realise a desire to leave their poverty or minority concentration neighbourhood, more often than natives, move to another poverty or minority concentration neighbourhood. In this chapter, I thus find ethnic selectivity in the realisation of mobility desires. These differences can lead to selective residential mobility and (involuntary) segregation. Neighbourhood selection of non-western ethnic minorities. Testing the own-group effects hypothesis using a conditional logit model Residential dissatisfaction will lead to a desire to move and people who realise their desire to move will select a new neighbourhood. Also in the selection of a destination neighbourhood there are differences between population groups. Neighbourhoods differ in population composition, amenities, dwelling availability and housing costs and population groups differ in resources, restrictions and preferences (Mulder and Hooimeijer, 1999), in their access to knowledge and opinions about neighbourhoods (Hedman, 2013) and in information about housing opportunities available to them (Bolt, 2001; Huff, 1986). In neighbourhood selection research, until now most studies characterise the neighbourhood based on a limited number of characteristics; they model the effect of personal characteristics on the probability to move to a poverty neighbourhood (Bolt and Van Kempen, 2003; Clark et al., 2006; Logan and Alba, 1993) or a minority concentration neighbourhood (Bråmå, 2006; Clark and Ledwith, 2007; Doff, 2010b; South and Crowder, 1998). However, in reality the selection of a neighbourhood will depend on multiple neighbourhood characteristics that are assessed simultaneously and in combination (Hedman et al., 2011). Ethnic minorities have been found to be more likely than natives to move to minority concentration neighbourhoods (Clark and Ledwith, 2007; Doff, 2010b; South and Crowder, 1998). However, this is not necessarily explained by the ethnic composition, also other neighbourhood characteristics correlated with ethnic composition might explain why especially ethnic minorities move to ethnic minority concentration neighbourhoods. In Chapter 4, I estimate the effect of various neighbourhood characteristics on neighbourhood selection of ethnic minority households. I use a conditional logit model, which allows me to simultaneously take into account multiple neighbourhood characteristics and thereby to distinguish the effect of the share of the own ethnic group, other ethnic minority groups and housing market characteristics on neighbourhood selection. Ethnic minorities are found to more often than others move to neighbourhoods with low dwelling values and high shares of social housing. These areas are often also ethnic minority concentration neighbourhoods, thus, housing market characteristics partly explain why ethnic minorities more often than others move to ethnic minority concentration neighbourhoods. Also when housing market characteristics are taken into account, I find evidence for own group effects; ethnic minorities are more likely to move to neighbourhoods with higher shares of their own ethnic group. Most likely, ethnic minorities select these neighbourhoods because they prefer to live among family or other own group members, and/or because they find a dwelling via their mono-ethnic network. This chapter focuses specifically on the four largest ethnic minority groups in the Netherlands. I find that for Surinamese and Antilleans the combination of housing market characteristics and own group effects explains why they more often than natives move to ethnic minority concentration neighbourhoods. Turks and Moroccans, however, are found to move more often to concentration neighbourhoods of ethnic minorities (other than their own ethnic group), also when housing market characteristics and own group effects are taken into account. Discrimination or fear of discrimination most likely explains why Turks and Moroccans are not willing or able to move to native majority concentration neighbourhoods. Mixed neighbourhoods; effects of urban restructuring and new housing development Many European countries use mixed housing policies to decrease the spatial concentration of low-income households. Within the Netherlands, large scale urban restructuring programs have been implemented in which inexpensive social rented dwellings in deprived neighbourhoods are demolished and replaced by more expensive and more often owner-occupied dwellings (Kleinhans, 2004). These urban restructuring programs have attempted to attract middle- and higher income households to deprived neighbourhoods. However, at the same time large numbers of expensive and mostly owner-occupied dwellings have been built on greenfield locations around the major cities. Urban restructuring programs might be less successful in attracting higher income households to deprived neighbourhoods when they have to compete with large scale greenfield development. In addition, greenfield development creates opportunities for relatively high income households to leave existing neighbourhoods, which will accelerate the process of selective outflow and income sorting and thereby increase the spatial concentration of low income households who are left behind. In Chapter 5 I study the effect of urban restructuring and new housing development on selective mobility patterns and income segregation. I compare three urban regions in the Netherlands with different patterns of urban restructuring and greenfield development. I use longitudinal register data to study income and income development of people who move to or from various neighbourhood types or to newly built dwellings and the effects of these selective mobility patterns on income segregation. I find that urban restructuring programs within deprived neighbourhoods are successful in attracting middle and higher income households, also when they have to compete with large scale greenfield development within the same urban region. Large scale greenfield development, however, leads to an outflow of relatively high income households from existing neighbourhoods. This outflow of higher income households leads to a further concentration of low income households in deprived neighbourhoods and an overall increase in residential income segregation. Residential segregation and interethnic contact in the Netherlands In Chapters 2 to 5 I study selective residential mobility and neighbourhood choice, while in Chapter 6 I study presumed neighbourhood effects mechanisms. According to the neighbourhood effects literature, one of the mechanisms through which neighbourhood effects transpire is via social interactions with neighbours (Ellen and Turner, 1997; Erbring and Young, 1979; Galster, 2012). Social interactions with natives provide ethnic minorities with the opportunity to learn the majority language, standards and values (Lazear, 1999), and with access to valuable information not present within the own ethnic network. Living in ethnic minority concentration neighbourhoods might reduce the opportunities for ethnic minorities to interact with natives and thereby hamper their integration and there life chances. Policymakers in many European countries therefore perceive concentrations of ethnic minorities as undesirable and try to create more mixed neighbourhoods (Bolt, 2009). It is, however, unclear to what extent the population composition of the residential neighbourhood determines social interactions, as people are found to increasingly have social contacts over larger areas (Boomkens, 2006). Therefore, in Chapter 6 I test whether the ethnic composition of the residential neighbourhood affects interethnic contact. I estimate a multilevel binary logistic regression model explaining whether or not ethnic minorities have contact with native Dutch people. This regression model includes both personal characteristics and neighbourhood characteristics including the share of native Dutch people in the neighbourhood. In earlier research (Gijsberts and Dagevos, 2005; Van der Laan Bouma-Doff, 2007) ethnic minorities have been found to have less contact with natives if the share of natives in the neighbourhood is lower, however, I find no effect of the neighbourhood ethnic composition on interethnic contact. Whether ethnic minorities have contact with the native majority is mainly explained by their individual characteristics such as educational level and household type. Also differences are found between ethnic minorities who live in the four largest cities -cities with high shares of ethnic minorities- and ethnic minorities in other cities with much lower shares of ethnic minorities. When these personal and regional characteristics are taken into account, the ethnic composition of the neighbourhood does no longer affect whether ethnic minorities have contact with the native majority. It is therefore unlikely that living in minority concentration neighbourhoods hampers life chances and integration of minorities via social interactive mechanisms. Ethnic residential segregation on neighbourhood level does not affect ethnic minorities’ social contact with the native majority and thus does not necessarily hamper integration and life chances of ethnic minorities. Conclusions: Selective mobility, segregation and neighbourhood effects The aim of this thesis is to gain more insight in both the causes and the consequences of segregation, through studying both individual residential mobility and neighbourhood selection and neighbourhood effects. Various authors have argued that selection bias is one of the main challenges in neighbourhood effects research (Harding, 2003; Sampson et al., 2002; Van Ham and Manley, 2012). It is not possible to completely eliminate selection bias from neighbourhood effects research, however, insight in selection will help to address selection bias (Manley and Van Ham, 2012; Van Ham and Manley, 2012; Winship and Mare, 1992). This thesis provides insight in both neighbourhood selection and neighbourhood effects and creates a link between these two fields of literature. It adds to the previous research as it studies selectivity in various aspects of the residential mobility process, thereby providing a more thorough insight in the causes of selective residential mobility and segregation. A central finding of this thesis is that there is non-random selection of people into neighbourhoods. Individual characteristics such as ethnicity, tenure, household type and income affect residential satisfaction, mobility preferences and behaviour and neighbourhood selection. Because of this non-random selection into neighbourhoods, a correlation found between neighbourhood characteristics and individual outcomes does not prove the existence of a neighbourhood effect. In this thesis I distinguish separate ethnic minority groups which allows me to decompose the causes of ethnic selective mobility. Both ethnic minorities and natives are less satisfied in neighbourhoods with higher shares of (other) ethnic minorities, however, ethnic minorities are more satisfied if the share of their own ethnic group in the neighbourhood is higher. Ethnic minorities thus prefer to live among their own ethnic group or close to ethnic specific facilities and these own group effects are found to partly explain why ethnic minorities more often than natives move to minority concentration neighbourhoods. However, not only preferences but also constraints due to housing market characteristics or discrimination cause ethnic minorities to move to ethnic minority concentration neighbourhoods. In addition, ethnic minorities are found to be less successful in realising their desires to leave their neighbourhood. Ethnic residential segregation is thus partly voluntary and partly involuntary. Besides ethnic selectivity, I also found selectivity in the residential mobility process with regard to household type, tenure and income. Residential mobility is selective with regard to income because higher income households are more successful than lower income households in realising residential preferences, not necessarily because their preferences are different. Household type and tenure are found to affect both residential preferences and the ability to realise these preferences. Neighbourhood effects can transpire via a number of presumed mechanisms. To provide a better insight in neighbourhood effects, it is important to study these mechanisms. (Andersson and Musterd, 2010). Social interactive mechanisms assume that neighbourhood effects transpire because the population composition of the residential neighbourhood affects with whom you interact (Ellen and Turner, 1997; Galster, 2012). However, this thesis shows that the ethnic composition of the residential neighbourhood does not affect whether ethnic minorities have contact with the native majority population. It is therefore unlikely that living in minority concentration neighbourhoods hampers life chances and integration of minorities via social interactive mechanisms. Directions for further research In further research, firstly, it is important to gain a better understanding of the potential mechanisms through which neighbourhood effects transpire. Social interactive mechanisms assume that (neighbourhood effects transpire because) the residential neighbourhood affects your social network and social contacts. In this thesis, I found that ethnic segregation on the scale of the residential neighbourhood does not affect whether working age ethnic minorities have social contacts with natives. It is therefore unlikely that, for this group and on this scale, neighbourhood effects transpire via social interactive mechanisms. More research on segregation on different spatial scales and on different population groups could give insight in when, where and for whom neighbourhoods affect social interactions and thus under which circumstances neighbourhood effects can possibly transpire via social interactive mechanisms. Besides social interactive mechanisms, neighbourhoods are also expected to transpire via job access, stigmatisation, the quality of local services and institutions and environmental mechanisms. Also for these mechanisms it is important to derive clear hypotheses about how the neighbourhood affects its residents and to subsequently test these hypotheses. Further research could for instance test if people in neighbourhoods with lower accessibility of jobs are more often unemployed, or if employers prefer employees from ‘good’ neighbourhoods over equally qualified ones from stigmatised neighbourhoods. Different neighbourhood effects mechanisms will work on different neighbourhood scales, be important for different groups of people, after different times of exposure to different neighbourhood conditions. Research that explicitly tests whether, for whom and under which circumstance these presumed mechanisms are at work, can provide insight in how, when, where and for whom the residential neighbourhood can possibly affect its residents. Secondly, neighbourhood effects research would benefit from more research actually trying to understand neighbourhood selection. We need to both empirically and theoretically link neighbourhood selection research to neighbourhood effects research. Empirically, selection research can be linked to neighbourhood effects research by incorporating models of selection into neighbourhood effects studies. Although models incorporating selection will not be able to completely eliminate selection bias from neighbourhood effects research, such research can show how incorporating selection affects the outcomes of neighbourhood effects models and thus give insight in the effects of selection bias. Theoretically, it is important to understand selective residential mobility and neighbourhood choice and to create a theory of selection bias. A theory of selection bias should explain how and why which factors affect both neighbourhood selection and individual outcomes (Van Ham and Manley, 2012). Such a theory could be used to design quasi-experimental studies, to invent new controls for selection bias, or to argue to what extent outcomes from neighbourhood effects studies are biased. This thesis provided some first ideas of what should be included in a theory of selection bias, but more research is needed. Differences in opportunities, differences in residential preferences and differences in access to information lead to individual differences in residential mobility decisions and outcomes and thus to selection bias. More insight is needed in why preferences and opportunities are different and in individual search strategies and decision-making processes; why do some individuals accept a certain dwelling in a certain neighbourhood while others continue searching for better housing opportunities? Policy implications This thesis finds that segregation is partly voluntary, caused be preferences to live among similar people and partly involuntary, caused by group differences in constraints induced by housing market characteristics or discrimination. To the extent that segregation is voluntary it will be neither possible nor useful to create stable mixed neighbourhoods (Cheshire, 2007). However, involuntary segregation can be, and has to be, addressed by policy-makers. Policies that reduce constraints and increase the options for households to move to a neighbourhood of their preference can reduce involuntary segregation and increase residential satisfaction. These policies, however, do not necessarily lead to more mixed neighbourhoods as people might use their increased freedom of neighbourhood choice to move close to similar people. Social interactions between people of various ethnic and socio-economic groups are important for emancipation and integration and to prevent segregated and separated worlds that can lead to fear and exclusion. However, as the population composition of the neighbourhood does not necessarily determine with whom people interact, creating mixed neighbourhoods is necessary nor sufficient to promote social integration. Other policy efforts that promote social contacts between various ethnic and socio-economic groups remain necessary. This thesis finds no neighbourhood effect of the ethnic composition of the residential neighbourhood on whether working age ethnic minorities in the Netherlands have contact with native Dutch people. However, working age people leave their small residential area on a daily basis and within the Netherlands, most high ethnic minority concentration neighbourhoods contain relatively high shares of natives. This level of segregation, on this spatial scale, is found to have no neighbourhood effect on contact. If, however, larger areas would become concentrations of very high shares of deprived households or ethnic minorities, residents will no longer have opportunities to meet and interact with more resourceful people or with the native majority. Therefore, continuing policy attention is needed to prevent high levels of segregation at larger spatial scales.

Introduction The residential neighbourhood is thought to affect residents because of presumed neighbourhood effects; the independent effects of a neighbourhood’s characteristics on the life chances of its residents. An enormous body of research has tried to measure neighbourhood effects, however, there are no clear conclusions on how much, if any, effect the neighbourhood has on its residents. There is non-random selection of people into neighbourhoods which causes a bias in the modelling of neighbourhood effects. Any correlation found between neighbourhood characteristics and individual outcomes might be explained by selection bias and can therefore not prove the existence of a causal neighbourhood effect. The question is; do poor neighbourhoods make people poor, or do poor people live in unattractive neighbourhoods because they cannot afford to live elsewhere (Cheshire, 2007). Therefore, insight in selection is important to gain more insight in neighbourhood effects (Van Ham and Manley, 2012). For neighbourhood effects research it is important to study selective mobility and neighbourhood choice and to combine neighbourhood effects research with neighbourhood selection research (Doff, 2010a; Van Ham and Manley, 2012; Van Ham et al., 2012; Galster, 2003; Hedman, 2011). The aim of this thesis therefore is to gain more insight in both the causes and the consequences of segregation and thus to study both individual residential mobility and neighbourhood selection and neighbourhood effects. Besides the neighbourhood effects literature, also the segregation literature will benefit from better insights in selective residential mobility because selective residential mobility is one of the main driving forces of segregation. There are two main research questions for this thesis. Firstly, I try to give insight in selective mobility and neighbourhood choice and thus to study where, when and why which people move. What is the effect of personal characteristics, neighbourhood characteristics and macro level housing market developments on individual neighbourhood satisfaction, moving wishes, moving behaviour and neighbourhood selection and on macro level selective mobility patterns and segregation? Secondly, I will test presumed neighbourhood effect mechanisms. Concentration areas of ethnic minorities are seen as undesirable, because their residents are thought to have less contact with the native majority which might hamper their integration and their life chances. It is, however, unclear to what extent social contact is affected by the residential neighbourhood. The second research question therefore asks whether ethnic minorities have less contact with the native majority if they live in minority concentration neighbourhoods. Segregation Segregation is defined as the population composition of neighbourhoods in relation to each other; that is, the concentration or underrepresentation of population groups in neighbourhoods compared to a city or national level average. Selective residential mobility is one of the main driving forces of segregation. Households move to a certain neighbourhood, either because they choose to live there, or because they are constrained in their choice options. Therefore segregation can be both voluntary and involuntary. This thesis focuses on selective residential mobility as cause of segregation, therefore it tries to understand why and where people move. According to residential mobility theory, personal characteristics determine residential preferences and if the residential situation is not in line with these preferences this will lead to dissatisfaction and a desire to move (Brown and Moore, 1970). Whether a dissatisfied household succeeds in moving to a dwelling and neighbourhood more in line with their preferences, depends on their personal resources and restrictions and macro level opportunities and constraints (Mulder and Hooimeijer, 1999). Successful households will move to a neighbourhood more in line with their preferences. There are, however, differences between households in which neighbourhoods are open to choice. Low income households will only be able to select neighbourhoods in which inexpensive dwellings are available. Similarly, households who depend on the social housing sector, or on the owner-occupied sector will only be able to select neighbourhoods where dwellings of this tenure are available. In addition, there are differences between households in which neighbourhood is (deemed) most attractive. People prefer to live among others who are similar to themselves and also facilities directed towards specific groups will make especially concentration neighbourhoods of the own ethnic or income group attractive. Besides residential preferences, also other factors will affect neighbourhood selection; population groups will differ in access to information on neighbourhood attractiveness or housing opportunities and discrimination, or fear of discrimination, can limit the opportunities of minority groups on the housing market. Neighbourhood effects It is typically assumed in European and American urban policy and academic research that spatial concentrations of low income households or ethnic minorities have negative effects on their inhabitants (Friedrichs et al., 2003). An enormous body of research has tried to measure neighbourhood effects; the independent effect of a neighbourhood on its residents when controlling for individual characteristics (see for a review Dietz, 2002; Ellen and Turner, 1997; Sharkey and Faber, 2014; Van Ham et al., 2012). The research attention for neighbourhood effects started with the seminal work of Wilson (1987). He argued that living in concentration areas of the jobless lowest class, isolated from role models, mainstream values and norms, and informal job networks and social contacts with employed, has a negative effect on your life chances. Neighbourhoods can affect their residents via a number of mechanisms (Ellen and Turner, 1997; Erbring and Young, 1979; Galster, 2012). Firstly, the geographical location determines job access and thereby labour market opportunities. Secondly, pollution, noise and disturbance affect health and (thereby) life chances via environmental mechanisms . In addition, neighbourhood stigmatisation can reduce life chances because others have prejudiced ideas and low expectations of the residents of stigmatised neighbourhoods. Also, the quality of institutions such as schools, museums, libraries and sport facilities will generally be lower in disadvantaged neighbourhoods, thereby reducing the life chances of residents. Finally, neighbourhood effects can transpire via social interactions with neighbours. Through collective socialisation and peer pressure people are thought to conform to local social norms (Jencks and Mayer, 1990). Positive role models and social network contacts with individuals with more social capital can help people advance in their work or educational career (Ellen and Turner, 1997). For ethnic minorities it can be important to have social interactions with the native majority to have the opportunity to learn the majority language, standards and values (Lazear, 1999) and to have bridging network ties that can provide access to valuable information not present within the own ethnic network (Buck, 2001). Many neighbourhood effect researchers believe that living in concentrated poverty has negative effects on individuals, and policymakers try to create mixed neighbourhoods to prevent these negative neighbourhood effects. However, although “it is perfectly plausible that poor people are made poorer by the characteristics of the neighbourhoods in which they live” (…) “a close examination of the best research available does not reveal any clear evidence to support it” (Cheshire, 2007: p. ix). Almost all neighbourhood effect studies struggle with selection bias (Cheshire, 2007). Any relation found between neighbourhood characteristics and individual outcomes might be a selection effect and therefore cannot prove the existence of a causal neighbourhood effect. Neighbourhood effects researchers have tried to reduce or eliminate selection bias. Firstly this is done by using quasi-experimental study designs, using households whose residential neighbourhood is determined by external factors (Sampson et al., 2002). Secondly, advanced statistical methods such as sibling studies, fixed effects studies, instrumental variables and propensity score matching are used to reduce selection bias or to control for selection (Harding, 2003). These advancements in methodology have improved our insight in selection bias and in neighbourhood effects, however, since there are no methods that can completely eliminate selection bias, there is still no clear evidence of causal neighbourhood effects. Generally, methods that apply more controls for selection bias find smaller neighbourhood effects. Instead of trying to eliminate selection bias, this thesis tries to provide insight in selection, in why and where which people move. In addition, this thesis tests presumed neighbourhood effect mechanisms. Social interactive mechanisms assume that neighbourhood effects transpire because the population composition of the residential neighbourhood affects with whom you interact (Ellen and Turner, 1997; Galster, 2012). Therefore I test whether the ethnic composition of the neighbourhood affects interethnic contact. Individual differences in determinants of residential satisfaction<

Selective mobility, segregation and neighbourhood effects

SUMMARY This paper empirically shows that the imbalance between an ethnic group’s political and military power is crucial to understanding the likelihood that such a group engages in a conflict. We develop a novel measure of a group’s military power by combining machine-learning techniques with rich data on ethnic group characteristics and outcomes of civil conflicts in Africa and the Middle East. We couple this measure with available indicators of an ethnic group’s political power as well as with a novel proxy based on information about the ethnicity of cabinet members. We find that groups characterized by a higher mismatch between military and political power are between 30% and 50% more likely to engage in a conflict against their government depending on the specification used. We also find that the effects of power mismatch are non-linear, which is in agreement with the predictions of a simple model that accounts for the cost of conflict. Moreover, our results suggest that high-mismatched groups are typically involved in larger and centrist conflicts. The policy implication is that power-sharing recommendations and institutional design policies for peace should consider primarily the reduction of power mismatches between relevant groups, rather than focusing exclusively on equalizing political power in isolation.