Rotterdam suffers from a ‘second-city syndrome’. In many countries of the world the largest two cities are natural rivals, even though that rivalry is experienced more strongly in the second city than in the first one. And indeed, inside the Netherlands Rotterdam (population 635,000) tends to look much more often at Amsterdam (population 835,000) than vice versa. At times, the two cities see each other as rivals: who will have the National Photo Museum, or host the Olympics (if they will ever be granted to the Netherlands)? Rotterdam’s Feyenoord and Amsterdam’s Ajax are legendary opponents in the national football league. At other times the relationship between the two is more of a joking nature, for example when Rotterdamers do not wish to pronounce the name of the Dutch capital city, and call it by its area code ‘020’ instead.

A major characteristic of both cities is that each of them claims to be very different from the other, even though they are only 60 km or a good half-hour train ride apart. In this chapter I will explore to what extent they indeed are different, focussing, in line with the theme of this volume, on how immigration has impacted on the two. What immigrants have they received, how have these immigrants found their way in the fabric of both cities, and how have the cities responded to these influxes? As we will see, there are similarities, possibly even more than the rivals may tend to believe, but significant differences also exist. These relate to the composition of the immigrant flows, including the more recent ones, to the situation of the immigrants, but also to either city’s economic, social and political infrastructure. Most data in this chapter were collected by the Statistical Offices of both cities for a comparative project in which I took part in 2012. It was the first detailed comparative exercise of this type related to immigration and immigrant integration.Footnote 1 Interestingly, and perhaps also surprisingly, many of the data that were previously available were not directly comparable, for example because different categorisations had been used, or because survey questions were phrased differently. Unfortunately, there has been no follow-up to this project so far, which explains why some of the comparative data used in this chapter are not as recent as one might wish.

1 A Tale of Two Cities

Amsterdam is a centuries-old trading centre that grew organically, and is now a hub for commerce and financial services. It boasts a large, historical city centre with a rich cultural life, which serves as a major tourist attraction. Rotterdam is a port and industrial city that underwent rapid growth just in the last 150 years. The city’s centre was bombed by the Nazis in the Second World War, and then completely rebuilt. Since that time Rotterdam has taken on the allure of a modern world city – particularly with respect to its architecture. However, this world city has a relatively unilateral economic structure that constantly threatens to become obsolete.

Despite these substantial differences, the two cities also have much in common. Both have their long tradition of immigration to thank for their growth and prosperity. At the end of the Dutch Golden Age, around 1700, 40% of Amsterdam’s population had been born abroad. The seeds of Amsterdam’s wealth were largely laid down by Antwerp Protestants, French Huguenots and Portuguese Jews. In later centuries, the percentage of immigrants gradually receded. However, it has been growing again over the past few decades (Lucassen and Lucassen 2011). At the moment, a quarter of Amsterdam’s population is foreign-born. If the children of those foreign-born residents are included in the count, it appears that just over half of Amsterdam’s population belongs to either a first or second generation of immigrants. Unlike in 1700, these immigrants have not only come from neighbouring countries, but from all over the world. And no wonder: distance plays a far less important role today than it did in the past.

Rotterdam’s immigration tradition is much more recent. It is only since its advent as a port and industrial city at the end of the nineteenth century that Rotterdam has experienced large-scale immigration. Initially, immigrants mainly came from the rural areas of the south of the Netherlands, but in the last 50 years they originated from a large number of countries, in particular Suriname, Turkey, Morocco and, more recently, Poland. Percentage-wise, as large a share of Rotterdam’s population consists of immigrants as is the case in Amsterdam; nevertheless, there are definite differences between the immigrant populations of both cities when it comes to important aspects like origin and educational level.

2 Demographic Essentials

To the superficial observer the immigrant situation in Amsterdam appears rather similar to that of Rotterdam. The percentage of residents with an immigrant background is almost the same in both cities. On 1 January 2016, just under half (48.3%) of the population of Amsterdam were native Dutch according to the definition set out by Statistics Netherlands (CBS), namely that the person in question and both of his/her parents were born in the Netherlands.Footnote 2 This percentage was a fraction higher in Rotterdam, at 50.2%. However, since 2000 the portion of native Dutch in Rotterdam’s population has been declining faster than that of Amsterdam. In that year, 60% of Rotterdam’s population were still native, compared to 54.7% in Amsterdam. If the populations of both cities continue to develop in a similar manner, Rotterdam will soon overtake Amsterdam – and The Hague – as the Dutch municipality with the greatest proportion of immigrants.

Who are the immigrants in both cities? We will focus here on the largest four communities of non-Western origin in each of the two cities: Surinamese, Antilleans, Turks and Moroccans.Footnote 3 At first glance the differences between the cities seem to be relatively small, but they do exist. For years, Surinamese formed the largest immigrant community in both Amsterdam and Rotterdam, until they were surpassed by Moroccans in Amsterdam in 2010. Although Amsterdam in particular has an image of a city where many Surinamese settled down, the proportion of this immigrant population is almost identical in both cities: 9.0% in Amsterdam, 8.9% in Rotterdam.Footnote 4 Moreover, the number of Surinamese in Rotterdam has increased while it has decreased in Amsterdam. This is most likely due to the fact that the formation of a Surinamese middle-class in Amsterdam has further advanced than it has in Rotterdam. And it is among the middle-classes in particular that we see a large exodus from the city. In contrast, Rotterdam traditionally has more Antilleans than Amsterdam, both in percentage terms of the total urban population (3.6% compared to 1.5%) and in absolute numbers (21,000 compared to 11,500). Evidently, many newcomers like to settle in close proximity to the fellow country members preceding them.

The two other large immigrant groups, the Turkish and the Moroccans, are nearly equal in terms of size, at least when they are considered together: in 2010, Rotterdam’s population was 14.5% Turkish or Moroccan, in Amsterdam it was 14.3%. But the distribution between the groups differed starkly: More Turkish live in Rotterdam than Moroccans (47,000 compared to 39,000) while Amsterdam has 40,000 Turkish and nearly 70,000 Moroccans. This is most likely attributable to the fact that, half a century ago, at the time of the recruitment of migrant workers, Amsterdam businesses had a preference for those from Morocco, while Rotterdam businesses mainly focused on those from Turkey. No research has ever been done into the reasons for this. Unlike the Surinamese, there is not yet a decline in the growth of the Turkish and Moroccan populations, let alone in their absolute numbers – a sign that the formation of a middle-class and the subsequent move to the suburbs is not as advanced amongst these immigrant groups. There is certainly a trend of moving house amongst the Turkish and Moroccans, but they tend, as yet, to stay within the city limits, moving from the old late nineteenth and early twentieth-century neighbourhoods to the neighbourhoods that were built in the decades after the Second World War.

3 Patterns of Settlement and Segregation

Thus, we see that in Amsterdam, between 2000 and 2010, the strongest growth, in percentage terms, of non-Western immigrants occurs in the predominantly post-war boroughs of Nieuw-West (from 37% to 49%) and Noord (from 27% to 36%). In contrast, the older boroughs of West and Oost in this period show a slight decline in their non-Western immigrant population: West from 34% to 33% and Oost from 36% to 34%. The city as a whole, however, experienced an increase in its non-Western population – from 31% to 35%. A similar development took place in Rotterdam, where the boroughs of Charlois (from 33% to 46%), IJsselmonde (from 20% to 34%) and Prins Alexander (from 12% to 20%), built wholly or partially after the Second World War, grew relatively quickly. Boroughs with much older buildings and a traditionally large immigrant population grew more slowly: Delfshaven from 57% to 60% and Feijenoord from 50% to 57%. During the first decade of this century, the number of non-Western immigrants in Rotterdam’s total population increased from 30% to 37%. Rotterdam has since passed Amsterdam in this respect.

The migration from the older to the post-war neighbourhoods does not mean that there is an automatic decrease in the segregation of immigrants. It is worth noting that this is the case in Rotterdam, but not in Amsterdam. For example, the segregation index for the Turkish in Rotterdam decreased from 48 to 38 between 2000 and 2010, while it increased from 41 to 45 for the Turkish in Amsterdam. The trends are similar for the Moroccans: a decrease from 43 to 35 in Rotterdam and an increase from 39 to 42 in Amsterdam. The segregation index indicates what percentage of a specific group of a population would have to move to another neighbourhood in order to reach a perfectly proportional distribution of that population throughout the entire city. Amongst the native Dutch population in Amsterdam we also see an increase in segregation, while in Rotterdam there is a (slight) decrease. However, the native Dutch population of Amsterdam is still significantly less segregated than that of Rotterdam. In Rotterdam, 45% of the native Dutch population would have to move in order to achieve a proportional distribution throughout the city; only 27% would need to in Amsterdam.

4 Shifting Immigration Flows

The most notable differences between the immigrant populations of Amsterdam and Rotterdam are among those who do not belong to one of the largest four groups (Turkish, Moroccans, Surinamese and Antilleans). Within this ‘residual category’ a distinction is made in the population statistics between the ‘Western’ and ‘other non-Western’ immigrants. In 2010, 14.9% of the residents of Amsterdam were Western immigrants; in Rotterdam the percentage was 10.8. Strikingly, this population category grew much faster in the first decade of this century in Amsterdam than in Rotterdam (3 percentage points in Amsterdam compared with 1.3 in Rotterdam). In that same period, however, the category ‘other non-Western immigrants’ grew much faster in Rotterdam: from 7.5% to 10%, while Amsterdam only showed an increase from 9.9% to 10.1%. One may suspect a connection here with differences in the demand for labour, and perhaps in the general power of attraction between the two cities, but more on that later. One should not, however, assume without question that the Western immigrants remain mainly at the top of the labour market and the non-Western immigrants largely on the bottom. For example, all EU citizens, including the Polish, Romanians and Bulgarians are counted as Western immigrants. Many of them perform low-skilled labour. Conversely, the highly skilled knowledge migrants that originate from, amongst others, countries like India and China belong to the non-Western immigrant group. This provides a first indication of the declining usefulness of the classification criteria commonly used in Dutch immigration statistics (Ham and Van der Meer 2012).

Furthermore, we see in both cities that the first generation among all groups is decreasing in size in relative terms, while the second generation is growing. In fact, the growth in the total size of the ‘classic’ immigrant groups is solely the result of the increase in the second generation. New immigration – for instance, family migration – has nearly come to a halt in the ‘classic groups’; national figures also back up this pattern and the expectation is that it will remain so for these groups (Nicolaas et al. 2011).

This increases the average length of time the ‘classic’ immigrant groups have been established. The average Moroccan resident in Amsterdam has now been living longer in the city than the average native Dutch person. When it comes to the Western immigrants, we see just the opposite: the most recent immigrants include more and more people with a Western background. Many of them appear to leave again quickly, often also leaving the Netherlands. In general, immigrants from Western countries exhibit a higher mobility than those that have come here from non-Western countries. Thus, one can assume that slowly changing patterns of migration in the long run will lead to fewer immigrants taking up permanent residence in the Netherlands than was the case in recent decades (Entzinger 2014). There is no reason to believe that this is different in the two largest cities from anywhere else in the country. However, one may expect a relatively high proportion of the newcomers to the Netherlands to settle down initially in Amsterdam or Rotterdam. This has been the case for many years. All over Europe newcomers demonstrate a strong preference for settling in a metropolitan environment. Fellow countrymen often live there, there is often a package of provisions that matches their needs and there are better opportunities to earn an income.

5 The Educational and Employment Situation

Rotterdam’s population, on average, has a much lower education than Amsterdam’s. This is the case not only for the native Dutch, but also for their immigrant populations. In 2008, in Rotterdam the ratio between highly educated native Dutch (Bachelor’s degree or higher) and native Dutch with a low education (maximum lower professional education or vmbo-plus) was 1 to 1 (30% versus 31%). In Amsterdam, by contrast, there were almost three highly educated native Dutch for every native Dutch with a low education (48% versus 18%). Also in 2008, 56% of the non-Western immigrants in Rotterdam had a low education, compared to 41% in Amsterdam. Only 11% of Rotterdam’s non-Western immigrant population was highly educated, against Amsterdam’s 23%. So, Rotterdam housed five low educated non-Western immigrants for each non-Western immigrant with a higher education, while the ratio in Amsterdam was roughly 2 to 1.

The substantial differences between native Dutch and immigrants in terms of education somewhat obscure the fact that, over the past two decades, a remarkable increase occurred in the overall educational level. However, this increase has taken place in both cities almost equally as fast and – more importantly – the level of education of both the native Dutch and the non-Western immigrants also increased at about the same speed. Thus, the educational gap between the two cities has not really narrowed, nor has the gap between the different groups. Within the major non-Western communities, the Turkish and the Moroccans have the lowest average level of education, while the Surinamese, Antilleans and other groups occupy an intermediate position between them and the native Dutch population.

Notwithstanding the persistence of the educational gap, there has been a significant increase in the participation of immigrants in higher education: There is hardly any underrepresentation of immigrant students in higher professional education (hbo) in both cities to speak of anymore. There is still room for catching up at the university level, particularly for the non-Western immigrant groups. The conclusion must be that the share of highly educated immigrants in both cities is certainly increasing, but at the same time that the immigrants will remain overrepresented amongst the low educated for a long time still. The fact that with new waves of immigration, there are also new arrivals of low educated immigrants is also a cause.

A comparison of data in the field of labour for the two cities leads to conclusions that are remarkably similar to those for education. In 2010, gross labour force participation in Amsterdam (employed plus job seeking) was 5 percentage points higher than in Rotterdam (73% versus 68% of all 15–64 year-olds). In both cities participation among the native Dutch is roughly 20 percentage points above the level of persons of Turkish or Moroccan descent, with the Surinamese once more occupying an intermediate position. In the past two decades, labour force participation of non-native Dutch has increased remarkably – notwithstanding conjunctural fluctuations – but the participation level of the native Dutch has increased at almost the same speed. Consequently, the gap between immigrants and non-immigrants has hardly narrowed, a phenomenon similar to that in education.

The lower employment rate of Turkish and Moroccans appears to be caused in both cities largely, though not exclusively, by the low number of women participating in the labour market. In both cities only about one in ten native Dutch women aged 25–34 are not part of the labour force. Of Turkish and Moroccan women in this age group, by contrast, one in two do not participate. One can assume that the difference in the age categories above 35 years is at least as large, if not larger. One promising trend, however, is that the non-participation among young Turkish and Moroccan women of the second generation is at about half the level of the first generation, although it is still well above that of the native Dutch women. There may be cultural reasons for these differences. However, there is also ample evidence that persons with an immigrant background experience more obstacles when entering the labour market than their native Dutch counterparts. They do not always have the same networking and language skills and they may be victims of (indirect) discrimination. In addition, low-educated immigrants also face tougher competition than those with higher qualifications in a labour market that constantly puts up its demands. This is a bigger problem in Rotterdam, where the supply of low-skilled labour is substantially larger than in Amsterdam. I will come back to this later.

6 The City as a Way Station

As already mentioned, big cities exert an almost universal attraction to immigrants. Amsterdam and Rotterdam are no exception. The nineteenth-century immigrants to both cities came mainly from the Dutch countryside and from neighbouring European countries. In the twentieth century, especially in the second half, we saw a surge of migration from more distant regions, even outside of Europe. This development seems far from over, despite the increasing call in political circles for a stricter immigration policy. The demand for labour, the fact that both cities have significant immigrant communities established and the attraction this has for new immigrants, as well as the greatly improved communication and transportation facilities will make Amsterdam and Rotterdam attractive locations for newcomers in the future as well. The fact that both cities have a large number of relatively cheap accommodations for hire will certainly play a role.

However, the two cities are not only destinations of choice for newcomers: they are ever more becoming way stations. Gradually, a process of social ascent starts taking place in the larger, longer established communities of predominantly lower educated immigrants, a process which is often associated with geographic mobility in the form of a departure to the suburbs, which offer more space and tranquillity. We see here a repetition of the emancipation process that took place in the decades after the Second World War among the native Dutch. Thus far, the process is more evident among the Surinamese than among the other two major ‘classic’ immigrant groups, the Turkish and Moroccans, but eventually, for many of them social and geographical mobility will go hand in hand, all attempts at housing differentiation within existing neighbourhoods notwithstanding. The gentrification process that some older neighbourhoods in both Rotterdam and Amsterdam are undergoing, may keep some of the upwardly mobile within the city limits. Given its present scale, however, it is not very likely that this will involve large numbers.

It is not expected that all members of the ‘classic’ immigrant groups will undergo a process of social and geographical mobility. The data for both cities clearly show a growing division, also within the migrant communities. Some of them are prospering (sometimes as entrepreneurs), others remain in a situation of deprivation, characterized by low education, poor housing, little prospect of work, poor health and crime. This situation can easily continue in the third and even subsequent generations. The less successful ones will remain in the ‘disadvantaged’ neighbourhoods, and the homes left behind by the departure of the more successful ones will become free and occupied once more by newcomers. More often than before, these will not be (large) families who will permanently settle in the Netherlands, but singles or small groups living as ‘passers-by’: migrant workers from Poland, the Balkans, but also from outside the European Union, among them illegal immigrants.

This will cause the least attractive part of the housing stock in the big cities to attain even more of a way station character, with all of the attendant risks: neglect of houses, lack of social cohesion, deprivation, public health risks and crime. If we do not want to leave these areas to their fate, we need to invest heavily in the quality of housing and living environment, properly oversee the enforcement of rules, and also invest in integration, education and facilities for health, sports and welfare. Paradoxically, the residents of these neighbourhoods will probably not always know how to value these investments, as their involvement in the neighbourhood and even the entire urban society is rather limited. The local government should not expect the social involvement to increase dramatically due to a better social climate. The romantic notion of the old city neighbourhood with its sense of community is really a thing of the past, if it even really existed to the extent that people now sometimes assume. Yet this is not an argument against investing, otherwise important parts of both cities could slide into becoming no-go areas for the rest of the urban population.

The way station character of both Amsterdam and Rotterdam will not only become ever more manifest in the socio-economically lower levels, but also at the top. The figures show that more and more highly educated immigrants from both Western and non-Western countries are settling in the Netherlands, Amsterdam as yet being considerably more popular than Rotterdam. There should also be attractive housing available for them and their arrival can affect the social fabric of the city as well. Although they generally require less public attention than the socio-economically disadvantaged, they will also require special educational facilities, leave their mark on the local associations and perhaps demonstrate less involvement in their surroundings than the native Dutch population because many of them know they will not remain for long and therefore will not always take the trouble to learn Dutch.

7 The Importance of the Economy

Amsterdam and Rotterdam will each in their own way develop even more into ‘international’ cities – and become ever more distinct from the rest of the Netherlands (with the possible exception of The Hague, the third largest city of the country, where a similar development may occur). This internationalization will have an effect on all areas: the nature and level of facilities, the social fabric, education, health, political participation and so forth. Although both cities experience this development, there are significant differences. These have mainly to do with the fact that the two urban economies are decidedly different. Amsterdam is envisaged in the literature as a typical global city, though obviously not with the character and size of a New York or London, but one of the second or perhaps third echelon (Sassen 1991, 2006; Van der Waal and Burgers 2009). Some important features of such a global city are that a large proportion of economic activity has a strong international focus (e.g. in the form of housing the headquarters of multinational enterprises), that there is a highly differentiated economic palette and that it contains a particularly dominant service sector. The most important branches of economic activity in Greater Amsterdam are (in decreasing order): financial institutions, trade and commerce, consultancy and research, and information and communication. These four taken together account for well over 50% of the gross regional product in the Amsterdam area, which amounts to €75,000 per inhabitant. This is almost one-and-a-half times higher than anywhere else in the Netherlands, which illustrates the great economic importance of the city (Jonkers 2017).

Global cities may also be described as cosmopolitan, having a very diverse population and a rich cultural scene. Global cities are certainly not only for the elite, and their labour market is best described by using the hourglass model. The relatively large, higher educated, high earning share of the population generates a lot of demand for domestic and other services, which are provided by the lower educated workers. While global cities generally radiate dynamism, the threat of dichotomization, polarization and segregation always looms above the market. Amsterdam more emphatically satisfies the image of a global city than Rotterdam does (Van der Waal 2010a).

Rotterdam is better typified as a post-industrial city, characterized by one dominant economic activity, its port. And even though employment in the port has declined enormously in the past few decades, a large part of the Rotterdam economy is still directly or indirectly related to the port, such as the very important transport and logistics sector. To illustrate this: in 2014, 444.7 million tonnes of goods passed through the port of Rotterdam as against 97.8 million tonnes through Amsterdam. This made Rotterdam by far the largest port in Europe (Antwerp is second), and Amsterdam the fifth largest. The main branches of economic activity in the greater Rotterdam area are (in decreasing order): trade and commerce, transport and storage, industry, and health and welfare. Jointly they contribute to almost half of the gross regional product, which stands at €43,000 per inhabitant, less than 60% of its Amsterdam equivalent (Jonkers 2017).

The Rotterdam labour market much less takes the shape of the hourglass model than that of Amsterdam, but is characterized by employment opportunities at all levels, including the intermediate levels. At first glance, Rotterdam would therefore offer better possibilities for low-skilled workers than Amsterdam, but there is also a downside. Because the low-skilled work opportunities in Rotterdam are related less directly to the demand from the highly skilled segment of the labour market than in a global city like Amsterdam, the risk is greater that low-level and mid-level functions will relocate elsewhere: why employ Dutch truck drivers when the Polish are cheaper? Rotterdam therefore in effect experiences stronger outside competition than Amsterdam, which is all the more problematic as the proportion of the lower educated in Rotterdam is much larger than in Amsterdam. Rotterdam will have to do its utmost to retain employment opportunities for its low-skilled workers. In the past decade, it needs be said, the city has been highly successful in retaining those workers. In the 2000s, unemployment among lower educated workers declined even more than in Amsterdam, as did the number of benefits claimants. Since the economic crisis of 2008, however, this pattern has reversed. In 2016, the overall unemployment level in the Greater Rotterdam area stood at 8%, against 6% for Amsterdam. For a long time, the level of reliance on public assistance was very similar in the two urban areas, but since 2011 it has slightly increased in Rotterdam, while it has remained constant in Amsterdam. In 2016, 37 inhabitants per thousand benefitted from public assistance, while the corresponding number in Rotterdam stood at 41 (Jonkers 2017). One should keep in mind, however, that reliance on public assistance tends to be higher among the low-educated, and Rotterdam has many more of them than Amsterdam has. For both cities, but certainly for Rotterdam, it is and continues to be of great importance to invest in good job training possibilities, in a smooth transition from school to employment, as well as in retraining and permanent education for workers whose knowledge threatens to become obsolete.

As we have seen repeatedly, the proportion of lower educated workers in the Rotterdam population is considerably larger than in Amsterdam’s, both among the native Dutch and among those with an immigrant background. This fact could account for a number of differences between the two cities, for example the differences in health, in social and political involvement and in the extent to which the population identifies with their neighbourhood, city and country. Overall, the Amsterdam scores on these indicators tend to be higher than those in Rotterdam. It could also explain why Amsterdam at the borough and neighbourhood level has a slightly stronger ethnic segregation than Rotterdam. Ethnic and socio-economic lines fall together more decidedly in Amsterdam than in Rotterdam. Perhaps this is an explanation of why in Rotterdam the Freedom Party scored more than twice as high at the 2017 parliamentary elections as in Amsterdam (15.6% versus 6.8%). This populist party is well known to attract mainly native, lower educated voters. These voters are more likely than higher educated natives to experience displacement and competition from the immigrant labour supply (Van der Waal et al. 2011). It may equally explain why Rotterdam’s largest local political party, Liveable Rotterdam (Leefbaar Rotterdam), which also has significant populist traits, has no equivalent in Amsterdam’s local political scene.

Yet there are contraindications that are more difficult to interpret. Crime statistics are traditionally higher among the lower than among the higher educated, yet Amsterdam still has significantly more crime than Rotterdam. This may result from the fact that Amsterdam, as a major international tourist city, attracts a lot of foreign ‘scum’. Also, the more repressive security policy that Rotterdam has implemented in the last 15 years – encouraged by Liveable Rotterdam – may play a role here: the relatively high number of suspects arrested in Rotterdam may have contributed to the reduction in the number of crimes. It is also notable that the Amsterdam residents have many more inter-ethnic contacts in the private sphere than the Rotterdam residents. This is all the more surprising since in Amsterdam ethnic and socio-economic boundaries seem to coincide stronger than in Rotterdam, while most people usually prefer having contacts not only within their own ethnic group, but also with people of similar educational and socio-economic levels.

8 The Cultural Climate

A possible reason for this last paradox may be found in the research of some of my close colleagues at Erasmus University (Van der Waal et al. 2011). Following American researchers, such as Richard Florida (2004), they introduced the concept of ‘cultural climate’, which does not so much denote a summary of characteristics and attitudes of individual citizens as it does a specific urban environment or climate that affects the ideas of the local population. In American studies, this concept has been operationalised through the so-called Bohemian Index, derived from Florida (2002). The index refers to the number of city residents involved in producing art and culture. The larger their number relative to the total urban population, the higher the tolerance for diversity among the population, or so comparative research in American cities has discovered. This applies to the Netherlands as well. Amsterdam in particular scores very high on this index, while the Bohemian Index rate for Rotterdam is just slightly below the average for all Dutch cities (Van der Waal 2010b: 126; Crul and Heering 2008: 123).

Similarly, it is not surprising that, in the latest (2017) general elections held in the Netherlands the two political parties that are the strongest advocates of cultural diversity – GreenLeft (GroenLinks) and the left-wing liberals of D66 – ended first and second respectively in Amsterdam, with a total share of 38.5% of all votes. In Rotterdam these two obtained 24.7% of the votes. By contrast, the two largest parties in Rotterdam were the right wing liberal VVD and the populist Freedom Party. They obtained 32.1% of all votes, as against 22.1% in Amsterdam.

Van der Waal et al. show that the cultural climate of a city is more decisive for the way ethnic groups interact with each other than the system of economic opportunities (Van der Waal et al. 2011). This could indeed explain why, despite sharper social-economic differences between natives and immigrants in Amsterdam, there still seems to be more frequent inter-ethnic contact than in Rotterdam. One should, furthermore, not lose sight of the fact that the number of Western immigrants is much higher in Amsterdam than in Rotterdam and is also growing strongly. Part of these immigrants is made up of highly educated Europeans and Americans. It is plausible that a certain share of the inter-ethnic contacts in Amsterdam take place between highly educated natives and highly educated (Western) immigrants, and not between highly educated natives and people who belong to one of the classic ‘minority groups’. Nevertheless, the latter two categories will also regularly meet, namely in the context of the service economy so typical of a global city like Amsterdam (Van der Waal 2010a). Although Amsterdam has more inter-ethnic contacts than Rotterdam, considerably more discrimination appears to occur there as well. One might assume that more contacts would also lead to more opportunities for discrimination to occur, but classical contact theory suggests rather the opposite: the more contact, the more mutual understanding increases. In any event, further investigation into how socio-economic and cultural differences affect inter-ethnic contacts, ethnic stereotyping and discrimination in both cities is desirable. The results would undoubtedly contribute to an effective diversity policy.

9 Differing Approaches

This brings us to the government’s role in general terms and, more particularly, to its role in promoting social participation and harmonious interethnic relations. Amsterdam and Rotterdam both have been pursuing an integration policy for several decades, even if not always by that particular name. The main goal of this policy has always been involving immigrants, both individually and as groups, with local society. Several studies comparing the Amsterdam and Rotterdam integration policies have reached quite different conclusions as regard their contents and effectiveness. Godfried Engbersen in his book Fatale remedies (‘Fatal Remedies’) finds that the ‘rhetorics of integration’ has notably differed between the two cities: Rotterdam favours a forceful approach, focusing security and law and order and not shying away from intervening in private affairs.Footnote 5 Amsterdam, on the other hand, tends to use a softer ‘multicultural’ model, characterized by the famous desire of ex-mayor Cohen to ‘keep things together’. However, Engbersen says, in implementing these contrasting models the cities are more akin than divergent, a conclusion shared by Van Ostaijen and Scholten (2013) in their comparison of policy documents on integration issued by each of the two cities over the past few decades. Though the Rotterdam local government at times may talk about a hard-line approach to integration, Engbersen argues, it has for years now involved the larger Islamic organizations in the city in its policy-making process. The Amsterdam authorities, often described as being more ‘soft’, have certainly acted vigorously against criminal youth (Engbersen 2009: 171–191).

In contrast to Engbersen, Justus Uitermark in his thesis Dynamics of Power in Dutch Integration Policies emphasises the differences between the integration policies of the two cities. He notes that the Rotterdam integration policy has been remarkably consistent, whatever the political composition of the local government at any one time. The desire to prevent strongly concentrated migrant populations in some neighbourhoods has been a central theme of the policy for quite a while now. Rotterdam has also pursued a vigorous civic integration policy and systematically provided professional support to immigrant organizations when shaping its integration policy. Even when Liveable Rotterdam is part of the Municipal Executive (2002–2006 and, once again, 2014–2018) contact with the Islamic organizations has been maintained as ever. This consistency, according to Uitermark, has clearly benefited the transparency and effectiveness of the policy. In Amsterdam, Uitermark finds, the policy has been less consistent over the years. Policy targets have been adjusted often and though the city offered much support to immigrant organizations, it often changed the organizations in favour. Furthermore, support was primarily of a financial nature, not of a professional one, as in Rotterdam. And this is why, Uitermark concludes, ‘minorities in Rotterdam are more socially and politically involved, organizations for minorities are more capable of taking action collectively and is there less of a presence of (Islamic) extremism’ (Uitermark 2010: 280).

10 Future Perspectives

Amsterdam and Rotterdam have a more diverse population than ever before, yet they also display substantial differences, which stem from, in part, their distinct economic structures and social-cultural climates. In the future, both cities will continue to be major poles of attraction for international immigrants. An ever-increasing share of the population in both cities has an immigrant background, while an increasingly smaller share settles in for the long term. As a result, the two largest cities in the Netherlands are more and more developing into way stations. Many foreign immigrants return to their own countries sooner or later, or even choose to migrate on to a new country. Though the immigration waves of a few decades ago resulted much more frequently in permanent residency than was presumed at the time, one cannot assume that newly arriving international immigrants will continue to settle in permanently. And among those that do remain selective mobility, both in the geographic and social senses, will eventually occur. This pattern is already apparent amongst the older immigrant groups. There is a tendency amongst the disadvantaged immigrants to remain in the city whilst those that are more successful fan out to the suburbs or integrate into Dutch society to such an extent that they are no longer viewed as immigrants (or the descendants of immigrants). Thus, the issues of immigration and integration will continue to present important challenges to both cities far into the future.

For a long time it was assumed that an integration policy oriented specifically at newcomers would be sufficient to allow them to become full members of the local urban society. That idea increasingly appears to be outmoded. The immigrant issue is so encompassing that an integration policy alone is far from adequate to address it. Policy areas like the economy, employment, education, housing and safety are in many ways far more important because they provide the basis for all of the citizens in a city to participate in society regardless of their level of education and whether or not they have an immigrant background. Both Amsterdam and Rotterdam appear to be increasingly aware of this, although their approaches up to this point have differed slightly. Amsterdam seems to invest more in fostering a sense of connectedness with the neighbourhood, while Rotterdam is seeking to promote participation through organizations. Ethnic background is but one factor in an interplay of forces featuring a great many other factors, though. As the number of immigrants rises and the length of their stay increases, the composition of the population becomes more diverse and will consist of minorities only – especially on the level of individual neighbourhoods – the ‘ethnic background’ factor starts to lose distinctiveness and, thus, relevance (also see Entzinger 2014).

Moreover, the way station nature is becoming ever more prevalent in some neighbourhoods of the cities. The number of people who settle in for the short term only will continue to grow and this population will generally only feel a limited sense of connectedness with urban society. Municipal executives will have to take this increasing way station nature of the bigger cities into account. How can cities with such a rapidly-changing, heterogeneous population base develop social cohesion? If this is the key question, promoting the integration of a few specific immigrant groups cannot remain the main objective of governmental policy – which both Amsterdam and Rotterdam have long since realised. It is much more about creating the conditions for an urban society in which everyone feels welcome and everyone, no matter how short their stay, can contribute as well. In such urban societies the distinction between ‘native’ or ‘immigrant’ will no longer occupy such a central position as in past decades. The ethnic background of a citizen can be a relevant factor, also in issues of policy – but it is only one amongst many. It is not a dominating factor that appears to subordinate all others, as is so often the case now. As time goes by, the boundary between immigrant and native will continue to erode. Having an immigrant background is no longer, by definition, an indication of social deprivation. The immigrant population is becoming more heterogeneous and is mixing more and more with the native population. That is how things have always been in the past in the Netherlands, and that has been the way of things (almost) everywhere else.

11 Conclusions

The approach advocated here requires a governmental policy with a primary focus on the urban society in its totality. The local government must continue to strive for a healthy urban economy, a safe living environment, and for high-quality social, cultural and educational facilities that are equally accessible to everyone. This does not mean that in the nearby or somewhat distant future questions may not arise again that pertain specifically to one or several communities. That is logical in an urban society that consists purely of minorities. That is, however, something quite different from implementing an integration policy geared towards all minority groups. Especially now that no actual majority community seems to exist anymore for minorities to integrate in, there is no longer a rationale for a group-oriented integration policy.

Amsterdam and Rotterdam are developing into what Vertovec has labelled ‘superdiverse cities’ (Vertovec 2007). In the light of this, and despite the differences identified between the cities, we should conclude that a ‘classic’ local integration policy aimed at individuals and groups would quickly lose its impact. This certainly applies to integration policies that are based on ethnic differences. It appears much more sensible to implement an integration policy for the most relevant policy areas that, as a matter of principle, looks at all citizens as being equal, but that, like any good policy, when necessary takes account of the differences between them, even if these differences stem from their immigrant background or culture. The objective is no longer primarily the integration of (immigrant) citizens into an existing urban society, but rather to promote a sufficiently integrated urban society in which as many citizens as possible feel welcome and at home, even those whose involvement is and will remain limited. The local government cannot accomplish this by itself. Businesses, educational institutions and other societal organisations must also help in preventing people from living side by side instead of with each other, even in those urban societies that increasingly function as way stations.

Thus, the capacity of governmental policy to shape society should not be overestimated. Urban societies develop largely autonomously and many actors other than the local governments can exert their influence on that development. This relatively autonomous progression also applies to the integration of immigrants. The single greatest influence public authorities can exert on the integration process concerns issues of legal status and law enforcement. In particular, legal status is extremely important for many immigrants: having a secure right to stay is a prerequisite for successful integration. This is, however, primarily a matter for the national government. Cities have very little to say about the legal status of immigrants, no more than they get to specify which newcomers settle into their borders. The scope of local governments is somewhat greater in the social-economic arena, but is limited here as well. In our economy it is the market that determines what happens; cities may steer market processes in the economy, but to a limited extent only. The cities, however, do have a specific responsibility, ever increasing due to decentralisation, for the social wellbeing of their citizens. Finally, the role of the government in liberal-democratic societies such as ours is limited when it comes to the cultural arena. Especially in the bigger cities, local governments are being confronted with a multitude of cultures and religions. Their duty then appears to be, above all, to guarantee the peaceful co-existence of people of all cultures, and to resolve or de-escalate any conflicts that arise. In addition, the importance of the local government setting an example when it comes to promoting equal opportunities for all its citizens and to actively combating discrimination should not be underestimated (Collett and Gidley 2012).

If this comparison of Amsterdam and Rotterdam has accomplished anything, it is surely the understanding that it makes little sense to focus exclusively on the course and the management, direct or indirect, of integration processes as long as one fails to involve the context within which the processes are taking place. The two cities certainly differ in the impact that immigration has had on their social, economic, cultural and political fabrics. Yet, both cities clearly show that immigration and its consequences exert such a powerful influence on urban development in general that it becomes impossible to look at them independently from each other. In other words: good integration policy is really nothing more and nothing less than good urban policy.