Insecure tenure in Amsterdam: who rents with a temporary lease, and why? A baseline from 2015

Abstract

Given that insecure leases impact negatively on ontological security and subjective well-being, and given increasing pressure on European housing markets, more insight into insecure leases is timely. In this article, we assess the occurrence of temporary leases in the city of Amsterdam in 2015, and explore the characteristics of the tenants. We employ hitherto underused local survey data (N = 17,803). Although permanent contracts are still dominant, the majority of young adults aged 18–23 are renters with a temporary lease. Students, those with a Western migration background, those who moved because their previous rental contract was terminated or because the previous dwelling was too expensive, and those who moved from abroad were particularly likely to have a temporary lease. Families were unlikely to have a temporary lease. Given recent developments – in 2016 temporary leases were legally established as a regular tenure in the Netherlands – the number of temporary leases may increase sharply from the reported baseline of 2015.

Keywords:

• Security of tenure
• rental housing
• the Netherlands
• temporary contracts
• multinomial logistic regression
• policy

1. Introduction

Secure housing is important for people’s psychological well-being. Uncertainty about when you have to leave your current home and find a new one can cause stress and undermine ontological security. In the Netherlands, where in 2015 42.6% of all households rented (Statistics Netherlands, 2019a), regular tenancies are of unlimited duration and difficult for the landlord to terminate unless the tenant does not fulfil certain basic conditions, such as paying the rent on time. In contrast, fixed-term tenancies, tenancies with unlimited duration which the landlord can terminate easily, and tenancies that depend on certain transitional conditions linked to either the tenant or the house all share the trait that they increase the risk of uncertainty for the tenant. For brevity we refer to these non-permanent leases collectively as temporary leases.¹ For a long time, the Netherlands has been known for its high levels of tenant protection, with permanent renting contracts traditionally being the norm (Huisman, 2016a, 2016b) and starting rents as well as increases regulated by government. Consequently, renting is not very different from home-ownership in the sense of security of tenure. This stands in stark contrast with the situation in, for example, the UK (Fitzpatrick & Pawson, 2014), the USA (Desmond & Gershenson, 2016) and Australia (Darab et al., 2018; Hulse & Milligan, 2014), where renting contracts can easily be dissolved by the landlord.

Yet, it seems that the Netherlands may be on its way to losing its rental protection, since over the last two decades, ever more variants of temporary leases have been introduced (Huisman, 2016a). The youth contract followed the introduction of the student contract, and rental contracts that end when the apartment is sold came after those that are automatically terminated when the dwelling will be renovated or demolished. At the same time, informal arrangements as well as irregular/illegal forms of renting such as anti-squat (house guardianship) thrived. Despite this proliferation of temporary rental contracts conditional on characteristics of the house or the tenant, the number of such contracts is not registered anywhere. We do not know how many households in the Netherlands have a temporary contract, nor do we know if and how these households differ in significant ways from other households.

The reasons for the gap in our knowledge have been summarized as follows (Huisman, 2016b). An important assumption is that temporary rent is simply a bonus addition to the stock because it brings previously vacant spaces into use. Secondly, temporary leases have been regarded as only a short, transient and thus insignificant phase in people’s lives. Also, newly introduced contract forms have each been regarded as incidental solutions to incidental problems, while their combined impact has been overlooked. Lastly, there are inherent methodological difficulties associated with measuring temporary rent, such as the invisibility of such forms of tenure in official databases and low response rates amongst people with temporary tenancies.

In this article we focus on the insecurity in the rental sector that is connected with temporary leases, where fulfilling the financial contract (i.e. paying the rent each month) does not necessarily guarantee continuation of tenure. Understanding the phenomenon of housing insecurity is timely since a shift from stable to insecure renting is emerging (or ongoing) in a number of countries. In the Netherlands the phenomenon of temporary leases is relatively new, while in countries such as England,² Australia and New Zealand, countries with already precarious private rental sectors where temporary leases are the norm, the comparatively secure social housing sector is now also increasingly becoming insecure through the introduction of temporary leases (Fitzpatrick & Pawson, 2014). Combined with the relative as well as absolute increasing numbers of households renting after the Global Financial Crisis in several European and Anglo-Saxon countries (i.e. Denmark: Statistics Denmark, 2017; Ireland: CSO, 2017; US: JCHS, 2017; UK: DCLG, 2017), and many liberal governments promoting the private rental sector, more insight into temporary leases is imperative. This also holds for knowledge on the characteristics of households with temporary contracts.

With its history of a strong rental sector and a recent widening of options for temporary leases, the Netherlands is a prime, concrete example of a society in which temporary lease could quickly gain importance. In the Netherlands, Amsterdam is the city with the tightest housing market and the largest rental sector (see Section 2). This article, therefore, contributes to the international literature by providing insight into temporary lease in the city of Amsterdam, as well as developing an analytical framework into the distinguishing factors for households to end up in temporary lease.

Without insight into the incidence of temporary leases in the Netherlands or which households have such leases, or knowledge about the effects of the previous widenings, the Dutch parliament recently decided to further widen the possibilities for the legal use of temporary leases. In July 2016 a law was passed, the Housing Market Throughput Law (in Dutch: Wet Doorstroming Woningmarkt), which removed the exceptional status from temporary leases, establishing them as a normal form of tenure for private landlords. The stated goal of this widening is to increase the supply of housing, as well as the availability of dwellings for specific target groups. Temporary leases are also presumed to reduce financial risks for potential landlords, making it more likely that they will lease out their properties. Beyond this landmark change, housing corporations are now also allowed to use temporary leases in more cases than before. Specifically, five year contracts for young adults and shared housing were introduced, as well as conditional contracts for instance students, internal and foreign immigrants, those in acute need and large families (Dutch Civil Law 7:271-274; Staatscourant, 2016). This law has the potential to change the Dutch housing market significantly. The normalization of temporary leases could considerably accelerate the already speedy residualization of renting, by stimulating home-ownership as the only means of escape from insecure tenure. Alternatively, analogous to what happened in the UK from 1989 onwards, it could lead to the complete precarization of the private rental sector, and the growth thereof. Certainly, since the introduction of the law a large housing corporation decided to completely switch to the use of temporary contracts by narrowing its target group to students and young adults (De Key, 2015), and press reports suggest a sharp increase in their use (Damen & Bontjes, 2017; Segaar, 2019).

To be able to assess the shift towards more temporary leases empirically over the years, one requires a baseline. Providing such a baseline is one of the goals of this article. The other goal is to gain insight into the characteristics of those living with temporary tenancies. We focus particularly on the capital of the Netherlands, Amsterdam. Amsterdam is an interesting case for studying temporary leases because of its tight housing market and its attractiveness to students, immigrants and internal labour migrants. By employing local survey data from the Housing in Amsterdam (Wonen in Amsterdam, WIA) survey conducted in 2015, we estimate the volume of temporary tenancies and assess who is using this form of tenure, using descriptive analyses and multinomial logistic regressions of housing tenure. We use the most common tenure in Amsterdam, permanent leases, as a comparison and also present results on the chance of home-ownership. The WIA survey was the first in the Netherlands that cautiously started to ask questions about temporary tenure. The year 2015 is particularly relevant because it is, essentially, the point just before the introduction of the new law that widened the possibilities for temporary leases and thus useful as a baseline for future monitoring.

2. The Dutch housing market and the Amsterdam context

While traditionally a country where the majority of the population rented, by the end of the 20st century the Netherlands was well on its way to becoming a nation of home-owners. By 2015 57.4% of all households owned their own home (Statistics Netherlands, 2019a). The switch from renting to home-ownership can be largely attributed to Dutch housing policy, which since the 1980s can best be characterized as the active promotion of home-ownership complemented by the deregulation and residualization of renting. Direct as well as indirect subsidies, varying from for instance tax rebates to governmental mortgage guarantees, helped many households to acquire homes. For the last decades, more than three-quarters of all rental housing was owned by housing corporations (calculated from Woon, 2019a): not-for-profit organizations which can be described as quasi-autonomous non-governmental organizations or quangos, because they function in close cooperation with the government in executing rental policy.

In line with the general political developments in the Netherlands, neoliberal arguments that the long-established system of rent controls applying to most rental housing should be viewed as market-distorting subsidies fell on fertile ground. Rent controls were held responsible for continuing (localized) housing shortages, hampering investors from developing more rental housing. Deregulation of the rental stock is an attempt to mitigate this. As a result, the share of regulated rental housing has been constantly decreasing and the share of unregulated rental housing increasing. Even within the regulated sector rent levels have been raised significantly in the last 15 years, enlarging the proportion of income households have to spend on rent,³ and making renting a less favourable alternative to buying a home than before.⁴ Indeed, between 2009 and 2015 the number of homes affordable for low-income households was halved, whereas the number of dwellings with free-market rents more than doubled (Blijie et al., 2016). Less affluent households are partly compensated for the high rent levels by individual housing benefits. To curtail the resulting increasing impact on the national budget, tighter restrictions on what sort of housing low-income households can rent were put in place in 2016 (Dutch Housing Law, art. 46). The shrinking stock of regulated housing is more and more earmarked for specific target groups, and viewed as a temporary safety net or social service for more vulnerable people who are not (yet) capable of managing independently on the market. Bar some very curtailed exceptions, such as for officially recognized refugees, homes with regulated rents from housing corporations are distributed according to time on a waiting list.

Relatedly, in spatial policy by the end of the 1980s the goal of regional deconcentration of economic growth was interchanged for a focus on stimulating economically already prosperous areas. The four largest cities of the Netherlands, together known as the Randstad, have grown continuously ever since. Spatial planning constraints have however hindered the construction of new housing to some extent, so that especially Utrecht and Amsterdam, as the main economic hubs of the country (Raspe et al., 2010; Statistics Netherlands, 2017), have been experiencing shortages of both rental and owner-occupied housing (Van Duinen et al., 2016). A former bulwark of progressive policy, Amsterdam has a large proportion of rental housing in comparison with the rest of the Netherlands (Table 1). In 2015 over two-thirds of the local stock was rental, and of this 82% was regulated rent. The city also lagged behind on the promotion of home-ownership, although since the beginning of the 2000s it has been catching up (Aalbers, 2004; Huisman, 2009). At the same time, the unusually large demand for housing in Amsterdam far exceeds the available supply. As a consequence, the prices of homes for sale as well as the levels of rent charged are among the highest of the country, and keep rising (Statistics Netherlands, 2019d; Gualthérie Van Weezel & Huisman, 2017). There are concerns that the conversion of homes to commercial, all-year-round vacation rentals (for marketing on platforms such as AirBnB, which are attracting a substantial amount of visitors to the city) is reducing the housing stock, and thus increasing pressure on the housing market (Nieuwland & Van Melik, 2020). Such conversion is deemed illegal by the municipality of Amsterdam, but enforcement proves difficult.

Table 1. Dutch and Amsterdam housing market (2015).

	The Netherlands^a		Amsterdam^b
Population	16,900,726		822272
Occupied dwellings	7,211,229		417100
Owner-occupation	4,119,362	57.4%	127500	30.6%
Renting	3,244,238	42.6%	289600	69.4%

Sources: ^aStatistics Netherlands (2019a, 2019e); ^bRIS (2016, p. 19), Municipality of Amsterdam & AFWC (2016).

Tables 2 and 3 give a detailed overview of the different sectors of the Amsterdam housing market. Table 2 describes the three long-established forms of tenure, namely home-ownership, permanent rent and irregular housing options. Home-ownership appears to be unattainable for the majority of the population. Because rent regulations are owner-neutral, homes with regulated rents can be either owned by housing corporations or by private landlords. Given the high demand for housing, waiting times for homes owned by housing corporations with regulated rents have risen constantly since the 1980s. By 2015 they started at approximately 8 years of waiting time for a home in the least attractive neighbourhoods, rising to over 20 years for more popular areas (AFWC, 2016, p. 31). Private rental dwellings with regulated rents are let out through the social network of the owner (Hochstenbach & Boterman, 2015). Few are coming up for rent and vacated dwellings are usually put into unregulated rent. Both corporation-owned and privately-owned homes with unregulated rents are distributed through the discretion of the landlord. Because of the substantial profit that can be made by renting out apartments, the amount of homes bought to let is on the rise (Van der Molen, 2017; Nul20, 2019). As can be gleaned from online housing ads, private landlords often prefer expats as renters. Such high-income migrants are expected to be able and willing to pay high rents for short-term leases without demurring, given their lack of acquaintance with Dutch renting culture. Sharing a home, defined as living together with one or more adults you have no family relation with, can happen across all sectors of the housing market.⁵ Considering more irregular but equally long established housing options, subletting of apartments is forbidden by law (Dutch Civil Law book 7:244), but can be highly profitable given the pressure on the housing market. Since squatting was made illegal in 2010, the number of squatted homes has most probably decreased, but as with subletting, no official data are available.

Table 2. Long-established forms of tenure Amsterdam housing sector (2015).

	Landlord	Net monthly cost	Entry conditions	Security	Share of market^a
Owner-occupation		Aver. 820 euro^b	Affluent enough to obtain mortgage (approx. 50,000 euro gross a year)^c	Good	30.6%
Permanent rent
Regulated rent
	Housing corporation	400–710 euro^d	> 8 years on waiting list, income < 34,911 euro gross a year^e	Good	42.9%
	Private landlord	400–710 euro^f	Discretionary freedom of landlord	Good	14.5%
Unregulated rent
	Housing corporation	710–1000 euro^d	Discretionary freedom of landlord, affluent enough to pay free-market rent	Good	2.7%
	Private landlord	710–1500 euro^g	Discretionary freedom of landlord affluent enough to pay free-market rent	Good	9.4%
Irregular housing options
Subletting	Tenant	400–1500 euro	Informal social network	Non-existent	–
Squatting	n/a	low	Willing and able to break law	Low	–

Sources: ^aMunicipality of Amsterdam & AFWC (2016); ^bWoon (2019b); ^cNibud (2015); ^dAFWC (2016, p. 12); ^eWoonbond (2014); ^fMunicipality of Amsterdam (2015); ^gMunicipality of Amsterdam (2016a).

Table 3. Forms of tenure introduced since 1997 Amsterdam housing sector (2015).

	Landlord	Net monthly cost	Entry conditions	Security
Campus contract
Shared	Only housing corporations	200–400 euro^a	Local college fee paying student, 1 year on waiting list + cooptation system	Contract ends when no longer fee-paying student
Non-shared	Only housing corporations	400–590 euro^a,b	Local college fee paying student, 3 years on waiting list
Youth contract
	Only housing corporations	Approx. 400 euro^c	Age between 18–22, no children, not for students, 4 years on waiting list^d,e	Contract ends at age 26
Temporary rent based on Law on Vacancies
	Housing corporation or private landlord	400–1500 euro	Discretionary freedom of landlord	Contract ends after 1/2/5 years
Anti-squat
	Agency, housing corporation or private landlord	80–235 euro^f	Discretionary freedom of landlord + cooptation systems, no children, willing to live in non-housing	Non-existent: notice to quit within 14 days

Sources: ^aDuwo (2016, p. 20); ^bAFWC (2016, p. 13); ^cStadgenoot (2013, p. 53–54); ^dVan der Tol (2016); ^eSchouten (2014); ^fVan Eijck & Naas (2014).

Table 3 deals with forms of tenure that were introduced from 1997 onwards. These are all temporary. Amsterdam housing corporations, who own most of the rental stock, have been actively lobbying for the introduction of temporary contracts (Huisman, 2016b; Nolles, 2013). According to them, the current system of distribution of scarcity by waiting time should be replaced by a system that continuously gives new, young single people and couples a chance to live in the city temporarily (households with children are excluded, Nul20, 2015). The idea is that when the contract ends, the tenant has perhaps gained enough income to afford free-market rental housing or become an owner-occupier. If things have not proceeded so favourably, the tenant can obtain a rented home in the periphery of the city. The campaigns of the housing corporations have led to the introduction of several new forms of temporary leases. Student housing is nowadays in Amsterdam solely let out with campus contracts (since 2005, see Municipality of Amsterdam, Amsterdam Universities and Housing Corporations, 2005), which are automatically terminated 6 months after the tenants finish or terminate their studies (Municipality of Amsterdam, 2016b). In 2012, several Amsterdam housing corporations started a pilot project with so-called youth contracts for young adults age 18–22, which ended once the tenant reached the age of 26. (Starting from July 2016, the conditions changed. Since then, one-third of all vacated homes with regulated rents from housing corporations can be rented out with youth contracts which are now available for those in the age category 18–27, and in this new form the lease automatically ends after 5 years, Amsterdam Tenant Association, Amsterdam Federation of Housing Corporations and Municipality of Amsterdam, 2016.) Temporary rent based on the Law on vacancies allows housing corporations and private home-owners to temporary let out their homes on a time-limited lease when they are to be renovated, demolished, or vacant awaiting sale. Selection of tenants is the personal discretion of the landlords. For such temporary rent, the landlord requires a permit. As such, they differ significantly from generic temporary leases currently available to private home-owners, which are based on the law that came into effect in 2016. Such leases can last at most 2 years, and afterwards the contract cannot be prolonged with the same tenant, but can be turned into a permanent lease, or the tenant can be replaced by a new tenant with a similar temporary lease. No permit is required. Finally, anti-squat or guardianship was originally introduced to prevent squatting of vacant properties, but given the successful criminalization of squatting, anti-squat seems to function solely as a way to let out homes without basic rights for tenants.

Demand for housing structurally exceeds supply. Careful observation of the column labelled ‘entry conditions’ shows that for newcomers to the city, housing options are scant. Affluent people might be able to afford to buy a home, or to rent in the unregulated rental sectors. Those on lower budgets cannot access regulated rental housing because they lack waiting time (for corporation-owned housing), and the chance that a private landlord will select them to rent out one of the few newly vacant homes with regulated rents is extremely low. Indeed, given the structure of the local housing market, there is a clear need for resources that most newcomers lack: an extensive social network that includes people who can help finding housing, and knowledge of the local housing market. Illegally renting a sublet home may be one of the most feasible options. Certainly, the conditions under which a newcomer to a city such as Amsterdam can secure a permanent rental contract are so limited that it is unrealistic to speak of a choice between permanent and temporary forms of tenure. Rather, temporary lease is frequently the only available choice.

3. Theoretical framework—preferences, needs, resources and constraints: housing decisions leading to temporary leases

Temporary leases lead to housing insecurity. This term refers to ‘residents’ limited capacity to determine how long they may remain in their home’ (Morris et al., 2017, p. 653). This insecurity is important because it impacts negatively upon people’s ontological security (Giddens, 1991; Saunders, 1990), that is, the stable psychological basis that people require to thrive and develop is undermined. Such an erosion of stability contributes negatively to subjective wellbeing and mental health. These effects sometimes cause people to live in an atmosphere of anxiety and fear (Darab et al., 2018; Fitzpatrick & Watts, 2017; Morris et al., 2017), and they can lead to reduced autonomy and privacy in the dwelling.⁶ In their study of Australian long-term private rental sector tenants in large cities, Morris et al. identified three typical responses to housing insecurity; ‘incessant anxiety and fear; lack of concern; and concern offset by economic/social capital and traded off against locational preference’ (2017, p. 653). Indeed, the burden of insecure tenure is not evenly distributed, falling more heavily on those with fewer financial resources and smaller networks, as well as cultural capital.

People's residential situations are usually understood as the result of their housing choices. Housing choices occur when people consider moving. Important motives for moving are events in the life course such as leaving the parental home, starting to live with a partner, having children, union dissolution, child launching, retirement and reaching old age (Mulder & Hooimeijer, 1999). Some moves are motivated by work, education or by housing-related preferences. It then follows that the type of tenure is simply one of many factors that households have to balance in their residential choices.

In the literature on the choice between renting and owning, it has indeed been argued that this choice depends on the benefits and costs of owning versus renting, the resources to overcome the costs, and the temporal and spatial context. The balance between benefits and costs is differentiated between individuals, households and life-course stages, and so are resources (Mulder & Wagner, 1998). Because moves into owner-occupied homes are associated with high transaction costs, home-ownership is less desirable for those who foresee they may move again soon, for example students, young adults more general, newly divorced people and singles. Owning also requires a stable income and a certain level of assets, making it more affordable to those with steady jobs, those in dual-income couples and those who have spent more time in the labour market (Mulder & Wagner, 1998).

To a certain extent, temporary leases might be viewed as just a type of renting. In that sense, one might argue that those population categories who are likely to rent – rather than own – would also be likely to have temporary leases. However, temporary leases are specific in the sense that it is hard to think of any benefit they would have compared with a more secure type of tenancy. In the legal framework of the Netherlands, tenants can terminate a permanent contract relatively easily, so legally a tenant does not constrain herself unduly by choosing a permanent contract over a temporary contract. As a result, temporary leases are very unlikely to be the first choice of people because they are temporary, even though some might prefer to rent a particular dwelling despite it being offered with a temporary contract.

For this reason, in our theoretical argumentation, we give a large prominence to urgency and scarcity in analysing why people live with temporary leases. With regard to preferences or benefits, we think the issue is not so much what are the circumstances in which some people might prefer to live with temporary leases, but rather what are the circumstances in which some people might be more prepared to accept temporary leases as an option. Concerning urgency, some moves are indeed not the result of a choice based on a preference, but the consequence of a pressing need arising out of circumstances.⁷ Such urgency limits the time available to search for suitable housing and may, therefore, lead to accepting a temporary lease if no other options are available. Regarding scarcity, it is important not to overlook or marginalize structural aspects of the context, that is, the local housing market, in the determination of housing decisions (Tu et al., 2017). Given the current constraints of the Amsterdam housing market, it is very difficult for many to obtain other, more secure housing. The absence of other options may be caused by urgency and/or restrictions in resources, such as money, time on the waiting list, knowledge about the local market and social capital (for example in the form of friends or acquaintances who help finding housing or introduce someone to a landlord), which are aggravated by the overall scarcity of housing. Another reason for accepting temporary leases would be that some people might mind less about the temporariness of their housing. People who only expect to live in Amsterdam for a short time, or expect to move soon again, may be less bothered by a time-limited contract. Alternatively, those with fewer ties binding them to their dwelling such as a partner, children or possessions may be less attached to security of tenure.

We hypothesize that a number of particular characteristics will increase the chance of accepting a temporary lease. Age will be an important factor, for a number of reasons. For many young adults, there is an often rapid succession of important events at the earlier moments in the life course, such as leaving the parental home, entering into higher education, the first job, changing jobs to establish a career, moving in with a partner and starting a family, and these events often coincide with moving (Bernard et al., 2014). Those who are younger, however, have fewer housing options, since they have lower incomes, less savings and no steady employment career (yet). Owner-occupancy through a mortgage, as well as renting a home with an unregulated rent will be often out of their reach (Filandri & Bertolini, 2016). Also, given that homes with regulated rent are distributed mainly through waiting time, and registration is only possible from age 18 onwards, those who are younger are at a disadvantage when looking to obtain such secure affordable housing. The availability of youth contracts also contributes to the expectation to find mainly young people with temporary leases. Furthermore, for youngsters temporary housing might match with the transitory character of their current life phase. In contrast, more senior citizens seem to be more attached to security of tenure. For instance, in their research among tenants in the social rented sector in England, Fitzpatrick and Watts found, with regard to the insecurity of their leases that: ‘Older people, people with health or disability issues, and families with children, tended to be the most concerned, though more pronounced anxiety was not confined to those groups’ (2017, p. 1032). Similarly, Darab et al. (2018) who focused on women above 45 in a more rural part of Australia, found that for this population category, security of tenure was the unanimously shared and foremost housing preference.

Relatedly, we expect being enrolled in education to enlarge the chance of having a temporary lease. Students have an urgent need to be housed in the vicinity of their educational venue, for practical reasons, while their resources are usually quite restricted (Feijten & Mulder, 2005). Dedicated student housing is the main housing option for students in Amsterdam, and at the present time, it always comes with a temporary lease, while not many other options are present. Students might not mind so much about the insecurity of their tenure, because they may expect to live there only for the duration of their studies.

Furthermore, we anticipate people who recently moved to Amsterdam from either inside the Netherlands or from foreign countries to have a higher chance of accepting temporary leases. Not already residing in Amsterdam, they may not have much time to look for adequate housing, while not many alternatives present themselves to them. They are likely to still miss the elaborate social network that is necessary to obtain housing in informal ways (Hochstenbach & Boterman, 2015), and to have insufficient knowledge of the local market. Affluent foreign migrants may specifically have a higher chance of obtaining housing with temporary leases, because of the strong preference of private landlords to rent out their more expensive homes to this group with this form of contract. Another factor is that some of those moving to Amsterdam may only expect to be in the city for a limited time. They may, therefore, be less attached to security of tenure.

With regard to international migration background, we expect those with Western migration backgrounds to have a higher chance of accepting a temporary lease because many of them will be expats envisaging a return to their country of origin. On the other hand, we expect those with a non-Western migration background to have a lower chance to accept a temporary lease. In Amsterdam, the main categories of those with a non-Western migration background are those with at least one parent born in Turkey, Morocco, Suriname or the Dutch Antilles. On average, people in these categories leave the parental home later than those who do not have a migration background (Stoeldraijer, 2014). When such young adults do move out, they more often go to stay with a member of the family or to share with others.

Household situation will influence willingness to accept a temporary contract as well. Those who live alone may be less attached to security of tenure than those who live with a partner, for instance because it is easier for them to move. Similarly, those who have no children living at home may mind less about temporary leases. Furthermore, most official forms of temporary lease, such as student contracts, youth contracts and temporary lease before demolition or renovation are not made available to households with children (Nul20, 2015). Those living alone will also have less spending power because they cannot combine their income with their partner to obtain alternative forms of housing. We also expect that those who share accommodation with others might have a lower chance, since they can join resources. Since divorce and relationship break-up play an important role in housing outcomes (Mikolai & Kulu, 2018), including the impact of these life-course events would have been valuable. Unfortunately the data for this are not available.

Considering income and employment status, less affluent people are excluded from several housing options, such as free-market rentals or owner-occupancy, while some forms of temporary lease (i.e. anti-squat) are more affordable. In Amsterdam, this particularly affects those who have not yet gained sufficient time (more than 8 years, Table 2) on the waiting list for homes with regulated rent distributed by housing corporations to households with lower incomes. Likewise, those who are unemployed, living on benefits or self-employed, may have a higher chance of accepting temporary leases, because they are often excluded from more secure, more expensive forms of housing.

People’s level of education will also influence their chances on the housing market. In general, highly educated persons obtain higher quality housing than less educated (Feijten & Mulder, 2005). They may be viewed as attractive tenants by landlords, or may obtain a mortgage more easily because of their favourable income prospects. Therefore, we envisage those with a higher level of education to be less likely to accept temporary leases. On a more speculative note, we wonder whether there might be a link between gender and the acceptance of temporary leases. Perhaps women are more attached to security of tenure, and it is easier for them to obtain, since landlords sometimes prefer women over men (Flage, 2018), because they are perceived as quieter and tidier tenants. At the same time, women have fewer options on the housing market, especially after divorce or relationship breakup (Feijten, 2005; Mikolai & Kulu, 2018).

Those whose rental contract was terminated or whose housing had become too expensive, have in common they have to make do with whatever is available because of the immediacy of their housing need. They will, therefore, have a higher chance of renting with a temporary lease.

Finally, since home-ownership is the second tenure in Amsterdam after permanent renting, we also include it in our analysis. For most characteristics, we expect the chance of being an owner-occupier to be the opposite of the chance of having a temporary lease.

4. Data and method

4.1. Dataset and sample

Housing in Amsterdam (WIA) is a survey jointly commissioned by the Department of Housing of the Municipality of Amsterdam, the seven Amsterdam boroughs and the Amsterdam Federation of Housing Corporations.⁸ The research has been repeated every two years since 1995 (Dignum & Kan, 2014). Included are questions about residents’ current and previous housing situation, their housing wishes, their socio-economic status and how they evaluate their surroundings. In 2015, a stratified random sample of 92,332 households was drawn from the municipal register of inhabited dwellings, excluding officially registered shared and institutional housing, such as student units with shared kitchens or nursing homes (Booi, 2016). Stratification entailed dividing Amsterdam into 85 small neighbourhoods. The questionnaire was delivered by post, with the possibility to complete it on paper or online, in Dutch or in English. In areas where the response initially was not sufficient, extra efforts through phone calls and calling at houses were undertaken. This resulted in 18,920 people completing the survey, which translates to a total response rate of 20.5%. For our analysis, households of which it could not be determined whether they rented with a permanent contract, a temporary lease or whether they were home-owners were removed. This resulted in a final data file with 17,803 respondents.

The low response rate is unfortunate, but WIA is the only feasible option at hand. From the population and dwelling registers, we have an indication⁹ of some of the sources of under- and overrepresentation. For example, while slightly more men than women appear in the sample (51.4% vs. 48.6%), according to the municipal statistics department Research, Information and Statistics (RIS) the city is actually home to fewer men than women (49.3% vs. 50.7%; RIS, 2015). Also, those of younger ages are underrepresented: 10.4% of the population falls in the age category 18–23, but in the sample this category only holds 2.5%. Of the age category 24–29 the population holds 15%, but the sample only 7.3%. Furthermore, owner-occupiers are overrepresented: in 2015, 28.9% of the dwellings were owner-occupied (RIS, 2016) rather than the 40.8% we find in the sample.

Another drawback from the survey is that it attempts to exclude shared accommodation – while temporary leases could very well be overrepresented in such accommodation. However, since registration of such units is incomplete, some people who share with individuals other than partners or children still appear in the sample. Furthermore, given that many people with an anti-squat contract live in vacant office buildings, anti-squat is likely to be underrepresented because non-residential buildings are excluded from the sampling frame. For reasons of socially acceptable answer patterns, we also assume that subletting will be underreported. In all, those with temporary leases are almost certainly underrepresented in the data. This implies our estimation of the share of temporary leases in Amsterdam will be conservative. We have considered using weighted data to correct for selective nonresponse, but decided against this because the numbers having temporary leases by category of respondent are rather small and applying large weights for some categories might ‘blow up’ to accidentally high numbers in such categories.

4.2. Measures

All variables were derived from the survey, except for international migration background which was retained from the data that formed the sampling frame (obtained from the municipal statistics department RIS). The dependent variable has three categories: having a temporary lease, having a permanent lease, and home-ownership. Age was measured in years. Those enrolled in higher education were coded as currently fulltime in higher education. For education level, we used the highest obtained qualification. Lower education included primary school, pre- and short vocational secondary education (vmbo/mbo-kort), middle education encompassed longer vocational education, high school and pre-university education (mbo-lang/havo/vwo), higher education included universities and universities of applied sciences. For the variable international migration background we considered all who had at least one parent who was born outside the Netherlands. We follow Statistics Netherlands (2019f), who split this category up into the two subcategories of Western and non-Western migration backgrounds. They define Western countries as Europe (excluding Turkey), North America, Indonesia and Japan, and non-Western as all other countries. For household composition we considered those who indicated that they formed a household with someone who is not their partner or their child, as sharing. For source of income, transfers included benefits, study loans and pensions. We split household income into lower, middle and higher incomes. For lower incomes, the upper threshold is the maximum income to be eligible for social housing (gross €34,911 per year in 2015), middle incomes include up to 1½ times modal income (gross €49,500 per year in 2015), according to the standard Dutch policy indicator (modal = gross €33,000 per year in 2015) and higher incomes include all above this. Income was self-reported, and in line with general survey trends, almost a quarter of the respondents declined to answer. We coded these as a separate category. We also ran analyses of income as a numerical variable, and with those who for which no data on income was available excluded (not shown here, but available on request). The results were very similar to those we present, with only small changes in parameters and p-values. The variables on reasons for moving (‘rental contract was terminated’ and ‘home was too expensive’) are based on questions concerning reasons for moving to the current home.

4.3. Method

Because the dependent variable is categorical, we employ multinomial logistic regression models. We set permanent lease, the most common tenure, as the base outcome. The model is given as: $$\text{log}\hspace{0.17em}\left(P_j/P_{\mathit{permanent}\mathrm{ }\mathit{lease}}\right)\mathrm{ }=x{b}_j$$ where x is a vector of characteristics for a given individual, P is the probability for an individual of being in a certain housing situation j (j = 1,2; 1 for temporary rent and 2 for home-ownership) and b_j is a vector of unknown parameters.

Although home-ownership is not our main focus, we think it is important to show results on it because it is the second most common tenure in Amsterdam. Because students are a specific population category with a large proportion having a temporary lease, we show the outcomes of two models: one for the complete sample, and another from which those currently fulltime enrolled in higher education (n = 905) are excluded.

5. Results

5.1. Descriptives: who rents with a temporary lease

The main characteristics of the people in the sample and the percentages living in the different tenure types by category of the independent variables can be found in Table 4. Overall, 4.4% of the sample has a temporary lease, versus 54.8% with a permanent lease and 40.8% who is owner-occupier. Concerning those with temporary leases, more women than men are in this category. To provide more insight into the distribution of age in the sample and the distribution of housing tenure by age, we provide for the descriptive analysis not only the average age, but also split up age into 5 categories. The distribution over the age groups is very uneven. The majority of the youngest part of the sample turns out to rent with a temporary lease, namely 52% of those between 18–23 years. In the next category, containing people between 24 and 29 years, the proportion drops to 22%. This tapers off to half a percent of those over 50 years. The average age of the whole sample is 51 years, with a standard deviation of 16.00. For temporary lease, the average age is 29 years (SD: 10.49), for those with a permanent lease 54 years (SD:16.30) and for home-owners 50 years (SD: 14.16).

Table 4. Descriptive statistics: Who has which tenure in Amsterdam.

Variable	Sample count	Sample percentage	Temporary lease percentage	Permanent lease percentage	Home-ownership percentage
Total sample	17,803	100.0	4.4	54.8	40.8
Gender
Female	8648	48.6	5.2	58.7	36.2
Male	9155	51.4	3.7	51.1	45.2
Age
18–23 years	451	2.5	52.3	38.6	9.1
24–29 years	1304	7.3	22.0	50.6	27.4
30–39 years	3189	17.9	4.7	47.3	48.0
40–49 years	3235	18.2	1.9	46.9	51.2
50–95 years	9624	54.1	0.5	61.2	38.2
Currently fulltime in higher education
Yes	905	5.1	40.0	38.6	21.4
No	16,898	94.9	2.5	55.6	41.8
Level of education
Lower education	3729	20.9	1.6	83.9	14.6
Middle education	3230	18.1	6.3	61.2	32.5
Higher education	10,126	56.9	4.9	40.4	54.8
No answer	718	4.0	4.6	78.0	17.4
International migration background
No migration background	11,383	63.9	4.0	48.7	47.3
Western migration background	2637	14.8	6.4	49.4	44.2
Non-Western migration background	3783	21.2	4.4	76.7	18.9
Household composition
One-person household	8140	45.7	5.5	62.4	32.1
Couple without children	4793	26.9	3.9	45.2	50.9
Couple with children	3212	18.0	1.6	42.1	56.3
Single parent	1152	6.5	1.7	72.2	26.0
Sharing with others	506	2.8	16.4	63.6	20.0
Main source of income
Employed	8187	46.0	4.8	43.6	51.5
Self-employed	2295	12.9	3.0	40.7	56.4
Transfers	6140	34.5	4.8	71.2	24.0
No answer	1181	6.6	2.7	74.0	23.3
Household income
Low income	5962	33.5	6.6	76.2	17.2
Middle income	2351	13.2	2.5	47.0	50.5
High income	5228	29.4	2.6	26.5	71.0
No answer	4262	23.9	4.7	63.8	31.4
Reason for moving: rental contract terminated
Yes	579	3.3	17.6	54.1	28.3
No	17,224	96.7	4.0	54.8	41.2
Reason for moving: home was too expensive
Yes	451	2.5	9.3	51.7	39.0
No	17,352	97.5	4.3	54.9	40.8
Location previous home
Amsterdam/ never moved	14,132	79.4	3.1	56.1	40.8
Elsewhere in the Netherlands	2863	16.1	8.3	45.8	45.9
Abroad	539	3.0	18.0	59.2	22.8
No answer	269	1.5	3.0	72.9	24.2

A substantial share of those currently in higher education have a temporary lease, namely 40%. Those with a middle level of education, and, to a lesser extent, those who finished a form of higher education, are somewhat overrepresented in the temporary lease category. This can be partly explained by the overlap with those who are currently in higher education: they have mid-level education levels as their highest qualification. Slightly more people with a Western migration background have a temporary lease than those with a non-Western background and those with no immigration background. Though, as described above, the survey attempts to exclude those in shared accommodation, still 2.8% of the respondents indicate they are living together with others who are neither their partners nor their children. Within this category, 16.4% has a temporary lease. Concerning income, people with temporary leases are somewhat overrepresented in the low-income group, which indicates that on the basis of their income alone, they would be eligible for homes with regulated rents. While at a lower rate, people with temporary leases can also be found in the middle and higher income categories.

Considering the link between the reasons for moving and temporary leases, of those who had to move because their rental contract was terminated or their dwelling had become too expensive, respectively, 17.6% and 9.3% had a temporary lease as their next tenure. Finally, almost 80% of those in the sample states their previous home was in the city. Those who moved in from elsewhere in the Netherlands or abroad are overrepresented in the category temporary leases with 8.3% and 18.0%.

In terms of affordability, the large majority of households with permanent leases, pay a monthly net rent between 400 and 710 euros (a regulated rent). Very few households pay less and those paying free-market rents of over 710 euros remain a minority. For households with temporary leases, the picture is reversed. Here, especially those with a campus, youth or antisquat contract usually pay less than 400 euros per month. Between 400 and 710 euros is less common for those with temporary leases. At the higher end of the spectrum, those with temporary leases are again overrepresented. This should be interpreted in the context of the distribution of regulated housing according to time on the waiting list combined with the maximum income threshold mentioned above. A picture arises of temporary leases on the one hand as an alternative for those with low incomes, who do not yet have enough time on the waiting lists, and on the other hand, luxury furnished apartments for affluent households.

5.2. Analysis: temporary leases versus permanent leases

5.2.1. Model for the full sample

According to the multinomial logistic regression for the full sample (Table 5), there is a statistically significant negative association between age and the chance to have a temporary renting contract: with every additional year of age, the odds of having a temporary rather than a permanent lease are estimated to be 0.900 times as high. This ties in with our expectations: more senior citizens are more likely to be more attached to security of tenure, and they will have more resources available to access permanent leases or owner-occupation. Compared to men, women have a significantly lower chance to have a temporary lease rather than a permanent lease (odds ratio: 0.780, implying that women’s chances are 0.780 times men’s). We speculated this might be so, because women might be more attached to secure housing and they might be in a better position to obtain this as well. Those who are currently in higher education have a significantly higher chance to have a temporary lease than those who are not (odds ratio: 3.805). Again, this confirms our hypothesis, we expected students to have a higher chance of accepting temporary rent, given their need to be close to their institutions and their restricted resources. However, people’s highest completed level of education does not seem to influence the chance of having a temporary lease. We expected that those with lower education would have a higher chance of accepting temporary leases, given that they have to compete on the housing market with those with higher education levels, who could be viewed as more attractive tenants or more likely to obtain mortgages. Perhaps the system of distributing social housing through waiting lists counters such an effect. As expected, those with a Western migration background have a significantly higher chance to have a temporary contract than those who do not have a migration background.

Table 5. Multinominal logistic regression: predicting the likelihood of temporary leases in Amsterdam.

		Model 1 students included
	Temporary lease			Home-ownership
Predictor	B	SE	Odds ratio	B	SE	Odds ratio
(Intercept)	1.372	0.279		−2.389	0.113
Gender: Female	−0.248***	0.091	0.780	−0.162***	0.038	0.850
Age	−0.106***	0.005	0.900	0.013***	0.002	1.013
Currently fulltime in education	1.336***	0.124	3.805	−0.014	0.108	0.987
Level of education (ref: lower education)
Middle education	−0.183	0.183	0.833	0.867***	0.067	2.381
Higher education	0.157	0.169	1.170	1.351***	0.059	3.860
No answer	0.165	0.268	1.180	0.193*	0.116	1.213
International migration background (ref: no)
Western migration background	0.549***	0.121	1.731	−0.098*	0.052	0.906
Non-Western migration background	0.117	0.120	1.124	−0.791***	0.055	0.453
Household composition (ref: one-person household)
Couple without children	−0.077	0.117	0.926	0.127**	0.048	1.135
Couple with children	−0.666***	0.170	0.514	0.434***	0.058	1.544
Single parent	−0.868***	0.249	0.420	−0.127	0.085	0.881
Sharing with others	−0.444***	0.163	0.642	−0.705***	0.131	0.494
Main source of income (ref: employed)
Self-employed	0.146	0.150	1.157	0.023	0.056	1.023
Transfers	0.113	0.111	1.120	−0.804***	0.054	0.448
No answer	−0.085	0.220	0.918	−0.677***	0.086	0.508
Household income (ref: low income)
Middle income	−0.333**	0.163	0.716	1.152***	0.058	3.166
High income	0.082	0.137	1.086	1.754***	0.056	5.776
No answer	−0.146	0.114	0.864	0.807***	0.053	2.241
Reason for moving: rental contract terminated	0.612***	0.145	1.845	−0.643***	0.110	0.525
Reason for moving: home was too expensive	1.040***	0.212	2.829	0.070	0.117	1.072
Location previous home (ref: Amsterdam/ never moved)
Elsewhere in the Netherlands	0.249**	0.107	1.283	0.125**	0.050	1.133
Abroad	1.062***	0.164	2.893	−1.001***	0.122	0.368
No answer	0.451	0.417	1.570	−0.330**	0.166	0.719

Notes: model 1: n = 17,803 model 2: n = 16,898.

* p < .10; **p < .05; ***p < .01, the reference category is permanent lease.

Couples with children, lone parents and those living with others have a significantly lower chance of having a temporary contract, compared with those who live alone. This is in line with our hypothesis. Contrary to expectation, we found no difference between couples living together and one-person households. We considered that more spending power and more difficulty in relocating as a couple would lead to a lower chance of accepting temporary lease, quod non. However, we do find an enhanced likelihood of owning a home for couples without children. It could be that the Amsterdam housing market is so tight that mainly those couples who can either afford to own or are prepared to rent temporarily stay in Amsterdam, while others tend to leave the city.

While we surmised that being employed would lower the chance of having a temporary lease, and being self-employed or relying on transfers might increase the chance, people’s main source of income is not significantly associated with their chance of having a temporary lease. Similarly, we expected those with lower incomes to have an increased chance of having a temporary lease. There is indeed a significant negative effect for those with a middle income, indicating a lower chance for them to have a temporary renting contract than those with a lower income. However, those with higher incomes are not found to have a lower chance than those with the lowest income. Some of those with high incomes might live in luxury or expensive furnished apartments with a temporary lease, but we do not have the information necessary to explore this idea.

Those who had to leave their previous home because their renting contract was terminated and those who moved because their home had become too expensive have a significantly higher chance of having a temporary contract than those who moved for other reasons, with odds ratios, respectively, of 1.845 and 2.829. This confirms our idea that people are more likely to accept temporary leases when they have an urgent need. Finally, compared with those who already lived in Amsterdam, those who moved from elsewhere in the country or from abroad, have a significantly higher chance to end up in temporary lease. This is in line with our expectation, based on the idea that those who lack a social network will be more likely to end up with a temporary lease.

5.2.2. Model with those in full time higher education excluded

With students excluded (Table 6), most of the associations found in the model for the full sample persist, but there are a few exceptions. In this model, the negative association between gender and the chance of temporary rent is slightly smaller and not statistically significant. Also in contrast with the model with the full sample included, here, those with middle levels of education have a marginally significantly lower chance of having a temporary contract compared with those with a low level of education (.05 < p < .10). This might be because in the first model, students mostly form part of the category middle education, since we measured the highest level of education achieved. However, still no effect is visible for those with a high level of education.

Table 6. Multinominal logistic regression: predicting the likelihood of temporary leases in Amsterdam.

		Model 2 students excluded
	Temporary lease			Home-ownership
Predictor	B	SE	Odds ratio	B	SE	Odds ratio
(Intercept)	0.675	0.306		−2.355	0.116
Gender: Female	−0.179	0.109	0.836	−0.169***	0.002	0.845
Age	−0.097***	0.005	0.908	0.012***	0.039	1.012
Level of education (ref: lower education)
Middle education	−0.357*	0.209	0.700	0.849***	0.067	2.338
Higher education	0.230	0.178	1.258	1.357***	0.060	3.884
No answer	0.254	0.278	1.290	0.180	0.117	1.197
International migration background (ref: no)
Western migration background	0.610***	0.135	1.840	−0.104 **	0.053	.901
Non-Western migration background	0.195	0.143	1.216	−0.812***	0.057	.444
Household composition (ref: one-person household)
Couple without children	0.223	0.137	1.250	0.130***	0.049	1.139
Couple with children	−0.388**	0.178	0.679	0.449***	0.059	1.567
Single parent	−0.497**	0.253	0.608	−0.111	0.086	0.895
Sharing with others	0.146	0.227	1.158	−0.704***	0.144	0.495
Main source of income (ref: employed)
Self-employed	0.140	0.163	1.151	0.022	0.057	1.022
Transfers	0.091	0.156	1.095	−0.792***	0.056	0.453
No answer	0.143	0.227	1.154	−0.679***	0.087	0.507
Household income (ref: low income)
Middle income	−0.061	0.178	0.941	1.160***	0.059	3.190
High income	0.178	0.154	1.195	1.770***	0.057	5.868
No answer	0.010	0.147	1.010	0.836***	0.055	2.308
Reason for moving: rental contract terminated	0.738***	0.168	2.091	−0.703***	0.115	0.495
Reason for moving: home was too expensive	1.180***	0.228	3.256	0.081	0.119	1.084
Location previous home (ref: Amsterdam/ never moved)
Elsewhere in the Netherlands	0.215	0.139	1.239	0.120**	0.052	1.128
Abroad	1.061***	0.178	2.891	−1.054***	0.125	0.349
No answer	0.381	0.441	1.464	−0.352**	0.167	0.703

Notes: model 1: n = 17,803 model 2: n = 16,898.

* p < .10; **p < .05; ***p < .01, the reference category is permanent lease.

The significantly negative associations between households living with children, either a couple or a single parent, are weaker than in the model for the full sample. Students seldom already have children, this likely fortified the effect in the model with students included. The significant negative relation between sharing with others and having a temporary lease vanishes in this second model. It could be that especially students team up to rent together in order to obtain a permanent contract. Also, in the model with students excluded, middle incomes cease to be significantly negatively associated with whether people have temporary leases. Perhaps the fact that students usually have a low income explains this. Finally, the positive effect for those whose previous home was elsewhere in the Netherlands is smaller than in the model for the full sample and no longer statistically significant. The significantly greater chance for those whose previous home was abroad nevertheless persists.

5.3. Analysis: home-ownership versus permanent leases

As expected, on several points, the chances for owner-occupancy mirror those for temporary rent. The chance of owning a home, for instance, significantly increases with age (odds ratio 1.013), and couples with children are particularly likely to own (odds ratio 1.378). Those who relocated from abroad, and those moving because their lease came to an end, are unlikely to own their home. This falls in with our ideas, since upon arrival from foreign parts a temporary lease might be more attainable as well as more practical than immediately buying a home.A cluster of variables indicating socio-economic status is significantly associated with chances of home-ownership but not with the chances of having a temporary lease: having high education levels and a middle or high income is positively associated with home-ownership, whereas the parameters for having a non-Western migration background and for transfers as the main source of income are negative. This is also in line with expectations.

For two characteristics, significant effects are found in the same direction for temporary leases as for home-ownership. Women have a significantly lower chance for owner-occupancy while (in the full sample) they are also less likely to have a temporary contract. Those whose previous home was elsewhere in the Netherlands are more likely to own than those who already lived in Amsterdam, or to have a temporary rental contract. This may reflect their lack of access to the regulated rental market, which may lead them to resort to less regulated housing options: temporary rent and owner-occupancy.

6. Conclusion

In this article, we have made the first step into assessing what increases the chance of having a temporary lease, in this case in Amsterdam. We argue that given the importance of secure housing for people’s wellbeing, the recent shift in the Netherlands from permanent contracts to time-limited leases deserves attention. Although permanent contracts are still dominant, we find that the majority of young adults aged 18–23 are renters with a temporary lease. From a multinomial logistic regression analysis, we found that students, those with a Western migration background, those who moved because their previous rental contract was terminated or because the previous dwelling was too expensive, and those who moved from abroad were particularly likely to have a temporary lease. Households with children were unlikely to have a temporary lease.

It is important and urgent to analyse housing insecurity because there are indications that the phenomenon is on the rise in several European countries. While in Anglosaxon countries temporary contracts have become the norm with private landlords since the 1980s, now those in social housing are also confronted with temporary contracts. At the same time, more households rent, both in Europe as in Anglosaxon countries, and they rent from a private landlord, since governments are promoting buy-to-let. The contribution of this article is that it analyses the size and the characteristics of the households renting in the temporary sector. The exploration of motives and opportunities provides a deeper understanding of theoretical elements.

The contribution of this research is that it presents, for the first time, data on who has a temporary lease and what increases the chance of having such a contract. The limitations of the current study are a reflection of the problems with the limited availability of meaningful data. The response rate is rather low, and data from other sources showed there exists a bias towards home-owners and other more advantaged groups, which influences the results. The estimation of the number of those with temporary leases will be conservative. The design of the survey also limited the analysis, since some relevant variables are not measured. For instance, no information is available about divorce or relationship breakup. The next instalment of the survey will remedy some of these concerns.

One of the goals of the current study is to create a baseline. Temporary leases as a normal form of tenure were introduced in the Netherlands in 2016, and it took landlords some time to come to grips with the new possibilities. The current research allows us to trace the development of this new contract form over time. This will also give some indications as to whether temporary rent will be a phenomenon mainly connected to age, or that a cohort effect will occur. Will temporary rent become a new rung on the housing ladder, related to the earlier stages of the life-course, with temporary leases as the lower rung, permanent rent as the middle and owner-occupation as the top rung? Or will the development be more analogous with that of the labour market, and will youngsters of the current generation never obtain a permanent renting contract?

Concerning policy, the recommendation is to start to seriously register and analyse temporary rental arrangements in the Netherlands, and to use this as input for new policy. While this might sound obvious, in the context of temporary leases we observed that further extensions are repeatedly being implemented while the results of the previous round of changes are only starting to become clear. Policies based on factual data rather than on intuitions hopefully will deliver better results.

The first findings indicate that temporary leases might well be more than a marginal phenomenon. As we hypothesized, temporary leases might be one of the only options for newcomers on the housing market, may they be young adults, or moving from within or without the Netherlands. Those with an urgent need seem also have to rely on temporary leases, given other options are out of reach.

Acknowledgements

We thank the Municipality of Amsterdam and the Amsterdam Federation of Housing Corporations for granting us access to the WIA data. We are grateful to Kees Dignum and Vera Berkers of the Department of Housing of the Municipality of Amsterdam for accommodating us at their offices for the data analysis. The comments of five anonymous reviewers helped to strengthen the manuscript, as did Steven Kelk’s remarks.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

Additional information

Funding

The research was funded through an Ubbo Emmius Award of the Faculty of Spatial Sciences at the University of Groningen.

Notes on contributors

Carla J. Huisman

Carla J. Huisman is a sociologist specialised in social-spatial inequality. Her current Research interests are the precarisation of housing markets, as well as the influence of spatial clustering on integration outcomes of recent refugees. She is a postdoctoral researcher at the Technical University of Delft.

Clara H. Mulder

Clara H. Mulder is professor of Demography and Space at the Population Research Centre at the University of Groningen.

Notes

1 Since we focus on situations where a home is rented out to be the primary residence of the tenant, we exclude short-term holiday rentals such as Airbnb.

2 With regard to private renting, first Scotland and to a lesser extent Wales and Northern Ireland seem as of late to be inching towards a slightly less unregulated sector in terms of landlord registration, the curbing of ‘no-fault eviction notices’ and rent controls (Moore 2017), while England very recently and unexpectedly seems to want to fall in with this trend (Elgot, 2019).

3 The average proportion of their income renters spent on housing expenses rose from 28.3% in 1990 to 38.8% in 2015. Owner-occupiers spent 20.5% in 1990, and 28.3% by 2015 (Statistics Netherlands 2019b, 2019c).

4 In addition, to encourage households with average incomes to move into owner-occupancy or homes with free-market rents, steep income-dependent rent increases have been introduced yearly since 2013 (Van Duijne & Ronald 2018).

5 Neither sharing nor furnished lodgings find their way into the table: they are not forms of tenure. Two friends can buy or rent an apartment together, and it is possible to buy a furnished apartment, or let one.

6 Examples are tenants in precarious renting situations who are denied the possibility of having a garden (Darab et al. 2018), pets (Power 2017) or starting a family (Heijkamp & Borštnik 2016).

7 For instance, relation break-up or divorce, eviction as a result of rent arrears or defaulting on the mortgage all create an immediate need for alternative housing. Similarly, temporary contracts themselves bear a negative-compulsory dimension: to move house may no longer be a decision per se, but rather the result of a landlord terminating a contract or the expiration of a fixed-term.

8 The resulting survey data are in general not available for analysis by others; ‘The use of the file is reserved for the municipality of Amsterdam and the Amsterdam housing corporations’, codebook 2015.

9 Since the population register does not include undocumented or unregistered inhabitants, it does not provide a complete picture of the population of Amsterdam.