THE MISMATCH OF HOUSING COSTS AND INCOME IN DUTCH HOUSING Frans M. Dieleman and Ronald van Kempen

1 Introduction Almost all countries in Western Europe have built up an extensive social rental sector since 1945. The main reason was to provide decent housing for people with a modal or low income. Yet this does not imply that social rental units are always or exclusively occupied by households of these income categories. The Netherlands is a case in point. Although more than 40 percent of the country's dwellings belong to the social rental sector, these units provide housing for a large share of the households with an income above modal. The Dutch government has actively promoted the function of the social rental sector for a wide range of income groups. Having done so since 1964, this policy has helped prevent the concentration of the poorest population in the social rental sector. Diverse household types seek dwellings in that sector; the units are desirable and the occupants are satisfied. The social rental sector in the Netherlands is not stigmatized as a housing market segment destined for the have nots. In the 1980s, the perception of how the social rental sector should function changed. This shift occurred in response to the need to cut back on government expenditure, which put pressure on the field of housing as well. Henceforth, the sector would primarily serve the priority categories of the policy. Thus, it was to provide shelter for households with a low to modal income, t Then, in the mid-1980s, a policy based on this standpoint was implemented under the Secretary of State for Housing, Mr. Heerma. A government white paper, "Housing in the Nineties" (MVROM, 1989) ushered in this policy. And this document introduced the theme of mismatch on the housing market. The concept of mismatch means that (a) inexpensive rental dwellings are occupied by households with a relatively high income, and that (b) expensive rental dwellings are occupied by households with a relatively low income. The aim of the policy is to reduce both forms of mismatch. In the first form, many social rental dwellings are built with subsidies for households that do not really need that assistance. In the second form, mismatch provokes individual households to ask for a housing allowance.

Neth. J. of Housing and the Built Environment, Vol. 9 (1994) No. 2. 159

This article deals with the theme of mismatch on the Dutch housing market. It is based on an investigation we have recently completed (Dieleman, Hooijmaijers and Van Kempen, 1993). First, we give a brief review of the nature of housing policy in the Netherlands in the past, with special attention for changes since the white paper on "Housing in the Nineties" appeared in 1989. Against that background, Sections 4, 5, and 6 discuss consecutively trends in the mismatch, national policy, and local responses to this situation. The article ends with an evaluation. There, we tally the advantages and the drawbacks of trying to rectify housing mismatch.

2 The state and the housing market All Western European countries intervened in their housing markets after the Second World War. They took a wide range of measures, including rent control and a variety of subsidies (Lundqvist, 1992). Almost all countries created a social rental sector as well. The Netherlands went the farthest in this respect, by West European standards. The large housing shortage in the Netherlands has almost continually set the agenda for governmefit policy. The 1947 Census of Population revealed a shortage of about 300,000 dwellings. Numerous measures were introduced -- rent control, housing distribution, subsidized loans to housing associations and municipalities, and property subsidies for the construction of new dwellings -- to alleviate the housing need and to restrict its negative effects (Van Weesep and Van Kempen, 1993; Meusen and Van Kempen, 1994). The property subsidies were primarily used to expand the social rental sector. Diverse attempts were made in the 1960s to return to a more free housing market. However, two situations arose, making it politically inadvisable to reduce state influence on the housing market (Van der Schaar, 1987). First of all, the number of households had grown very rapidly, especially in the second half of the decade, as a result of several factors: (a) the postwar baby boom generation had come of age; (b) families were getting smaller; and (c) immigration was rising. Therefore, the housing shortage remained large, retaining a high priority on the political agenda. Secondly, the effort to build sufficient dwellings in the non-subsidized sector was undermined from time to time by unfavorable market conditions. For instance, at the end of the 1960s, the interest rates on long-term loans were high. Moreover, construction costs had risen sharply. These factors led to a decline in non-subsidized housing -completions. The government intervened with an anti-cyclical construction policy. This entailed stepping up the building program for subsidized housing. As it turned out, the government was to do this again in the early 1980s. During the 1950s and 1960s, there was continuous state intervention in the housing market. This was the most powerful engine of growth for the social rental sector in that period. From 1961 to 1970, over a million new dwellings were built. Slightly more than a fourth of these were not subsidized (Van Weesep and Van Kempen, 1993). Rogmans (1991) points out that the conditions prevailing in the early 1970s allowed the state to take a lesser role in the housing market. The rising affluence of Dutch society was spread widely throughout the population. In combination with increasing 160

Table 1 Individual housing allowance 1975-1991 Period 1975 1979 1984 1988 1991

Number of recipients

Total budget (million Nlg)

348,000 395,000 715,000 880,000 953,000

339 473 1,271 1,551 1,779

Source: MVROM (1992d).

access to mortgage loans, from the end of the 1960s on, the new affluence sparked growth in home ownership (Harloe and Martens, 1985). The proportion of owneroccupier dwellings in the new construction rose from 40 percent in 1974 to 60 percent in 1978. The government used subsidies to promote new construction in this sector as well (Dieleman and Jobse, 1991). But the homeowner market collapsed at the end of the 1970s. Again, the state intervened, providing huge subsidies for social housing construction. By 1984, only 30 percent of new completions were owner-occupier units. Between 1971 and 1980, 1.2 million new dwellings were put up; less than a quarter of these were built without subsidy. Thus, the expansion of the social rental sector kept going at the same pace in the 1970s and into the early 1980s. By the end of the 1980s, the tenure structure of the Dutch housing stock was as follows: 45 percent owner-occupier; 41 percent social rental; and 13 percent private rental (MVROM, 1992a). The government introduced a system of individual housing allowances (IHS) in the 1970s, as part of the expansion of the Welfare State (Dieleman, 1994). This system was supposed to give households with a relatively low income the opportunity to move into dwellings that were fairly new, of good quality, and (consequently) relatively expensive. During the 1980s, the rents were increased sharply, even though income growth was slight in this period. Not surprisingly, the number of recipients of housing allowances swelled to a torrent (Table 1). By the end of the 1980s, more than 25 percent of all renters received a housing allowance.

3 Current housing policy The Dutch government was slower than other West European countries to make fundamental changes in its national housing policy (Harloe and Martens, 1985; Boelhouwer and Priemus, 1990; Lundqvist, 1992). The Dutch state did cut back the property subsidies and the housing allowances, and hefty rent hikes were carried out in the 1980s. But the relations between the national government and the social rental sector, for instance, did not change fundamentally in that decade. 161

Various reasons have been given for the Netherlands' tardy moves to make radical changes in the national housing policy. Van Kersbergen and Becker (1988) and Lundqvist (1992) explain the delay primarily in terms of the basic organizing principle of Dutch society: 'columnization'. This means that civil society replicates the sectarian 'columns' of Catholicism, Protestantism, and secularism. Thus, the constituency of each political party is a vertical cross-secti0n of society. And each party has a vested interest in maintaining strong government support for housing. Therefore, none of the parties were able to win votes by launching an attack on the continued existence of the social rental sector. Yet other analysts point out that many organizations have been established in the Netherlands on the borderline of the public and the private sector (Andeweg, 1988; Silver, 1990). These organizations are fairly well armed to withstand sudden changes in government policy. Housing associations form a case in point. Although these are non-profit organizations and as such regulated by law, they are also recognized by law as private enterprises (Priemus, 1990). Fundamental changes in the national housing policy only became manifest when the policy outlined in the government white paper "Housing in the Nineties" (MVROM, 1989) was implemented in the early 1990s. It goes beyond the scope of this paper to make an exhaustive analysis of this about-face in Dutch housing policy. But in broad strokes, the changes may be summarized as follows (see also Van Kempen and Van Weesep, 1994): 1. The national government has reduced its responsibility for housing; this has been largely transferred to the municipalities and housing associations. 2. Construction subsidies for new social rental dwellings have been reduced (and will be terminated as of 1995). The housing associations have to dip into their own reserves to finance maintenance on existing housing estates. For new construction, they have to fired funding on the capital market. These financing constraints partly explain why the share of social rental dwellings in new housing construction declined at the end of the 1980s. 3. The national government conducts a rental policy that is geared to help the housing associations keep operating their huge social rental sector, largely without government support. 4. The system of housing allowances is retained, at least for the time being, in order to ensure that people on a low income will have access to decent housing. In the white paper on "Housing in the Nineties" (MVROM, 1989), the problem of mismatch on the housing market is identified as one of the main causes of the high costs of rental housing in the Netherlands. If households with a high income who occupy social rental units were to move on to the owner-occupier sector or into more expensive rental dwellings, then many inexpensive social rental units would become available. In that event, less new construction would be necessary in the social rental sector. Moreover, the demand for housing allowances (Individuele Huur Subsidie, IHS) could be reduced. The following sections elaborate on the policy that is supposed to address the problem of mismatch on the housing market.

162

4 Mismatch in the rental sector: characteristics and developments In the Netherlands, the social rental sector is huge, whereas the private rental sector has declined sharply under the influence of government policy after the Second World War. Accordingly, the types of household that occupy the social rental sector are highly diverse. In fact, nearly everyone who prefers to rent rather than to buy is relegated to the social rental sector. For this reason, the social rental sector houses many people, both young and old, with a relatively high income (Dieleman, 1994). Consequently, many people with a fairly high income pay a relatively low rent. This is called 'inexpensive mismatch'. In contrast, there are many people with a tow income who live in an expensive rental dwelling. They are able to do so by taking advantage of the opportunity for a housing allowance (IHS). This situation is called 'expensive mismatch'. A policy has been enforced since the end of the 1980s at the national level to reduce the problem of mismatch. Inexpensive mismatch 2 For the Netherlands as a whole, the incidence of inexpensive mismatch increased over the period 1986-1990, both in relative and absolute terms. In 1986, 34 percent of all inexpensive rental units were occupied by households with an income above the normative rent. By 1990, the inexpensive mismatch had risen to 38 percent. Whereas 21 percent of all renters paid less than the normative rent in 1986, this figure had risen to 23 percent in 1990 (MVROM, 1992b, p. 9). In absolute numbers, this was 641,000 households in 1986 and 726,000 in 1990 (MVROM, 1992d, p. 12). The increase of 85,000 households had several antecedents: (a) it was the result of moves; (b) it may be attributed to rising incomes among households in inexpensive rental units; and (c) it reflects adjustments in definitions (see Fig. 1).

Figure 1 Mismatches in inexpensive housing in 1986-1990 640,000 households

730,000 households

with housing costs

with housing costs below the norm

below the norm

Moves to dwellings with rent meeting the norm

120.000

Moves without rectifying mismatch

60.000

Immobility of mismatched households

435.000

= 145,000

j

ii!~i~:i~ ~

~

60.~'~

housing market resulting in rents below the norm Moves without rectifying mismatch

Immobility and

Exits from the housing market

490.000

income growth exceeding cost increases

Moves to dwel-

lings with rents below the norm

Source: MVROM, 1992b 163

(a)

(b)

(c)

During the period 1986-1990, an estimated 3 205,000 households with housing costs below the norm moved out of their dwelling. This represents 30 percent of the total number of inexpensive mismatches. Of these movers, 59 percent moved into a dwelling where the housing costs were in accordance with their income level. Furthermore, 29 percent moved on to another inexpensive rental dwelling. For the remainder (12 percent), an inexpensive rental dwelling was vacated due to decease or institutionalization of the occupant (MVROM, 1992b, p. 10). In the same period, there was an influx of 240,000 households whose housing costs were below the norm. Sixty percent of the new inexpensive mismatches were starters on the housing market. Consequently, their moves did not yield a vacated dwelling. Another 25 percent of the inexpensive mismatches moved without changing their ratio of rent to income. The remaining 15 percent moved out of a dwelling where the housing costs met the norm and into a unit where the rent was below the norm for their income level. Between 1986 and 1990, 55,000 households who had been living in rental units with housing costs that met the norm saw their incomes rise to the extent that they no longer fell into the priority categories. In other words, their housing costs met the norm in 1986 but were too low for their income in 1990 (MVROM, 1992b, p. 10). The rising incidence of inexpensive mismatch is also a reflection of the procedures applied to indexing of rents and incomes (Maas, 1992). Between 1986 and 1990, the incomes were adjusted for inflation, and the rents were raised by parliamentary decision to diminish the property subsidies. The percentage of the rent increases was actually much higher than the rate of inflation. Therefore, the increase in the number of households whose housing costs are below the norm is somewhat artificial.

The incidence of inexpensive mismatch in 1990 is thus largely due to the fact that many households did not move. On the one hand, they may have had a high income in 1986 and thus already have been included among the inexpensive mismatched households. On the other hand, their income may have grown to the extent that they came to be counted among the inexpensive mismatched households in the course of time. At the same time, a sizeable number of households with housing costs below the norm left the rental sector in a short period of four years. If the movers had only been allocated dwellings that met their normative housing costs (according to the white paper on "Housing in the Nineties"), then the inexpensive mismatch would have been reduced significantly during that period. Many starters, in particular, were allocated a dwelling with a rent that was too low with respect to their income. Apparently, the conditions on the housing market do not always permit an appropriate allocation (as defined in "Housing in the Nineties"). Alternatively, the gatekeepers of the rental sector may not always consider this an important aim. We return to this point in the next section. Young people (under 40 years of age) and households in which at least one person derives income from wage labor are over-represented among households paying less 164

rent than they can afford. That group includes relatively few elderly persons (aged 65 and over). Numerous people in this group, both in relative and in absolute terms, live in housing estates that were built in the period 1945-1970. The over-representation of younger households among those who could afford to pay a higher rent is in accordance with the observation that many starters are allocated an inexpensive rental unit. This suggests that the phenomenon of inexpensive mismatch takes root at the time the dwelling is allocated (MVROM, 1992c). Expensive mismatch 4

The incidence of expensive mismatch decreased somewhat between 1986 and 1990; it dropped from six to five percent of the total number of renters (MVROM, 1992d, p. 5). This decline is partially the result of a rise in income among low-income households who live in expensive rental dwellings. At present, we cannot say whether or not a more stringent allocation policy and the enforcement of the so-called approval threshold (as of January 1, 1989) for eligibility for housing allowance (IHS) will actually have an effect. At the time it was introduced (1989), the approval threshold meant that in principle, no dwellings would be allocated to households who were eligible for a housing allowance of more than Nlg 250 (Nlg 275 at present), as determined on the basis of the ratio between their income and the rent charged for the dwelling. The existence of expensive mismatch is often associated with the application of the housing allowance (IHS). The use of IHS increased during the period 1986-1990, rising from 21 to 25 percent of the total number of renters. However, most of the increase may be ascribed to households who moved from a low-rent dwelling to a rental unit with a moderate rent level. (That is, in 1990, the basic rent would have been between Nlg 490 and Nlg 650.) Households occupying dwellings with a moderate rent level are not classified under expensive mismatch. Only seven percent of the priority category receives an IHS grant of Nlg 250 or more per month. Of those households receiving that amount, 45 percent may be counted among the expensive mismatch group. Out of all households belonging to the priority category and having moved recently to a dwelling where the housing costs are above the norm, 26 percent receive an IHS grant of more than Nlg 250 (MVROM, 1992c). It is obvious that a policy aimed at diminishing the expensive mismatch will partially reduce the number of housing allowance recipients who are awarded more than Nlg 250 a month in IHS benefits. The profile of expensive mismatch households may be described as follows. The size of the household hardly differs from that of other renters. Relatively many of all those who fall into the category of expensive mismatch are at least 65 years of age. They are dependent on a social security (AOW) benefit and a pension for their income. This over-representation shows that mismatch is to some extent the result of inertia on the housing market. And it is also due to tile fact that the elderly, possible under force of circumstance, are allocated dwellings that suit them in terms of amenities but are too expensive, in view of their income. A fairly large number of expensive mismatch households live in dwellings built after 1970 (MVROM, 1992c). 165

5 Mismatch and national housing policy The national housing policy seeks to reduce the amount of households with housing costs below the norm. By the year 2000, the number should be 550,000, down from 726,000 in 1990. Similarly, the number of people whose housing costs are above the norm should be reduced to 120,000 by the year 2000, down from 168,000 in 1990 (MVROM, 1989, p. 148). The means to achieve these targets are mostly sought in the promotion of filtering (MVROM, 1989, p. 54). Several instruments, briefly described below, can be used to this end (see Dieleman, Hooljmaljers and Van Kempen, 1993).

Strategic new construction One important way to make inexpensive rental dwellings available to low-income households is to build more dwellings in the more expensive rental and owner-occupier sector. This will certainly increase the flow out of the inexpensive sector (Tweede Kamer, 1990, p. 4). Households whose income is too high in relation to the rent level can also be enticed to move out of their inexpensive dwelling in another way. This entails building special units for the elderly, more expensive rental dwellings, and owner-occupier dwellings. These sectors can form important elements of a strategic policy for new construction. The need for new construction in the market sector was recently emphasized in a trend report (MVROM, 1992e) and in a letter from the Secretary of State for Housing to the Second Chamber of Parliament, reiterating this position. In the big cities, relatively many households live in a situation of inexpensive mismatch. These cities are the best places to build more expensive dwellings, because the policy of strategic construction would be most likely to succeed there (Tweede Kamer, 1990, p. 5). Housing allocation instruments Filtering can also be promoted by applying strict allocation procedures. The Housing Act and the Housing Decree stipulate that municipalities and social landlords must give people with a low income priority in allocating inexpensive rental units. The parties who allocate dwellings (that is, the municipalities, housing associations, and private landlords) will have to try various ways to improve the allocation procedures. They must find ways to ensure that the inexpensive units are assigned to the low-income households. They can do this, for instance, via adoption and strict enforcement of financial eligibility criteria (in accordance with the white paper on "Housing in the Nineties"; see also Heerma, 1991, p. 5). Another way is by exercising close control on the income figures of households at the time a dwelling is allocated (Tweede Kamer, 1990, p. 7). Yet another way is through active mediation for filtering households. It is also important that municipalities make contractual agreements with landlords. These agreements also apply with regard to the decision on whether or not to apply the approval threshold. Of course, this point is particularly important in the prevention of expensive mismatch. As a potential sanction, just in case the housing allocation does not diminish the mismatch sufficiently, the Secretary of State for Housing, Physical Planning and Environment has the option to introduce temporary rent contracts and temporary housing permits. 166

Regional cooperation among municipalities and landlords The amelioration of mismatch and the promotion of filtering may be more successful when municipalities and (other) landlords have regional coordination of their policy for housing allocation and distribution. Other forms of cooperation, including joint registration of home-seekers and exchange of dwellings to be assigned, may lead to a more effective use of the regional housing stock in its totality (MVROM, 1991, p. 26). Information The public should be provided with information on the housing supply, in both the social and the market sectors. This would enable home-seekers to find suitable housing on their own initiative and by using their own resources. The national government underwrites these measures as opportunities to reduce the mismatch on the housing market. But the policy has to be elaborated at the local level, and it is up to the municipalities and the housing associations to do so. To gain insight into this process, we conducted an investigation among administrators of housing associations and municipal project managers in the field of housing. The findings give an impression of the way in which these parties put national policy into practice at the local level (Dieleman, Hooijmaijers and Van Kempen, 1993). The next section is devoted to this topic.

6 Housing policy on the local level At the national level, mismatch is considered to be one of the main problems of the housing market. But at the policy level of the municipalities and housing associations, that is certainly not the major issue. The majority of the interviewees consider the qualitative housing shortage to be the biggest problem. Because of the shortage of dwellings for high-income households, these people remain living in inexpensive accommodation. The shortage for low-income households is partially caused by the inexpensive mismatch. The dismantling of the program for property subsidies has decreased the options for immediate construction of new dwellings for low-income households. And this is the reason why filtering is seen as the answer to the housing problems of low-income people. Municipalities and housing associations promote filtering in various ways. They make an effort to stimulate higher-income households to vacate their low-rent dwelling. In this manner, they expect to increase the opportunities for those with a low income. In many cases, strategic new construction is perceived as the solution. Often, it is the only way to enlarge the range of opportunities for people with a low income. This approach includes building more expensive rental dwellings, owner-occupier dwellings, and special units for the elderly. The intent of these efforts is to entice households with a relatively high income to move out of the inexpensive rental units. In turn, strategic new construction has a side effect: the amelioration of mismatch. That which is deemed to be strategic may differ from one municipality, or even housing association, to the next. Some examples are presented below. 167

In the municipality of The Hague, a shortage of more expensive dwellings was noted. Because of the shortage, households with a relatively high income live in inexpensive rental dwellings. When they do move, they tend to go to areas outside of The Hague. In order to reduce this lack of (attractive) more expensive dwellings, a number of projects are being undertaken in The Hague. The initiatives include construction of single-family dwellings priced at Nlg 140,000; apartments in the price range of Nlg 120,000 to 190,000; and dwellings built without subsidy, costing upwards of Nlg 190,000. In view of this supply, the opportunities for higher-income people are increased. And when these households have been occupying inexpensive rental units within the city, those dwellings will thus become available. Potential homeowners who vacate a rental dwelling in The Hague, of which the basic rent (excluding services) was Nlg 550 or less, are given priority to buy a home in those projects. In this way, one stone can bring down several birds: the supply is enlarged for both higher-income and low-income groups; the higher-income groups do not have to move to the surrounding municipalities; and at the same time, this helps prevent The Hague from developing a population profile that is increasingly homogeneous. The housing stock owned by a housing association in Rotterdam consists primarily of relatively inexpensive rental dwellings. In fact, this is true of so many housing associations. Therefore, it is impossible for a household living in one of these units to move on to a more expensive one under the same management. The housing association would like to expand the opportunities for housing careers among the membership. For this reason, it intends to build about 500 fairly expensive rental dwellings (as well as a small number of inexpensive owner-occupier dwellings) in a new development. These new units are explicitly targeted to those higher-income households that are currently occupying a relatively inexpensive rental dwelling. This effort was not designed primarily as a way to reduce the (inexpensive) mismatch. But by offering this opportunity to households with a higher income, dwellings in the inexpensive rental stock do indeed become available. Obviously, the options to steer housing behavior are largely determined by how tight the market is. When a carefully selected scarce product is added to the housing stock, it is generally quite feasible to set conditions on who will gain access to the units, as well as on the type of dwellings they leave behind. In the big cities of the Randstad, the housing market is extremely tight. There, a policy of this type may be successful. But housing markets with an oversupply of dwellings are not amenable to such policy. There, the authorities have no chance of steering housing behavior in the direction of filtering. Clearly, when the supply is ample, the social landlords are grateful for every new potential tenant. New housing projects are not always developed for households with higher incomes. The elderly also constitute an important, target group. That is not surprising; Dutch society is graying. Many elderly persons live in a dwelling that is actually more suitable for another type of household, such as low-income families. Measures to promote filtering that are specifically aimed at the elderly can have the desired effect; 168

elderly persons would leave their inexpensive rental dwellings, making the units available to low-income households (see Filius, Hooimeijer and Dieleman, 1990). In order to get the filtering process going and to ensure its success, it is crucial to build the right types of housing in the right place (which is often in the same district or neighborhood). Many municipalities build projects specifically for the elderly. For the time being, this effort is confined to small numbers of dwellings for the elderly. These units are then exclusively available to senior citizens. In most cases, additional conditions are made with regard to the dwelling that is being vacated. For instance, in Apeldoorn and Maastricht, the projects involve the construction of units for the elderly that are primarily intended for older persons living in relatively inexpensive single-family homes. In principle, the construction of dwellings for the elderly may be a successful aspect of a policy to reduce mismatch, even if that is not its prime goal. Yet in view of the small numbers of units that have been built thus far, the quantitative effect is still negligible.

Expensive mismatch: an unresolvable problem? In municipalities where expensive mismatch is widespread, efforts to combat the problem are often considered futile. In 's-Hertogenbosch, the existence of expensive mismatch is the result of the small amount of inexpensive rental dwellings in this city. Because the housing stock is expensive, households with a low income have a very hard time finding an inexpensive dwelling. The introduction of the approval threshold (described above) forced people to wait even longer for an inexpensive dwelling. In a short time, the waiting time of two years went up to four or five years. The construction of new inexpensive dwellings is out of the question. Alkmaar is also confronted with the problem of many expensive and few inexpensive rental dwellings. There is a high incidence of expensive mismatch (14 percent of all rental dwellings). Alkmaar was previously a growth center. Accordingly, many relatively expensive dwellings were built there in the 1980s. Thus, rents are on average higher there than elsewhere in the Netherlands. The pressure on the inexpensive rental stock is increasing, due to the approval threshold. In view of the relatively small number of inexpensive dwellings, people with a low and/or declining income are usually forced to remain in an expensive rental unit.

7 Evaluation The previous sections suggest that the municipalities and the housing associations are not very serious about combatting the problem of mismatch. At that administrative level, the main task is mostly perceived as diversifying the stock. In this effort, the housing preferences of various types of household are taken into account more than ever before. A side effect of diversifying the local housing stock is that eventually there will be greater opportunity to combat the problem of mismatch. We too feel that the existence of inexpensive mismatch does not pose a major 169

problem. In our opinion, efforts to tackle it need not be given high priority. The social rental sector in the Netherlands is very large. Therefore, besides low-income households, other types of household --including those with modal income or above-seek access to this sector. Some examples of other groups are starters on the housing market; young dual-earner households who are not ready to buy their own home; and the elderly. In fact, inexpensive mismatch cannot be avoided when the social rental sector is as large as it is in the Netherlands. At the same time, inexpensive mismatch has a major advantage. It prevents segregation by income in the social rental sector. Therefore, the social rental sector in the Netherlands does not have the stigma of a housing market segment exclusively for the poorest of the poor. Lacking this stigma, the social rental sector is not shunned by other households. In many other countries, the rental sector does have this negative stigma. Of course, inexpensive mismatch should not hinder the provision of housing to households with a low income. But the bottleneck is not the existence of inexpensive mismatch. Rather, the provision of adequate housing in the big cities is hampered by the overall shortage of dwellings in the Randstad (Dieleman, Hooijmaijers and Van Kempen, 1993). We do see expensive mismatch on a large scale as a problem. This phenomenon puts political pressure on the instrument of housing allowance. It is hard to understand --and even harder to explain-- why 953,000 households need (in 1991) a housing allowance, when the Netherlands has such a huge and accessible social rental sector. This is where previous policy comes home to roost. In the Netherlands, local administrators carry out the housing policy, and the national government has to pay the bill. Local managers of the rental sector allocate the dwellings. In doing so, they are unconcerned with the claims for housing allowance that their allocation policy generates. Each year, the national government pays out nearly two billion guilders in housing allowances to renters; that is, they pay as long as the political support is forthcoming. It would be useful to temper the use of the IHS program before the political support for this important instrument is undermined. To this end, the municipalities and housing associations would have to attempt --more than at present-to minimize the claims on IHS by taking this into account at the time of allocation of a dwelling.

Notes 1 As of 1986, this referred to singles with a net disposable annual income over Nlg 22,000. It also referred to multi-person households with a net disposable annual income over Nlg 30,000. Households with an income under these thresholds are considered priority groups of housing policy. The 1990 incomes are indexed, whereby the thresholds are set at Nlg 22,300 for singles and Nlg 30,400 for multiperson households. 2 By inexpensive mismatch, we mean a household with a relatively high income that lives in an inexpensive rental dwelling. See note 1 for the income thresholds we apply. In 1986, an inexpensive rental dwelling was defined as a unit with a basic 170

rent (not including services) of Nlg 450 or less. In 1990, the threshold was Nlg 490 per month. The figures presented in this section are estimates. They are based on the Housing Demand Survey, a sample survey, which is regularly conducted by the Netherlands Central Bureau of Statistics (see MVROM, 1992b). Expensive mismatch occurs when households with a low income occupy expensive rental dwellings. See note 1 for the income thresholds applied. In 1986, a dwelling was considered expensive if the basic rent (excluding services) was Nlg 600 or more; in 1990, the threshold was Nlg 650.

References

Andeweg, R.B. (1988), "Less than nothing? Hidden privatisation of the pseudo private sector: the Dutch case", West European Politics (11), no. 4, pp. 117-128. Boelhouwer, P. and H. Priemus (1990), "Dutch housing policy realigned", Netherlands Journal of Housing and Environmental Research (5), no. 1, pp. 105-119. Dieleman, F.M. (1994), Social rented housing: valuable asset or unsustainable burden? Urban Studies (31), no. 3, pp. 447-463. Dieleman, F.M., A.B.C.M. Hooijmaijers and R. van Kempen (1993), Veldtocht tegen de scheefheid: een onderzoek naar de effecten van scheefheidsbestrijding op lokaal niveau, Utrecht: Faculteit Ruimtelijke Wetenschappen, Universiteit Utrecht. Dieleman, F.M. and R.B. Jobse (1991), "Multi-family housing in the social rented sector and the changing Dutch housing market", Housing Studies (6), pp. 193-205. Filius, F.C., P. Hooimeijer & F.M. Dieleman (1990), "Meer woningverlating dan nieuwbouw in 2000", Bouw (45), pp. 34-35. Harloe, M. and M. Martens (1985), "The restructuring of housing provision in Britain and the Netherlands", Environment and Planning A (17), pp. 1063-1087. Heerma, E. (1991), Brief aan de Colleges van Wethouders en de Colleges van Gedeputeerde Staten over de evenwichtige verdeling van de woningvoorraad, Zoetermeer: DGVH. Lundqvist, L.J. (1992), Dislodging the welfare state? Housing and privatization in four European nations, Delft: Delft University Press. Maas, R.A.W.M. (1992), "Te goedkoop en te duur wonenden niet zondermeer uitwisselbaar", Woningraad (19), pp. 52-55. Meusen, H. and R. van Kempen (1994), Dutch social rented housing: a British experience? Bristol: School for Advanced Urban Studies. MVROM Ministerie van Volkshuisvesting, Ruimtelijke Ordening en Milieubeheer (1989), Nota Volkhuisvesting in de Jaren Negentig; van bouwen naar wonen, 's-Gravenhage: SDU uitgeverij. MVROM Ministerie van Volkshuisvesting, Ruimtelijke Ordening en Milieubeheer (1991), Beleidswijzer: Evenwichtige ~;erdeling van de woonruimte, 's-Gravenhage: DGVH. MVROM Ministerie van Volkshuisvesting, Ruimtelijke Ordening en Milieubeheer (1992a), Volkshuisvesting in cijfers 1992, 's-Gravenhage: MVROM. 171

MVROM Ministerie van Volkshuisvesting, Ruimtelijke Ordening en Milieubeheer (1992b), Profiel van het te goedkoop wonen, 's-Gravenhage: MVROM. MVROM Ministede van Volkshuisvesting, Ruimtelijke Ordening en Milieubeheer (1992c), Regionale profielen van de woonruimteverdeling I/I, 's-Gravenhage: MVROM. MVROM Ministerie van Volkshuisvesting, Ruimtelijke Ordening en Milieubeheer (1992d), Achtergronden bij dure scheefheid, 's-Gravenhage: MVROM. MVROM Ministede van Volkshuisvesting, Ruimtelijke Ordening en Milieubeheer (1992e), Trendrapport volkshuisvesting 1992, 's-Gravenhage: MVROM. Priemus, H. (1990), "Changes in the social rented sector in the Netherlands and the role of housing policy", paper presented to the International Research Conference on Housing Debates and Urban Challenges, Pads, 3-6 July. Rogmans, B. (1991), "De afbouw van de wederopbouw", Intermediair (27), no. 25, pp. 18-23. Silver, H. (1990), "Privatising public housing in Britain and the Netherlands", paper presented to the Seventh International Conference of Europeanists, Washington D.C., 24 March. Tweede Kamer der Staten Generaal (1990), Volkshuisvesting in de Jaren Negentig, 's-Gravenhage: SDU uitgevedj. Van der Schaar, J. (1987), Volkshuisvesting: een zaak van beleid, Utrecht: Aula. Van Kempen, R. and J. van Weesep (1994), "Housing policy, gentrification and the urban poor: the case of Utrecht, the Netherlands", Urban Studies (31), no. 7, pp. 1043-1056. Van Kersbergen, K. and U. Becker (1988), "The Netherlands: a passive social democratic welfare state in a christian democratic ruled society", Journal of Social Policy (17), pp 477-499. Van Weesep, J. and R. van Kempen (1993), "Low income and housing in the Dutch welfare state", in Hallet, G. (ed.), The new housing shortage; housing affordability in Europe and the USA, London: Routledge, pp. 179-206.

172