Social rights and spatial access to local social services: The role of structural conditions in access to local social services in Estonia
Kersti Kriisk, Tallinn University, School of Governance, Law and Society
Renate Minas, Stockholm University, Department of Social Work
Provision of local social services is an important measure to alleviate poverty, support vulnerable groups and promote citizens' independent living. Access to such services is inevitably the outcome of a complex interplay between several actors with different expectations, obligations and rights, and can be impeded by various obstacles. These include shortage of resources and legislative and implementative fragmentation (Daly, 2002; Rauch, 2005). While legislative fragmentation is about dispersion of authority in legislative processes resulting in ambivalent or unclear regulations, the implementative fragmentation entails distribution of power and resources across levels of government, as well as fragmentation of authority in the implementation process. Both types of fragmentation can be decisive for access to social services; legislation is a crucial precondition for provision, yet, it does not guarantee that a service actually is provided at local level. Implementative fragmentation refers to the authority of local social services and social workers create organisational solutions that can promote or limit access (Minas, 2005). Local conditions in terms of population density, size of municipality or distance to county centre, as well as financial resources have also been discussed as playing a decisive role in access to local social services. This issue of spatial access is important all over Europe and elsewhere, especially in small countries such as Sweden, Finland and Norway, or in sparsely populated areas, e.g. East Germany (Gustafsson, 1980; Larsen, 2002). In some countries, the discussion about the principle of proximity and capacity has emerged with renewed strength and reforms in form of mergers of municipalities have been introduced in Finland, Denmark and Norway (CEMR, 2008; Nordregio, 2015). Greater proximity to social service providers is thought to increase the likelihood that individuals in need will receive or apply for support, as shorter distances reduce the burden of commuting (Allard, 2004), whereas too small municipalities do not always live up to demands of tasks performance. The urgency of this issue is stressed by a projected overall population decline in rural areas by 2050, especially in the new EU Member States that in many countries will be accompanied by a decline in social services and in infrastructure (European Commission, 2008). That would adversely affect the situation of the most vulnerable. Consequently, it will be more difficult and costly to deliver social services to poor people in areas characterised by dispersed population and problems of remoteness (OECD, 2010). Thus, ensuring that citizens in general and vulnerable groups in particular have access to services irrespective of where they live, is of crucial importance for modern welfare states and for their ambition of upholding social rights. So far, however, existing studies on access to local social services have often concentrated on one target group (e.g. elderly, children) or on one type of services (e.g. care services), and do not provide a comprehensive picture of access to locally offered social services that municipalities are obliged or expected to provide. Furthermore, most studies concerned with access begin with the provision of existing services, ignoring the question of which services are provided at all (and which are not). This study will fill that gap by using Estonia as a case. Estonia is one of the smallest East European countries, but it is divided into a large number of municipalities, often with very low population density. The extremely high level of decentralisation of public and social service provision to the local level of government that enjoys an almost unlimited discretion in its decisions, sets an example for the importance of local structural features that may play a decisive role in access to local social services (van Mechelen & De Maesschalck, 2009; OECD, 2011).
The focus of this study is to analyse access to local social services and the impact that the structural conditions of Estonian municipalities have on it. More precisely, we will analyse the impact of local conditions, using indicators like remoteness, resources and spending, on the extent of (in terms of quantity) and the additional demand for local social services. We include all social services in Estonia, which the municipalities are expected to finance and administer locally, and where local authorities have full discretion to decide to whom, how and by whom these services are delivered. Empirical data on the extent of and additional demand for social services comes from a survey among local social workers, whereas data on structural conditions derives from Statistics Estonia and Ministry of Finance. Analytically, this article follows the discussion on social rights (Marshall, 1950).
The article consists of three parts. The first part deals with theoretical and conceptual considerations about social rights, as well as with access to social services and local conditions influencing that. The second part gives an overview about the Estonian welfare state’s institutional frame with emphasis on the social service provision. The third part includes empirical study about local social service provision in Estonia based on an online survey and administrative data. The paper ends with a discussion.
2 Access to local social services – universal social rights
As European societies developed, the introduction of social rights came to be associated with social citizenship. The provision of a minimum set of social rights ought to guarantee a minimum of welfare, according to the standards prevailing in society, to all citizens regardless of their position on the labour market or in society (Marshall, 1950). Social rights are often realised through social policy such as various social insurance programs and social services (Daly, 2002). Yet, unlike civil or political rights, social rights are based on limited resources and are not distributed freely among citizens. An important distinction between social insurance and social services is, however, the distribution principle behind them. Whereas social insurance compensations are contribution-based and their distribution follows the same rules all over the country, the access to social services is often based on an individual needs test proving a person’s eligibility. At the same time, it is expected that all citizens are treated equally, including equal access and eligibility standards irrespective of the citizens’ residence. In other words, access to social services should not depend on which part of the country a citizen lives (Elster, 1992). The provision of local social services can be contrasted to nationally provided services. While the latter aim to deliver uniform standards throughout the nation and deviations from those standards therefore produce illegitimate inequalities, the municipalities provide local services by responding to the demands of the local population within local context. This implies a trade-off between local conditions and local capacity to fulfil assignments given by the central government (Powell & Boyne, 2001). Kröger (1997) denotes this as the ‘dilemma of municipalities’. Regarding access to local social services, the question of social rights is thus of particular importance.
In general, knowledge about the spatial access to welfare and local strategies of equality and equity is fragmented between various research traditions, for example geography (e.g. Mohan, 2003), research on the commons (Young, 1990; Gyuris, 2014), research on poverty and social exclusion (e.g. Atkinson, 2000) or research on multi-level governance (Kazepov, 2010). These research traditions all cover important aspects of access to social welfare on various spatial levels, yet some crucial dimensions are still under-researched. Research on local social service provision is so far mostly concentrated on specific target groups, such as children (e.g. Belanger & Stone, 2008), or a type of services (e.g. Trydegard & Thorslund, 2001; Minas, 2005), whereas the package of locally provided services is ignored. Often it is crucial for a person to receive several services at the same time such as counselling, social transport, housing, etc. (Ferrera, 2005). Also missing are studies that start a step earlier and investigate local social services that municipalities are obliged or expected to provide, instead of looking at existing services. Therefore, an important part of the local welfare system has been understudied. We will fill that gap by including all social services in Estonia which the municipalities are expected to finance and administer locally, and direct attention at the impact of local conditions (remoteness, resources and spending) on the extent of (in terms of quantity) and the additional demand for local social services. Andreotti et al. (2012) talk about local welfare systems as representing dynamic arrangements in which the specific local socio-economic and cultural conditions prompt different mixes of welfare provision. In the following, access to social services will be conceptualised and operationalised.
3 Access to social services – extent of and additional demand for local social services
At first sight, the simplest way to measure access to social services would be to count the number of social services in each of the Estonian municipalities. This, however, would only partly cover the range of access to social services. The mere number of available social services does not necessarily reflect the demand and/or need for services at municipality level. Therefore, in this study, the access to social services is defined by two components: the extent of provided social services and their additional demand. Extent of services can imply two things: firstly, a local service is indeed provided and delivered by a municipal agency, and secondly, the municipality just provides the financial and administrative framework but the actual delivery occurs by another, e.g. private actor. The number of provided social services reflects the range of existing services and may indicate something about the prioritisation within the municipality. However, a smaller number of provided services should not per se be interpreted as a sign that people in need of specific services do not receive help. In order to reflect that, we use the concept of additional demand for social service provision.
Demand for social services as such is a complex and multidimensional concept and difficult to operationalise. In general, eligibility for local social services is based on place of residence and specific individual needs. However, these needs are not always related to certain demographic criteria or specific target groups.Following Young (1990), this category also allows to capture the differences in identity or activity in urban and rural areas, meaning that people cluster according to a variety of affinities or needs depending upon the engagement of people, which does not necessarily follow demographic criteria or defined target groups.In addition, municipalities have the discretion to react to individual needs with alternative measures: they can for example meet the need with a cash benefit or other services. Particularly for small municipalities it can be much more cost-efficient to meet needs individually instead of introducing a standard service (Riigikontroll, 2012; Ainsaar & Soo, 2009). Therefore, the concept of additional demand used in this study implies that a service is not provided but needed, or that a service is provided but not to a sufficient extent. Extent of and additional demand for social services in Estonian municipalities are the dependent variables in our study, measuring access to local social services. We will use that definition to study access to all major social services that the municipalities are expected to provide. Variation in access to these services between the municipalities, and variation between local social services will be explained by the following local factors: remoteness, resources and spending on local social protection.
4 The local context
Previous studies about local social services in other countries point out a large variety of variables explaining local disparities (Trydegård & Thorslund, 2001; Lien & Pettersen, 2004; Jensen & Lolle, 2013). In this study, we will use the concepts of remoteness, resources and spending to reflect socio-economic differences in local conditions.
Remoteness is an overarching concept including dimensions such as size of population, low population density, dispersion of villages and distance to county centre (Allard & Cigna, 2008; OECD 2010; Australian Institute of Health and Welfare, 2004). The concept allows us to broaden the discussion beyond focussing on population size (which is dominant in political discussion). Size of population is, however, crucial: a large number of inhabitants implies resources in the form of tax income, but it also means additional expenditure because of a need for more and diversified public and/or social services. Since local social services are sometimes provided only in county centres or other larger municipalities, distance to service provision is often pointed out as a risk factor for access to social services (Powell & Boyne, 2001; Pinch, 1997; Daly, 2002), as is distance to municipality’s central village/town. The geographical distance from the main centre of economic, political and cultural activity (OECD, 2010) is not only connected to municipality’s low population density, but also to other aspects of remoteness. Distance in time (e.g. traffic situation in larger cities can result in longer driving measured in time and not in kilometres) as well as administrative regulations regarding space are also important dimensions but will not be considered in this paper. An additional territorial factor often mentioned is the type of municipality, i.e. the distinction between urban and rural municipalities. Previous studies in Estonia (Ainsaar & Soo, 2009; Medar, 2004), as well as in UK and US (Pugh, 2003; Turbett, 2009; Belanger & Stone, 2008) have pointed out that social service provision differs considerably between urban and rural areas for various reasons.
4.2 Resources and spending
Resources in the form of revenues as well as infrastructure are a prerequisite for service provision. Yet, evidence on the impact of resources on local social service provision is scattered. Larger municipalities (by population size) are often assumed to have a better capacity in providing a wider range of public services, to raise the quality of the provided services and to deliver services more cheaply (Swianiewicz, 2010; Daly, 2002). This has been shown by Ainsaar & Soo (2009) in the case of local services for children and for families with children in Estonian municipalities. Reiljan et al. (2012), on the other hand, showed that municipalities with higher earning-related income have a lower share of social protection spending.
5 The institutional framework of the Estonian welfare state
Estonia is one of the smallest Eastern European countries both in terms of geographical size (45,227 km²) and population (about 1.3 million in 2013) (Statistics Estonia, 2016). At the same time, Estonia is divided into a relatively large number of municipalities (226). The median population size of Estonian municipalities in 2013 was about 1,600 inhabitants, having decreased from about 1,800 in 2006. In 2013, the smallest municipality had less than 60 inhabitants while the largest municipality – the capital city of Tallinn – had over 400,000 inhabitants. Hence, the number of inhabitants in Estonian municipalities differs by more than 6,000 times. Estonian municipalities consist of 193 rural and 33 urban municipalities (towns). This distinction is based on an administrative categorisation regulated within nationwide legislation, where it is defined that urban municipalities only consist of one settlement (e.g. a town), whereas rural municipalities consist of two or more settlements – mainly villages. Rural municipalities are often large sparsely populated areas. Several authors (Sootla et al., 2009; Swianiewicz, 2010) have pointed out that compared to other European countries Estonian municipalities have one of the smallest average population size and the lowest density, while the range of functions is one of the widest. Thus, the issue of access to and provision of social services is essential for the local population. The division of responsibilities between the central and the local government regarding the provision of social services is regulated by the Social Welfare Act and the Local Government Organisation Act. However, national regulations only provide an ambivalent and unclear description of the municipalities’ tasks. The Social Welfare Act defines in a very general way the range of social services that the municipalities are expected to provide. Recommended guidelines published by the Ministry of Social Affairs in 2010 define the minimum standards for local social services, but do not reduce the general ambiguity and uncertainty. A further characteristic of the Estonian local welfare scheme is the far-reaching local decision-making power with respect to the administration of local social services. This is a very important aspect in access to social services. Some of the services that the municipalities are expected to provide are mentioned in the legislation without a definition of the target group, and the municipalities can decide upon that by themselves. Fiscal autonomy is limited and local financial resources are mainly depending on the decisions by central government (Trasberg, 2009). A mismatch between local fiscal capacities and devolved responsibilities resulting in variation and fragmentation of local social service provision has also been pointed out (e.g. OECD, 2011; Riigikontroll, 2012; 2015).
Financing and administration of social services in general falls within the responsibility of local municipalities (whereas the actual delivery of a service can be organised in various ways). The municipalities finance and administrate for example services related to counselling, social housing, personal assistance, social transport, care and other services aimed at supporting independent living, improving the quality of life and promoting social integration. Child day-care centre services are, in contrast to many other countries, defined as a municipal responsibility categorised under educational services and not considered as a social (care) service. Main eligibility criteria for receiving local social services are the person’s place of residence and the existence of specific individual needs and/or belonging to certain target groups defined in local legislation. Overall, local social services are part of a complex structure of vague central government steering, local autonomy and financial scarcity.
As in other countries, the problem of capability deficit of small municipalities in Estonia is high on the political agenda (Sootla et al., 2009; Swianiewicz, 2010); the tasks that the local government should to carry out are under-resourced, and local governments are often too small to be able to fulfil their responsibilities and functions. Hence, the issue of reorganisation of the local level by merging municipalities is an important political issue. So far, the political discussion about reorganising the municipal level in Estonia is mostly concerned with merging municipalities in order to increase the population size in the municipalities. Yet, as Sootla et al. (2009) have pointed out that the amalgamation of municipalities might result in increasing the territorial area of a municipality even more, raising the costs of communication and in the long term even jeopardising accessibility to local services. Thus, the merging of municipalities is only one strategy among others, and it might be just as important to reform the municipalities’ financing structure and mandatory duties (European Commission, 2015; OECD ,2011; Reiljan & Ülper, 2011; Riigikontroll, 2012; Sootla et al., 2009).
6 Data and methods
The data used for analysing access to local social services in Estonia (extent and additional demand) is based on an online survey conducted in 2013. The survey gathered information about local social service provision in Estonian municipalities and includes data from 225 of 226 municipalities. The survey consisted of questions concerning the provision of the 18 main local social services that municipalities are expected to offer; the social workers were also asked to assess if additional provision is necessary. This survey compensates for a lack of systematic data on Estonian social service provision. So far, empirical data on local social services is fragmented. The Ministry of Social Affairs regularly collects data on six local social services from the municipalities (home care services, support person for adults and children, personal assistant, child care, social housing), while data about social centres, elderly care homes or shelter services is only collected from non-municipal service providers, and for some services (e.g. social counselling, social transport, etc.) no systematic data exists at all.
In the survey, local social workers were asked to describe the extent of and any additional demand for each of the 18 local services during the period of 2010–2013. The survey was directed at local social workers who were assumed to know the situation in the respective municipality and to be able to estimate the met and unmet needs for social services in the local area. It is difficult to estimate the validity of the answers of social workers. A rough comparison with data from the Ministry of Social Affairs showed that the extent of services provided was about 10% higher in the survey. However, the survey included service provision over three years in total, whereas official data is collected for single years. In addition, the official data collection does not include alternative ways of service provision. Furthermore, previous studies (Groenningsaeter & Kiik, 2009; Mitendorf & van Ewijk, 2015) have pointed out to local social workers’ dissatisfaction with service provision mainly due to lack of resources. Therefore, we assume that the local workers have not much underestimated nor overestimated the extent of and the demand for social services. Overall, the survey provides a unique opportunity for a description and an analysis of local social service provision in Estonia.
The independent variables are taken from administrative data from Statistics Estonia (distance to county centre and capital city), Ministry of Finance (resources and spending) and the Population and Housing Census conducted in 2011 (population size and density). In accordance with the commonly used definition of remoteness, we use the following indicators in our analysis: population density, distance to county centre and distance to the capital city (Tallinn). Resources are measured by income tax per capita in the local budget and by general spending per capita. We are further interested in social spending priorities as an indicator of local politicians’ willingness to invest in local social needs. This is reflected by taking into account the proportion of social protection spending in local budget and social protection spending per capita. However, high spending on social protection is an indicator that should be used with caution since high spending is not necessarily equivalent to efficient service provision (Green-Pedersen, 2007). Furthermore, besides spending on social services, it includes spending on social cash benefits, investments in local social policy (large-scale funding from European Union and/or state budget), administrative spending on local social policy, etc.
The resources and spending variables are calculated as average values for the years 2006–2013 to avoid a bias due to occasional funding in single years (single large-scale investments from central government and/or from European Union budget). Municipalities are divided into urban and rural categories.
The analysis has been conducted in two stages. In the first stage, the dependent variables (extent of and additional demand for local social services) are studied by combining all 18 local social services. This stage of analysis presents descriptive statistics (table 1), bivariate correlation analysis distinguishing between rural and urban municipalities (table 2) and multivariate regression analysis (table 3). In the second stage, the dependent variables are studied separately for each of the 18 local social services. In this stage of the study, we analyse differences in social service provision between urban and rural areas for each local social service separately (table 4). Mean comparisons (t-tests) are provided for subgroups of municipalities where the service is provided and where it is not, as well as for municipalities where additional demand is assessed and where it is not (tables 5 and 6).
We start with a general description of the municipal characteristics and local social services. Table 1 shows that Estonian municipalities differ in average only slightly according to the extent of and the additional demand for social services, but more as regards remoteness, resources and spending. On average municipalities provide 11 social services – 13 in urban and 11 in rural municipalities, respectively. In the opinion of the social workers, there is an additional demand for 7 social services, on average; this is lower in rural (7) than in urban areas (9). Thus, social workers in rural municipalities tend to be somewhat more satisfied with the social service provision compared to social workers in urban municipalities. However, the differences between rural and urban municipalities are rather small when looking at all services combined. As expected, population size, as well as distance to county centre and to capital city (remoteness) also differ considerably across the country, especially with respect to population density. The lowest population density is only 1.6 persons per square km; the highest is 2515.5 inhabitants per square km.
Table 1. Descriptive statistics
Table 1 illustrates further that urban municipalities have higher income tax per capita than rural municipalities. Based on average values of all municipal resources and spending, it seems that rural municipalities have on average more financial resources to provide local public services. However, if we compare the median values of all three spending indicators between urban and rural municipalities, we can observe that more than half of the rural municipalities spend less both in terms of general as well as social spending than half of the urban municipalities. This may indicate that single large-scale strategic investments from the government and/or EU budget have a larger influence on rural municipalities’ budget, even when we observe the average values of eight years.
In the next step, we will analyse the relationship between municipal characteristics and local social services (table 2).
Table 2. Spearman correlations between access to local social services and local conditions
A strong positive correlation can be found between the extent of services and the size of the population; thus, the more inhabitants there are, the more social services exist for them. In addition, population density correlates positively with the number of provided services, although less strongly. Distance to the capital city and to the county centre have a negative effect on service provision; less so for distance to the capital city. This might be explained by the fact that there are several so-called metropolitan municipalities in Estonia; for example, besides Tallinn in the north of Estonia, the second largest town is located in the southern part of Estonia. The correlation with distance to county centre is considerably stronger in urban areas than in rural areas. Municipalities with higher income tax per capita provide more services, especially in urban areas. However, municipalities with higher spending per capita provide fewer services. Neither of the social protection spending indicators are correlated with the extent of social services.
The additional demand for services and the chosen municipal characteristics are only weakly correlated; they show, however, the same direction as for the extent of social services. The weak but statistically significant positive correlation between the extent of provided social services and the number of services that need additional provision indicates that the more services there are provided, the bigger need for additional provision is expressed. Distance to county capital, income tax in budget per capita and social protection spending have no statistically significant correlation with additional demand for services. For urban municipalities, no statistically significant correlation at all can be noted.
Table 3. Multivariate regression models for the extent of local social services. Standardised regression (beta) coefficients and explained variance
In order to exclude the possible multicollinearity issue between local contextual factors, multivariate models on the extent of the provided services were calculated (table 3). The multivariate models emphasise the relevance of the distance to county centre in social service provision over other local context factors chosen, especially in urban areas. More services are provided in municipalities with higher population size or population density, but the increasing spending per capita reduces the number of provided services. For social services with additional demand, it was not possible to calculate any statistically significant and/or meaningful multivariate model that would have explained more than 8% of variations.
So far, looking at all services jointly, the analysis brings out a clear picture regarding access to local social services, and supports our assumption that local conditions play an important role. Particularly the remoteness indicators and resources are decisive indicators for access to social services. The supply of services is somewhat larger in urban than rural areas, but so is the demand for additional support. Nevertheless, how does the picture look like when we unpack the cluster of social services? In table 4, access to social services (extent of services and demand for additional support) is examined separately for each of the 18 services.
Table 4. Local social services' extent and additional demand in urban and rural municipalities
The table reveals that the extent of and the additional demand for all 18 local social services varies considerably. Three groups of services can be distinguished; one group of services that are provided to a high extent and two groups of services that are provided to a low extent. Social services most commonly provided in all municipalities are social counselling (in 98% of municipalities), general elderly care home services (in 95% of municipalities), social transport (in 91% of municipalities) and home care services (in 90% of municipalities). These services are provided to a large extent in both urban and rural areas, except for social transport which is considerably more relevant in rural areas. Additional demand for these services is generally low.
A second interesting group of local social services consists of those with more limited access, which are therefore provided in relatively few municipalities. These groups are furthermore divided into two subgroups. The first subgroup contains services such as personal support (for children and adults) and personal assistance services. These services are provided in around 50% of all municipalities, but at the same time, a majority of the municipalities claim that the demand for these services is higher than the municipalities are offering. In urban municipalities, the extent of and additional demand for these services are proportionally higher than in rural areas. This indicates problems with access to personal support (for children and adults) and to personal assistance services. The second subgroup of local social services (shelter services and childcare) is also provided to a low extent; however, only around 20% of the local social workers claim that the provision is not sufficient. Interestingly, there is a large difference between urban and rural areas. It seems that the need for shelter services is more pronounced in the former.
How does then access to the 18 local social services look like when it comes to the local context in terms of remoteness, resources and spending? This is presented in table 5 and table 6.
Table 5. Differences in municipalities’ remoteness according to the extent of and the additional demand for local social services
The extent of and the demand for additional social services is analysed through mean differences for all included remoteness indicators. The general picture that emerges shows that distance to county centre but also to capital city and population density are of particular relevance for the extent of services, whereas population size is of minor relevance. This picture differs from the results of the previous analysis of the joint services that indicated a very high impact of population size on the extent of local social services (table 2). The effect of distance to county centre and to capital city on extent is negative, meaning that the farther from the county centre a municipality is located the lower is the probability of service provision. Proximity to county centre is relevant for the extent of all the local social services’ that are provided in less than 50% of the municipalities. For the frequently accessible services, the t-tests are generally not statistically significant; that might be a consequence of the relative smallness of the municipal subgroups where those services do not exist.
The remoteness factors differ less in the additional demand for social services. Table 4 highlights that for personal support services and childcare services the additional demand is lower in remote areas.
Table 6. Differences in resources and spending according to the extent of and the additional demand for local social services
The importance of resources and financial priorities for the extent of and the additional demand for all individual local social services is presented in table 6. The previous analysis of all services together indicated that resources explained some of the variations of local social service provision, whereas social protection spending patterns did not have any impact at all. Analysing all local social services separately confirms that finding. Social protection spending is of minor or no importance for the provision of local social services. The resource dimension has larger impact. An interesting point is that services provided to a low extent, e.g. personal support services and personal assistance services (although with high additional demand) as well as shelter services for children and childcare (low additional demand) are provided more often in municipalities with higher income tax per capita. Other shelter services (all belonging to the group of services that are provided to a lesser extent and with low additional demand) are in turn less likely to be provided in municipalities with higher spending per capita.
In this part, we wanted to analyse the impact of local conditions on access to every local social service separately. The analysis emphasised pointed out the importance of local conditions for access to local social services, particularly the remoteness indicators and resources. Distance to county centre and to capital city as well as population density are more relevant indicators of accessibility of local social services than mere population size. At the same time, we found that personal support services and personal assistance services are more likely provided in municipalities with higher income tax per capita. Spending indicators had in contrast of minor essentiality. The analysis also showed that access to some services (social counselling, general elderly care home services, social transport and home care services) is widespread both in urban and rural areas, whereas access to other services is more limited. Services that are provided less likely are personal support for adults and children and personal assistance services. Yet, in a majority of the municipalities, social workers claim that demand for these services is higher than the municipalities provide. Shelter services and childcare service are also provided rarely but with minimal need for more provision. In this group, we have the largest differences between urban and rural municipalities.
8 Discussion and conclusions
The purpose of this study was twofold. We wanted to analyse access to local social services and the impact that municipal conditions have on this, but also to broaden the analysis by including the range of services municipalities are expected to provide. This allowed us to go beyond studying single target groups or individual services and to raise the issue of the access to local social services in general. Spatial access, as an expression of local capacities and central requirements to provide social services is an important topic, especially in small countries or sparsely populated areas. It directly concerns people’s lives and particularly affects the situation of the most vulnerable and their access to social rights. Estonia is a particularly interesting case. It is divided into a large number of municipalities. Its institutional frame of local social service provision can be characterised by an unclearly defined assignment from the national level, by limited local financial resources and by large local discretion regarding services administration. This frame sets local structural conditions in terms of remoteness, resources and spending in focus.
The range of possible factors explaining variations in access to local social services is wide and only a small fraction of these are included in this study. Yet, we believe that with our focus on remoteness, resources and spending we can add crucial knowledge to the discussion about spatial access to social rights. In addition, the data used are unique to Estonia, and they provide for the first time an opportunity to gain more detailed insights in the pattern of access to local social services. Access was defined as the extent of and the additional demand for social services.
All Estonian municipalities are expected to provide every local social service included in this study. Like previous studies (e.g. Pugh, 2003), we found considerable variations in access to local social services, both in general and when distinguishing between individual services. Certain services are provided almost everywhere, while others are provided less often, particularly in rural areas. Among the latter, for some services there was additional demand, while for others the supply was felt to be sufficient. Explanations for the varying access pattern to local social services might be: a) target groups and character of services can be varying, b) local social services can be provided more economically elsewhere, and c) local social services can be substituted by alternative measures.
In the first case, access to services is related to the target group size and character of service. Thus, we can see that access to services directed at broad target groups is more common than access to those directed toward smaller groups. Home care services or counselling services are for example directed at broad groups, whereas personal support services are targeted to smaller groups (e.g. disabled with specific needs). Also, social transport is a service that is used by various groups, and is particularly important for people living in rural areas. In general, rural municipalities provide more rarely services that are targeted at smaller groups. With respect to the services' features, some need to be delivered near or at the residence of the target group, as for example personal support services for disabled people since these are not as mobile as non-disabled. Proximity is here a crucial feature, both for service provision and the support person or personal assistant who needs to originate from the neighbourhood. These services are staff intense and costly and municipalities might not prioritise them. Small and remote municipalities may not always have the resources and capability to deliver services of this kind (and they might also have difficulties in finding staff), whereas it may be easier for larger municipalities and municipalities closer to county centres to organise personal support services. Another explanation for (limited) access could be that local social services can be provided more economically elsewhere. This explanation might apply to services that are provided to a lesser extent, but where the local demand for additional delivery is also low (e.g. shelter services). Municipalities far from county centres and those with low population density rarely provide those services. Services might be too costly; staff competence is not available or it is not as crucial that service delivery take place close to people's homes. Thus, it might be more economical to pool resources and provide these kinds of services in form of institutions. A similar pattern but with different explanation can be found in the case of childcare services. Also, these are only provided in few municipalities and additional demand is assessed as low (especially in rural areas). Many families with small children have in recent years moved to urban or rural municipalities close to larger towns or county centres. Those municipalities, however, have not been able to react to these changes and to provide sufficient number of day-care centre places, the main care service for small children. This has resulted in an increased demand for alternative local services – mostly childcare services. In contrast, rural municipalities were mostly able to provide child day-care centre service (Riigikontroll, 2015), and consequently express lower additional demand for alternative (childcare) services.
The impact of local conditions was shown to be strong in particular with regard to remoteness indicators and resources. Distance to county centre and to capital city, as well as population density were in our analysis more relevant for access to local social services then mere population size. The only indicator that is of minor or no importance is spending. A reason for that may be that the indicator of local social protection spending is a basket including various items such as spending on social cash benefits, social services, investments in local social policy, administrative spending on local social policy, etc. In addition, this indicator does not say anything about how efficiently or inefficiently the local social services are provided. The role of remoteness and local resources is an important finding and relates directly to the current discussions in Estonia and elsewhere about territorial-administrative reforms of local governments. Voluntary mergers of municipalities are an ongoing process and the main argument for the amalgamation of (small) municipalities is improved access to and quality of local services. Discussions of territorial-administrative reforms often solely focus on population size, whereas the division of resources and functions between the territorial levels is absent. However, the present study has pointed out that with respect to provision and delivery of local social services, proximity to county centre and capital city is an even more important factor than mere population size. Thus, it is necessary to broaden the so-far narrow focus of the debate to include several dimensions. Furthermore, this narrow focus on population size when debating the merging of municipalities can even have reverse effects in the sense that the distances within the municipality increase and the way to the municipal centre where locally provided services are usually delivered in fact gets even longer. That means new obstacles for access to local social services and an even bigger risk of jeopardising the already exposed situation of vulnerable groups in remote areas.
The study also shows that the connection between social rights and access to local social services is not uncomplicated. Local capacities and needs confronted with requirements to offer local social services demand creative and innovative solutions from local politicians, social workers and even local inhabitants; particularly in remote areas where resources are limited and population is spread out. There is an obvious risk that efforts to solve the ‘dilemma of the municipalities’ may result in certain target groups having to pay the price. Local services that need to be delivered individually and close to the needy person’s home are often targeted at the most vulnerable, while being also the most easily neglected or delivered in other (less costly) forms. This confirms our suggestion that studying the realisation of social rights at the local level requires a more nuanced approach.
Ainsaar, M., & Soo, K. (2009). Kohalike omavalitsuste toetus lastega peredele 2008 ja laste päevahoid 2008-2009 Eestis, Tartu: Tartu Ülikool.
Allard, S. (2004). Access to social services: the changing urban geography of poverty and service provision, Metropolitan Policy Program Survey Series, Washington DC: The Brookings Institutions.
Allard, S. W., & Cigna, J. (2008). Access to social services in rural America: the geography on the safety net in the rural west, Rural Poverty Research Centre Working Paper, (08-01).
Andreotti, A., Mingione, E., & Polizzi, E. (2012). Local welfare systems: a challenge for social cohesion, Urban Studies, 49(9), 1925-1940.
Atkinson, R. (2000). Combating Social Exclusion in Europe: The New Urban Policy Challenges, Urban Studies, 37(5-6), 1037-1055.
Australian Institute of Health and Welfare (2004). Rural, regional and remote health: A guide to remoteness classification, Canberra: Australian Institute of Health and Welfare.
Belanger, K., & Stone, W. (2008). The social service divide: service availability and accessibility in rural versus urban counties and impact on child welfare outcomes, Child Welfare, 8(4), 101-124.
CEMR (2008). Balancing democracy, identity and efficiency. Changes in local and regional structures in Europe. Council of European Municipalities & Regions. http://www.ccre.org/docs/changes_in_local_and_regional_structures_web_EN.pdf.
Council of Europe (2001). The Role of Local Authorities in the Field of Local Social Services, Strasbourg: Council of Europe Publishing.
Gyuris, F. (2014). Basic Education in Communist Hungary: A commons Approach, International Journal of Commons, 8(2), 531-553.
Daly, M. (2002). Access to Social Rights in Europe, Strasbourg: Council of Europe Publishing.
Elster, J. (1992). Local Justice: How Institutions Allocate Scarce Goods and Necessary Burdens, New York: Russell Sage Foundation.
European Commission (2005). Report on social inclusion 2005 An analysis of the National Action Plans on Social Inclusion (2004-2006) submitted by the 10 new Member States, Brussels: European Commission.
European Commission (2008). Regions 2020: Demographic challenges for European regions, Commission of the European Communities.
European Commission (2015). Country Report Estonia 2015: Commission Staff Working Document, Retrieved February 2016 from
Ferrera, M. (2005). The Boundaries of Welfare: European Integration and New Spatial Politics of Social Protection, Oxford: Oxford University Press.
Green-Pedersen, C. (2007). More than data questions and methodological issues: theoretical conceptualization and dependent variable 'problem' in the study of welfare reform. In J. Clasen, & N. Siegel, Investigating Welfare State Change: The 'Dependent' Variable Problem in Comparative Analysis, Cheltenham: Edward Elgar Publishing Limited, pp. 13-23.
Groenningsaeter, A., & Kiik, R. (2009). My work is exciting' - social workers in Estonia compared to Norway, European Journal of Social Work, 12(1), 25-39.
Gustafsson, G. (1980). Modes and effects of local government mergers in Scandinavia, West European Politics, 3(3), 339-357.
Jensen, P. H., & Lolle, H. (2013). The fragmented welfare state: explaining local variations in services for older people, Journal of Social Policy, 42(2), 1-22.
Kazepov, Y. (2010). Rescaling Social Policies: Towards Multilevel Governance in Europe. Farnham: Ashgate.
Kröger, T. (1997). The Dilemma of municipalities: Scandinavian approaches to child day-care provision, Journal of Social Policy, 26(4) 485-507.
Larsen, A. C. (2002). Municipal size and democracy: a critical analysis of the argument of proximity based of case of Denmark, Scandinavian Political Studies, 25(4), 317-332.
Lien, S., & Pettersen, P. A. (2004). Local government and welfare generosity: municipality spending on social welfare, Scandinavian Political Studies, 27(4), 343-365.
Marshall, T. H. (1950). Citizenship and Social Class and Other Essays, Cambridge: Cambridge University Press.
Medar, M. (2004). Ida-Virumaa ja Pärnumaa elanike toimetulek: Sotsiaalteenuste vajadus, kasutamine ja korraldus. Coping of the population of Ida-Virumaa and Pärnumaa: need for, use and administration of social services. PhD Thesis, Tallinna Pedagoogikaülikool.
Minas, R. (2005). Administrating poverty. Studies of intake organization and social assistance in Sweden. PhD Thesis, Stockholms Universitet.
Mitendorf, A., & van Ewijk, H. (2015). Work with social complexity in a neoliberal society: challenges for social work in Estonia, European Journal of Social Work, 1-14.
Mohan, J. (2003). Geography and social policy: spatial division of welfare, Progress in Human Geography, 27(3), 363-374.
Nordregio (2015). Reforms sweeping over the Nordic countries?, Nordregio News, 3, September 2015.
OECD (2010). Strategies to Improve Rural Service Delivery, Paris: OECD Publishing.
OECD (2011). Estonia: Towards on Single Government Approach, Paris: OECD Publishing.
Pinch, S. (1997). Worlds of Welfare: Understanding the Geographies of Social Welfare Provision, London: Routledge.
Powell, M., & Boyne, G. (2001). The spatial strategy of equality and the spatial division of welfare, Social Policy & Administration, 35(2), 181-194.
Pugh, R. (2003). Considering the countryside: is there a case for rural social work?, British Journal of Social Work, 33, 67-85.
Rauch, D. (2005). Institutional fragmentation and social services variations: A Scandinavian comparison PhD Thesis, Umea University.
Reiljan, J., & Ülper, A. (2011). The Necessity of administrative territorial reform, Discussions on Estonian economic policy: Theory and practice of economic policy, 190-217.
Reiljan, J., Ülper, A., & Puolokainen, T. (2012). Demand side factors of local public services in Estonia, Discussions on Estonian economic policy: Theory and practice of economic policy, 165-185.
Riigikontroll (2012). Avalike teenuste pakkumise eeldused väikestes ja keskustest eemal asuvates kohalikes omavalitsustes. Assumptions for provision of public services in small and remote local authorities, Tallinn: Riigikontroll.
Riigikontroll (2015). Lasteaiakohtade kättesaadavus valdades ja linnades. Accessibility of preschool places in municipalities, towns and cities. Tallinn: Riigikontroll.
Sootla, G., Kalev, L., & Kattai, K. (2009). Perspectives of local government amalgamations in a transition society: the case of Estonia, Studies of Transition States and Societies, 52-65.
Statistics Estonia (2016). PO0241: Population, 1 January by Year, Sex, Administrative Unit or Type, Retrieved January 2016 from http://www.stat.ee.
Swianiewicz, P. (2010). If territorial fragmentation is a problem, is amalgamation a solution? An East European perspective, Local Government Studies, 36(2), 183-203.
Trasbeg, V. (2009). Baltic local government's fiscal situation: two decades of reforms, Journal of Baltic Studies, 40(2), 179-200.
Trydegård, G.-B., & Thorslund, M. (2001). Inequality in the welfare state? Local variation in care of the elderly - case of Sweden, International Journal of Social Welfare, 10(3), 174-184.
Turbett, C. (2009). Tensions in the delivery of social work services in rural and remote Scotland. British Journal of Social Work, 39, 506-521.
van Mechelen, N., & De Maesschalck, V. (2009). Devolution of Social Protection Arrangements. In A. Cerami, & P. Vanhuysse (eds), Post-Communist Welfare Pathways: Theorizing Social Policy Transformations in Central and Eastern Europe, Hampshire: Palgrave Macmillan, pp. 181-198.
Young, I. (1990). Justice and the politics of difference, Princeton, New Jersey: Princeton University Press.
School of Governance, Law and Society
Department of Social Work
 This was the number of the municipalities at the time the study was conducted in 2013. By 2016 the number of municipalities has decreased to 213, due to some voluntary mergers.
 The authors thank Merle Piho who as a social work doctoral student at Tallinn University significantly contributed to data collection from social workers of local municipalities by means of a questionnaire.