1The tendency for organisations to focus on their core tasks and outsource support activities is a phenomenon which has emerged worldwide since the 1980s in both the private and public sectors [Coase, 1991; Hermann and Flecker, 2013]. In Belgium, and in the Brussels regional administrations as well, tasks have been outsourced and privatised increasingly. The main reason for this trend is the possibility to reduce organisational costs by outsourcing relatively simple activities or tasks which require little specific training – the cleaning of buildings and office spaces, security services and catering are obvious examples. Since many of these outsourced activities are done by relatively low-skilled workers, the outsourcing tendency is important for the labour market position of low-skilled workers, which has been weakening over the past decades (with higher unemployment rates, downward wage pressure and threats to job quality in most developed countries).
2In recent years, a countermovement has emerged, as the monetary economies derived from outsourcing seem to have been overtaken by the extra organisational costs: some previously outsourced activities are now being re-internalised, or “backsourced”, in a growing number of organisations. The aeroplane manufacturer Boeing and IT firms in Silicon Valley are well-known examples [Economist, 2016; Petitjean and Kishimoto, 2017]. While outsourcing often originates from a mere financial cost-benefit analysis, backsourcing often occurs due to dissatisfaction with the quality of services provided. Apart from these organisational arguments – and, certainly, for public organisations – insourcing can also be an instrument of labour market policy to improve the working conditions of groups in a weak labour market position or to meet other aspects of (corporate) social responsibility. While outsourcing is a means to rationalise (i.e. reduce) the operational costs of an organisation (public or private), backsourcing stems from a consideration of a more complete view of management costs, which are more difficult to assess properly, as well as the societal impact of a (public) organisation.
3The aim of this study is to investigate the outsourcing and backsourcing phenomena for the Brussels regional bodies by focusing on three research questions: (1) Which tasks have been outsourced, and why? (2) What has the impact of outsourcing been on the workers and organisations involved? (3) How do stakeholders experience outsourcing, and what are their attitudes towards a possible re-internalisation (i.e. backsourcing to public organisations)? In a cross-cutting approach, we focus on the specificities of the Brussels case, if any. In order to answer these questions, we have relied on data collected during a multidisciplinary research project [Vanroelen et al., 2018]. During the preliminary investigation, secondary information was collected on the Brussels regional public services as a whole (online information, annual and operating reports as well as other policy documents and available data sources). A general survey on outsourcing practices was also conducted with the heads of department or their representatives. Subsequently, a detailed case-study was carried out for four public administrations. For these selected cases, the study was supplemented by in-depth interviews with key persons, the analysis of internal documents and the legal analysis of employment contracts. In the conclusions, nine fundamental dimensions to be considered when out/backsourcing tasks are suggested for policymakers and stakeholders.
4The paper is structured as follows. Section 2 reviews the literature on outsourcing and backsourcing. In Section 3, the selection of the cases and the methodology is presented. Section 4 presents the main results and findings. Section 5 summarises the main lessons learned from the research for policymakers and stakeholders.
5Since the 1980s, both the private and public sectors have been characterised by a thorough rationalisation of business operations, and it is in this context that outsourcing has undergone a significant rise. Since the Thatcher era in the Anglo-Saxon world, the “New Public Management” approach imported the idea of outsourcing into public organisations, arguing that public monopolies are inefficient and costly for citizens [Hermann and Flecker, 2013]. In this context, Belgium was no exception, and a wave of outsourcing took place during the 1990s, including privatisations in the public sector. However, signs of a countermovement have emerged recently: more and more private companies and public organisations have decided to re-insource previously outsourced services – the so-called “backsourcing” phenomenon. The reversal to backsourcing in some well-known companies such as Boeing, has received a lot of attention in the news [e.g. The Economist, 2016]. At the same time, both local and national governments are taking control once again of a number of services which had been outsourced previously: a survey describes 835 public services which have been backsourced recently [Petitjean and Kishimoto, 2017].
6Why are some tasks outsourced, and others not? According to the Nobel Prize winner Ronald Coase (1991), outsourcing involves both costs and benefits. Specifically, he refers to the transaction costs associated with drafting a contract and checking compliance with it. The benefits of externalisation come mainly from lower prices due to economies of scale at the level of the external supplier which is able to specialise. The increasing degree of outsourcing over the past decades can thus be explained by falling communication and transport costs, which may have radically reduced transaction costs. Accordingly, activities are more likely to be outsourced if the benefits greatly outweigh the transaction costs. This should be expected for services which are subject to substantial economies of scale, services which are well defined and routine so that contracts with external suppliers are easy to draw up and monitor, and activities in which a significant reduction in costs is possible via privatisation if regulations and wages are more costly in the public sector. On the other hand, complex tasks entail high transaction costs, as they often lead to conflicts regarding compliance with the contract [Child, 2015]. Therefore, the first cases of backsourcing were related to IT activities, a sector characterised by complex and constantly evolving tasks [Deloitte LLP, 2014; Giest, 2018].
7The conceptual dichotomy between insourcing and outsourcing is too simplistic. In real-world practices, a wide variety of intermediate forms are possible between the “pure” outsourcing to a supplier in the private market and the performance of in-house tasks [Verhoest et al., 2004]. Between the two extremes of pure outsourcing and in-house management, there are other alternatives. For example, certain tasks can be assigned to an internal, autonomous unit (“internal privatisation”) or they can be outsourced to an external public organisation or to (external) not-for-profit organisations (see also [Bouckaert et al., 2010]).
8Choosing an appropriate form of management requires an in-depth evaluation of the various options as well as ex-post evaluations on the achievement of expected results and possible adjustments. The advantages of outsourcing can be, for instance, the promotion of a result-oriented approach, better control of operational costs, clarification of the role of the service provider, the use of specialist knowledge for selected service providers, flexibility, risk-spreading and quality improvement. However, the risks and possible disadvantages of outsourcing should also be considered, such as contract establishment (rather than focus on the real objectives of the service), transaction costs, “hidden costs” and unforeseen events, the loss of in-house expertise, poor coordination (with monitoring and evaluation), little flexibility to deviate from the contracts and possible worsened work- and employment conditions for employees. In the case of public organisations, outsourcing also raises the concern that the distance may grow between citizens and the services which are delivered.
9The cost savings resulting from outsourcing are not always achieved only through a more efficient production process, but often also through poorer (cheaper) terms of employment and working conditions. Goldsmith and Schmieder (2017) demonstrated that saving labour costs is an important reason to outsource. For example, the wages of outsourced low-skilled workers in Germany are 10 to 15 % lower than those of similar workers who are employed internally. The authors suggest that the trend towards outsourcing over the past decades has been an important cause of the increasing income inequality in Germany during this period. After all, outsourcing allows a level of wage inequality which might not be tolerated within organisations, as it would be resisted by workers and unions [Weil, 2014]. Although a great deal of empirical research shows that outsourcing has led to lower wages [e.g. Dube and Kaplan, 2010], it must be noted that there is also research which reports positive effects [e.g. Munch and Skaksen, 2009]. In any case, the extent to which working conditions change depends on the reasons for outsourcing (increased efficiency versus fewer regulations and lower wages) and on the power of trade unions and the regulation of working conditions in collective agreements and in outsourcing contracts. All of these elements highlight the complex framework to be considered when outsourcing or backsourcing a task.
- 1 www.rijksschoonmaak.nl
- 2 www.ubrijk.nl/organisatie/ubr-rbo
- 3 See [Vanroelen et al., 2018] for more details and references regarding the case of the Rijksschoonm (...)
10In various countries, there have been policy initiatives for backsourcing in the public sector. A very interesting and practical example of backsourcing in the public sector is that of the Netherlands, where the national government has recently backsourced its cleaning services.1 The Dutch government has set up the National Cleaning Organisation (Rijksschoonmaakorganisatie, RSO), giving the status of civil servant to previously outsourced cleaning workers. The RSO was launched on 1 January 2016, and by the end of 2017 had 700 cleaners working in 190 locations. By 2020, it plans to grow into an organisation of 2 000 cleaners. A similar organisation has been created to provide security services for government buildings (the Rijks Beveiligings Organisatie)2. This backsourcing policy (among other objectives) aims to ensure better working conditions for workers at the lower end of the labour market, namely job security and career opportunities. Moreover, a major benefit for the RSO is the possibility to exploit economies of scale thanks to the large amount of buildings owned by the Dutch government. On the other hand, the launch of the RSO met with resistance from the private sector, which had lost a market. Nine private cleaning companies went to court claiming that the RSO benefits from unauthorised state aid and that it violates public procurement and competition laws. However, the Dutch Court did not follow this line of reasoning and declared that the RSO does not infringe competition law since it does not provide services to third parties, and therefore does not carry out any economic activity. Along this line, the RSO only takes over outsourcing contracts when they end in order to avoid the risk of further legal issues. By appointing cleaners (usually low-skilled) as civil servants, the Dutch government wants to send the message that low-skilled tasks are also important in all government services3.
11In this context of general outsourcing with a recent counteracting trend towards backsourcing, the Brussels regional administrations are no exception. Based on a project for Talent.Brussels, the regional agency for human resources, an interdisciplinary research team from the Brussels Studies Institute (BSI) has explored the possibility of backsourcing low-skilled functions by comparing situations of outsourcing and insourcing in the context of the Brussels-Capital Region [Vanroelen et al., 2018]. The aim of the study was to investigate which tasks have been outsourced, the impact of outsourcing, and the attitudes of stakeholders towards a possible re-internalisation. The focus is on low-skilled activities, such as cleaning, security services, gardening and catering.
12As regards the methodology, the research followed four steps: 1) analysis of the literature and documents (legal documents, policy reports and internal materials); 2) a workshop in which the different heads of department participated; 3) an online survey among sixteen department heads from Brussels regional administrations; and 4) a series of seven in-depth interviews with department heads, HR experts and trade union representatives (five personal interviews and two group interviews). For some parts of the research (for example the interviews and the analysis of legal documents), three cases were selected for a more in-depth study:
13(1) Actiris, the regional employment service;
14(2) The Brussels regional public service (BRPS), specifically its Facilities Department and Brussels Mobility;
15(3) Brussels Environment, the regional public service for environment and energy.
16A more detailed description of data and a discussion of the underlying method is available in the full report [Vanroelen et al., 2018].
17Based on the questionnaire presented to all Brussels regional administrations, it appears that the tasks most commonly outsourced are cleaning, security and gardening services. IT support, building maintenance, catering and printing are also occasionally outsourced. Even though each department/agency has followed a different context-specific path, the most important reasons for outsourcing are the need for specialised knowledge and a lack of internal resources or staff. In addition, other stated reasons for outsourcing are greater efficiency, more flexibility and higher quality. Some respondents pointed out that the choice to outsource a certain activity has grown historically or was imposed on a top-down basis.
18When we focus on the organisations selected for the case studies (see the overview in Table 1), it is striking that security services are always outsourced. The outsourcing of security should be considered as a special case due to the specific regulations of this type of activity. Due to the so-called “Tobback Law” (adopted on 10 April 1990), strict requirements and authorisations are required in order to perform private surveillance and security tasks. Only specialised and accredited security firms are allowed to operate in this sector. It is also noticeable that cleaning is an activity which is at least partially outsourced by all selected services (although the various organisations deal with it differently, as we will discuss below).
Table 1. Outsourced tasks in selected regional organisations
Note: OUT: outsourced activity (shaded cells); IN: internal activity; MIX: partly internal and partly outsourced; n/a = not applicable or information not available.
Source: Vanroelen et al., 2018
19A closer analysis of the selected cases reveals considerable differences in the views and practices regarding outsourcing, which tend to be context-specific and path-dependent. More relevant findings come from the management views on this theme. It can be said that Actiris is very conscious and cautious when it comes to outsourcing support services, and also explicit as regards the consideration of criteria other than efficiency or the cost control element. Actiris management highlights the tension between mainstream outsourcing dynamics and the central role of these practices on the labour market in Brussels. The specific role of Actiris probably explains its decision to keep its catering service organised internally, and to experiment with mixed models for cleaning and security/reception services (though the Tobback Law sometimes forces outsourcing). The other regional bodies have more pragmatic views on the topic, which are mainly inspired by cost efficiency and manageability reasons. While every organisation has its own specificities such as road maintenance for Brussels Mobility, the case studies show that some regional organisations have started introducing demands for social and/or ecological specifications in their tender for public procurement. For example, Brussels Environment has included social and technical clauses which go beyond the simple cost and quality guarantees, such as specific attention to the ecological dimension(s) of companies operating as suppliers. Other organisations tender exclusively for social economy organisations. A general openness exists in all administrations regarding new management models offering alternatives to private market outsourcing, though a clear, precise model does not seem to have emerged yet due to the specificities of each context.
20In various cases, only part of the support activity is outsourced, i.e. specific sub-tasks. The most commonly outsourced (sub-)tasks are those which have to be done at atypical or flexible moments. In this way, internal employees are able to retain more interesting activities, outsourcing the most demanding and least interesting duties. A good example is the seasonal and very routine activities of garden maintenance, such as clearing leaves in autumn: few internal staff seem interested in this activity as such, as it is a task which is unlikely to further their careers. Another example is routine cleaning activities such as the daily maintenance of large office spaces, which usually takes place outside office hours. The management of such tasks often poses an extra challenge for mid-level management: outsourcing eliminates difficult coordination and HR-related activities (e.g. planning and management, motivation and absenteeism of staff and taking care of turnover). In a sense, outsourcing such activities is a way to create a “qualitative buffer” for internal employees (they can focus on more interesting sub-tasks within the organisation) and for the mid-level management which just has to outsource a potentially time-consuming task which is not a core activity. In this way, many management-tasks are also outsourced together with the support activity itself. As tender models for the outsourcing of support services have become a standard routine (e.g. including standard documents and procedures, a market of service providers which is adapted to this mode of operation, etc.), the risks for the organisation associated with outsourcing practices have become low and controllable.
21Indeed, the established character of outsourcing practices provides various preconditions offering incentives for outsourcing and at the same time obstacles for internal management or procurement scenarios. A first theme has already been discussed above: the unattractiveness of certain tasks for the internal staff and the related HR-management. Another relevant factor is the inflexibility of recruitment procedures within the public service. More specifically, in areas with high staff turnover (e.g. cleaning) or for activities in which the need fluctuates (e.g. specific tasks of green management), outsourcing often offers more possibilities in terms of flexibility. For surveillance and security, the strict regulation of the profession plays a role once again. The legislation on security firms makes it very difficult for the core security duties to be performed by public service employees themselves. Even if one is willing to split off related tasks from monitoring functions and organise them internally (e.g. reception functions at Actiris), one is obliged to continue to outsource the core security activities. Similarly, there are also situations which require large investments in order to be able to meet new and ever-stricter standards (e.g. food safety in industrial kitchens). In such cases, outdated equipment can be a reason for outsourcing. Another common precondition was the (apparent) political pressure to keep public institutions “lean” in terms of staffing. The logical consequence of internal procurement, namely the increase in the number of internal staff, was cited by several interviewees as an important obstacle, even though (internal) procurement could be useful from a cost or quality point of view. This obstacle seems to come down to a political and symbolic obstacle: it is sometimes more feasible from a political point of view to spend considerable outsourcing budgets instead of the sustainable financing of an internal staff framework. However, it must also be clear that widening the internal (statutory) framework is a decision with a budgetary impact which may continue for several decades, while outsourcing at a budgetary level allows a shorter-term perspective.
22A legal analysis of the employment and wage conditions of low-skilled workers, considering those who are employed directly by government institutions as well as private sector subcontractors, shows that there are significant differences between “internal” (public sector) and “outsourced” (private sector) employees. Differences are clear in particular regarding job security, wages, holidays and social security. Table 2 provides a complete overview of this comparison. For starting employees (considering the level D workers, as specified in Table 2), wages seem more interesting in the private sector, but the public sector applies a scaled system evolution based on seniority (the barème/barema) which provides a better progression. Public sector employees generally benefit from more holidays and greater job security, whereas in the private sector, many employment and wage conditions are laid down at sectoral level, and these rules offer only minimal protection. In terms of social protection, the statutory (i.e. public) system offers the most certainty. However, most workers from the private sectors surveyed benefit from additional coverage from the FSE-FBZ (Fonds de Sécurité d’Existence - Fonds voor Bestaanszekerheid). Overall, the contractual government employees, temporary workers, students and social economy employees seem to be the least favoured.
Table 2. Comparison between public and private sector for selected activities
Source: Vanroelen et al., 2018
23The results of this comparison are in keeping with what we have pointed out in the case studies: although a public sector status generally seems somewhat more advantageous than most private sector statuses, the advantage certainly does not apply in all cases nor to all aspects. This finding should be duly considered when policy-makers consider re-internalising certain activities in order to improve the employment conditions of low-skilled workers. For instance, it is worth bearing in mind that in the Dutch RSO, the trade unions were involved in writing a detailed agreement which regulates and guarantees the working conditions of workers who move from the private to the public sector as a result of the backsourcing process. Considerations regarding saving on wage costs often form the basis for outsourcing decisions. Numerous mechanisms allow subcontractors to reduce wage costs more effectively. However, focusing on direct wage costs alone may be misleading: net wages may in some cases be higher in the private sector, as shown in Table 2, but the public sector provides other benefits such as higher pension entitlements. Employees do not always see or consider more indirect or long-term benefits.
24A classic argument put forward in the scientific literature in favour of backsourcing is the improvement of the quality of services. The interview data from our case studies suggest that this argument is not prominent. Overall, the participants were positive about the quality of the services provided by external companies. One element which has been mentioned several times is the perceived lack of flexibility of the outsourced activities. Most of the time, the content of the assignment is clearly described in the (tender/subcontracting) specifications, making it impossible to adjust the assignment without incurring additional costs. This may be due to a lack of expertise in the elaboration of specifications. But progress has been made, for example by working with framework agreements, which allow a more flexible deployment of subcontractors. The discussion about the quality of services also touches on the earlier theme of the content of work for internal staff. It seems that activities which are hard to control and complex, small-scale and demanding new tasks are preferably done by an internal team. Only after an activity becomes controllable and plannable does it become eligible for outsourcing. A good example of this is Brussels Environment: new parks are first under internal management and only later are specific gardening tasks outsourced. All of this shows that, from a quality perspective, the greatest potential for backsourcing is probably in tasks with a relatively high added-value. For more routine tasks, quality control could be easier by transferring the management risk (e.g. ensuring that enough staff are available) to a subcontractor.
25Both outsourcing and internal management incur different costs which are hard to estimate and are often overlooked when making management decisions, for example, the costs related to setting up HR policies on recruitment, training, absenteeism and motivation. Therefore, any internalisation of routine tasks is likely to increase the need for internal management functions as well. Nonetheless, participants in our research point out that the outsourcing of activities also creates unintended management work related to the preparation and follow-up of public tenders. The correct elaboration of specifications requires specialised knowledge which is not always present in public institutions and which requires the recruitment of experts. It also appears that more (and more costly) legal conflicts have arisen as a consequence of appeals from candidate-outsourcing companies which have lost the bid for a contract. The hidden costs also refer to the monitoring and enforcing of the conditions stipulated in the outsourcing agreement. These hidden costs are likely to increase as more social or ecological clauses are included in the agreements.
26Through the in-depth interviews, we also investigated the attitudes (concerns, suggestions and expectations) of stakeholders towards a possible re-internalisation of certain services. Two proposals were made: uniting services at regional level and developing a more social subcontracting policy. Based on the interviews, the two main concerns can be summarised as follow: “wages in the public sector are not appealing enough to attract private staff”, and “reinternalisation would only result in the transfer of jobs from the private sector to the public sector” without the creation of new jobs. However, the respondents also saw benefits of reinternalisation, such as higher salaries, improved coordination and the possibility to involve workers in higher-quality work.
27Broader policy objectives are important for decisions regarding a possible reinternalisation. A decision based purely on arguments about quality, cost price or efficiency would be debatable in many cases. The respondents believe it is important to emphasise broader objectives such as the implementation of a (regional) labour market policy or the improvement of working conditions. However, they point out that in such a context, reinternalisation would require a clear political decision in order to pursue such standards in public organisations. However, even such broader objectives do not exclude the possibility of outsourcing, which might be done in different ways such as targeting the social economy or strengthening the sustainability requirements in the tender specifications. In such a context, however, the private market might no longer be inclined to provide services and an alternative form of management might be a better choice.
28Outsourcing and insourcing require considerable transition phases with associated costs. In the case of insourcing, it should be borne in mind that the transition of existing employees from private firms to the public sector may be complex. This is also the case with movements in the reverse sense, or when an outsourcing contract is won by another company. However, one must be aware that there is no guarantee that insourcing implies that wage and working conditions will improve for all external staff. As a consequence, an insourcing operation may create resistance among internal staff (in terms of what will change for them) or among middle management who will oversee the coordination of the new internal activities. Moreover, certain legal aspects of the public employment relation may prove difficult to reconcile with the nature of the work, such as the ban on night work. It is therefore possible that a large-scale reinternalisation would also require a restructuring of the public service itself in terms of its internal labour organisation.
29The transfer of the organisation of services in the public sector from one form of management to another is a complex matter which involves public management, business economics, legal and broader political aspects. The case studies have taught us that every function and institution entails a series of specific challenges which may be generalised only partially, and are often specific to the situation such that the concrete challenges can rarely be anticipated fully. In this perspective, the Brussels cases do not seem to differ from the general countermovement from outsourcing to backsourcing, though each case is hardly comparable. Nonetheless, several challenges have emerged throughout our research, identifying general questions beyond our cases. When considering a decision about outsourcing or backsourcing, these general questions can be synthesised as follow:
301. Clarify all objectives in an explicit way. The question of the most appropriate form of management is undoubtedly dependent on the set of objectives which policymakers want to pursue, which are often implicit. For example, an exclusive focus on cost efficiency in the short term is likely to lead to a choice other than a set of more articulated, forward-looking and complex objectives, such as labour market policy or the quality of working conditions, which also play a role.
312. Try to have a good idea of the substantive characteristics of the activities and tasks which they want to out/backsource. What are the job contents and required skills? How far ahead can the activities be planned and how much flexibility is needed? What are the operational expenses and existing regulations of these activities? Finally, can these activities form part of an attractive job content for permanent in-house employees?
323. Clearly identify which specific sub-tasks fall under a specific function, and then elaborate the management accordingly. Even within well-defined activities (e.g. cleaning), sub-tasks could be distinguished, for which the appropriate management form is not necessarily the same (e.g. cleaning could refer to either the routine maintenance of office spaces or to more specific duties).
334. What are the potential hidden costs of the current form of management? These hidden costs concern management as well as long-term personnel costs. On the one hand, it seems that the full long-term cost of an activity is usually not accounted for in current outsourcing contracts; on the other hand, in the case of reinternalisation as well, the potential hidden costs are rarely considered explicitly.
345. What are the options for achieving economies of scale? Some activities are carried out in many different public institutions and therefore lend themselves to a potential integration into a single organisation (public or private) which exploits economies of scale. Some other activities are very specific, so that the scope for alternative management forms is more limited.
356. In the case of a potential reinternalisation, policymakers should consider the compatibility of the regulations regarding public sector workers and the way in which the work is organised. Sometimes, legal hurdles may exist, such as regulations regarding working hours, wage schedules, recruitment procedures or job-specific legislations such as the Tobback Law for surveillance and security.
367. The implications for the relevant labour market should be evaluated carefully. For instance, if a policy is aimed mainly at improving employment opportunities in a specific labour market, it would make no sense to create employment if there is already a shortage of workers in that field.
378. Map the required transitional measures and implications for all stakeholders and not just the body directly involved. A change in management type has implications for both internal and external staff members as well as for other legal entities involved (e.g. the current subcontractors). This can only be done on a case-by-case basis, which should then involve an evaluation of the costs and benefits for all stakeholders.
- 4 In this field, the Federal Planning Bureau published a relevant study while this paper was under re (...)
389. Take into account the type of contract for internal staff members, especially for young people with low qualifications. The Brussels Regional Government has decided to encourage the “Youth Guarantee” through a policy aimed at providing each young person who is registered at Actiris access to a one-year (and in some cases, a two-year) internship, training or employment period4. However, even if the trainees keen to remain in their jobs are guided by the Brussels Region, this system sometimes leads to frustration regarding the impossibility to continue these types of contract and the negative effects in terms of turnover rates and staff motivation. Thus, the challenge lies not only in providing direct access to experience and employment, but also in providing those involved with meaningful, valued and longer-term job perspectives.
The authors are grateful to the Brussels Studies Institute (BSI) for their support in coordinating this research and to Talent.Brussels for funding and research cooperation.