1 Introduction

High levels of competition in the global economy are inducing companies to make strategic decisions to identify the activities that it would be better to conduct internally and those that it would be more appropriate to outsource. In line with the above, they not only outsource the manufacture of their products, but also services that traditionally used to be carried out in-house in the company itself. As a result, business services procurement has become a major part of the way that companies acquire external resources (Murray and Kotabe 1999; Axelsson and Wynstra 2002; van Weele 2005; van der Valk 2008). Services outsourcing is considered substantially different and more complex than the purchase of goods (Stock and Zinszer 1987; Fitzsimmons et al. 1998). As Thomas (1978) and Nie and Kellogg (1999) stated, knowledge of the outsourcing of goods should not simply be transferred to the field of services without any further consideration. This all points to a specific body of knowledge being required for the latter.

Services outsourcing is a strategic decision in the firm, as it tends to improve financial performance (Boyson et al. 1999; Wilding and Juriado 2004; Merino and Rodríguez-Rodríguez 2007; Hsiao et al. 2010), flexibility (Wilding and Juriado 2004; Sahay and Mohan 2006; Núñez-Carballosa and Guitart-Tarrés 2011; Arias-Aranda et al. 2011), customer service (Sum et al. 2001; Sohail and Sohal 2003) and productivity (Razzaque and Sheng 1998; Sum et al. 2001; Sahay and Mohan 2006), among other things. In addition, it enables companies to focus on their core business (Bhatnagar et al. 1999; van Laarhoven et al. 2000; Bustinza et al. 2010; Hsiao et al. 2010) and to access new markets (Sink and Langley 1997; Fernie 1999) and the latest technologies and expertise (Boyson et al. 1999; Wilding and Juriado 2004; Sahay and Mohan 2006; Núñez-Carballosa and Guitart-Tarrés 2011).

Defining the services outsourcing strategy is a complex issue, as there are several dimensions that have to be determined. Following Gilley and Rasheed (2000), there are two basic dimensions that define outsourcing strategy: breadth and depth. Breadth refers to the number of services that are outsourced, while depth refers to whether each of the services is outsourced totally or partially.

In general, the existing empirical literature that analyzes the financial impact of outsourcing has concentrated on only one of the above-mentioned services strategy dimensions, breadth. To be specific, the empirical studies using objective data have analyzed whether there is any relationship between the level of outsourcing (in terms of breadth) and financial performance. Moreover, none of these studies has focused exclusively on the case of services outsourcing, and only two have been found to take into consideration both services and materials outsourcing separately: Görzig and Stephan (2002), and Görg and Hanley (2004). Görzig and Stephan (2002) state that there is a partial relationship between the number of activities outsourced and financial performance, but determine that a greater level of materials outsourcing further improves performance. Görg and Hanley (2004) do not find that services outsourcing improves financial performance, although they state that an improvement does occur with materials outsourcing. On the other hand, studies by Gilley and Rasheed (2000), Jiang et al. (2006) and Salimath et al. (2008) that analyze the outsourcing of materials and services jointly do not arrive at the same conclusions. While Salimath et al. (2008) find a positive effect on profitability depending on the level of outsourcing, Jiang et al. (2006) only find a relationship at the cost level, and Gilley and Rasheed (2000) do not find any relationship at all with profitability. In a specific analysis, Kotabe and Mol (2009) determine that the practice of materials outsourcing influences financial performance. In short, it seems clear that empirical studies that analyze the financial consequences of the level of outsourcing are not conclusive. Moreover, these studies are scarce, and even more so in the case of studies that focus on services outsourcing.

The question that has arisen around services outsourcing is, therefore, whether the contradictory results of previous studies might be due to the fact that they focus only on the outsourcing strategy’s breadth, and ignore another major aspect, its depth. The objective proposed in this paper is to determine whether the depth of the services outsourcing strategy has an impact on financial performance; in other words, whether outsourcing services activities totally or partially has any financial effects and, if so, of what type. As far as is known, there is no empirical study in the literature that focuses on this aim.

The present study therefore seeks to contribute new empirical evidence of the financial consequences of outsourcing services for manufacturing companies. For this, a sample of 1,522 Spanish industrial firms that outsource services was used. The sample was obtained from the Ministry of Industry, Energy and Tourism Survey on Business Strategies (Encuesta Sobre Estrategias Empresariales, ESEE), which is conducted with the utmost rigor and provides an accurate snapshot of the Spanish industrial sector. A cluster analysis identified three groups of firms according to the depth of their services outsourcing strategies. The differences between costs, productivity and profitability were then analyzed for each of the three groups of firms using the analysis of variance (ANOVA) statistical technique.

This article is relevant for research into outsourcing, firstly, as it adopts a different focus for the analysis of outsourcing; compared to the simplistic view as to whether there is a high or low number of activities outsourced, it seeks to consider another dimension of firms’ services outsourcing strategies: their depth. Secondly, it focuses on manufacturing companies in Spain. This is a geographical area that has not been addressed hitherto, as all the previously cited studies have concentrated on companies in the USA, Ireland, Germany and The Netherlands. This case is also of special interest, as services outsourcing is widely used in Spanish industry (López 2002). Thirdly, the sample of companies used is also very large (of previous studies, only Görzig and Stephan (2002) used a larger sample) and includes not only large companies, but also small and medium-sized companies (SMEs); the latter are especially important in the case of Spain (López 2002). Fourthly, this study analyzes costs, productivity and profitability. This provides a broader overview of the financial effects of outsourcing compared to the more limited analyses of the majority of studies undertaken to date, which target the profitability level. Only Jiang et al. (2006) use ratios directly linked to cost reductions, productivity and profitability improvements, but without specifically analyzing the services outsourcing strategy or focusing on the influence that the depth of the strategy has.

To achieve the aim of this paper, the following section analyzes the most relevant empirical antecedents to the present study. Section 3 specifies the hypotheses that are to be tested. Section 4 defines the sample used and the methodology followed. Section 5 sets out the findings of the study, and the final conclusions and discussion are subsequently presented in Section 6.

2 Literature background

To the best of our knowledge, there is no empirical study that focuses on the impact that the depth of outsourcing (of either services or materials) has on financial performance. The sparse literature that analyzes the relationship between outsourcing and financial performance basically focuses on the level of outsourcing measured in terms of breadth, and therefore these are the empirical antecedents to the present study. Moreover, none of these studies has exclusively focused on the case of services outsourcing.

Two studies (Görzig and Stephan 2002; Görg and Hanley 2004) make a distinction between services and materials when analyzing the financial impact of outsourcing. In a study of a broad sample of German manufacturing firms, Görzig and Stephan (2002) analyze profit margin and return per employee, and find that the level of materials outsourcing improves these indicators. A high level of services outsourcing is also found to improve return per employee, but profitability diminishes, although this can be explained by the difficulty for monitoring the quality of outsourced services. In the second study, Görg and Hanley (2004) use a sample of 215 companies in the Irish electrical equipment manufacturing sector. The authors state that there is no relationship between services outsourcing and financial performance measured by profit margin. However, in line with the previous case, they determine that materials outsourcing improves the profit margin.

Other studies by Gilley and Rasheed (2000), Jiang et al. (2006) and Salimath et al. (2008) analyze the outsourcing of materials and services jointly. The first of these takes a sample of 94 North American manufacturing companies and relates the intensity of outsourcing (measured by the number of activities that are outsourced and the extent to which this is done) to profitability measures, such as the profit margin and return on assets. The study concludes that the intensity of outsourcing has no significant impact on these indicators, and the authors state that one of the limitations of their study is the need to explore other financial aspects, such as costs.

Jiang et al. (2006) use a sample of 51 US listed companies that outsource part of their operations to examine the differences between the performance of these companies and their non-outsourcing competitors during the year following notification of outsourcing. These authors take a greater number of financial indicators into consideration. They use profitability (return on assets and profit margin), productivity (inventory turnover, asset turnover and employee productivity) and cost efficiency ratios (general cost divided by sales and operating cost divided by sales). Finally, the authors conclude that the only relationship that exists is between outsourcing and the cost-related ratios.

Salimath et al. (2008) focus their study of the financial effect of outsourcing on an analysis of the profit margin in a sample of 278 entrepreneurial firms in the US industrial sector. The findings show that outsourcing leads to a significant improvement in the profit margin.

Focusing solely on materials outsourcing, Kotabe and Mol (2009) analyze a group of 1,100 Dutch manufacturing companies and determine that a relationship exists between materials outsourcing and financial performance measured by return on value added. They also state that this relationship is non-linear and negatively curvilinear, whereby an optimal level of outsourcing exists for each company and, when this level is surpassed, negative effects start to be produced.

Table 1 summarizes the studies cited and the financial variables that they examine. A Yes indicates studies that have concluded that a relationship exists between outsourcing and their specific financial indicator, whereas a No indicates the opposite. In short, the findings of the studies that analyze the financial consequences of the level of both services and materials outsourcing measured in terms of breadth are not conclusive. Therefore, it is necessary to take into consideration other aspects when focusing on the financial effects of the services outsourcing strategy, such as the depth of the strategy.

Table 1 Empirical studies of the financial impact of outsourcing
Full size table

3 Hypothesis development

As previously stated, the present study analyzes the impact of the depth of the services outsourcing strategy on costs, productivity and profitability.

Firstly, on the theoretical level, the improvement in costs can be explained by Transaction Cost Economics Theory (Williamson 1975, 1985), according to which the decision is made to outsource a good or service if the cost of purchasing it externally is less than that of producing/providing it in-house. In fact, some empirical studies state that outsourcing leads to an average 20% annual cost reduction (e.g., Domberger et al. 2002). Hence, cost reduction has been identified as the main reason why industrial companies outsource (Casale 2004).

One major source of cost reduction comes from the specialization of service providers, and the fact that they provide a service to a high number of companies enables them to achieve economies of scale that can be transferred to the firm that is outsourcing (Roodhoof and Warlop 1999). Moreover, the transfer of services to external providers implies, on the one hand, a reduction in personnel requirements, which is bound to result in an improvement in personnel costs for the outsourcing company (Ellram et al. 2008; Tate et al. 2009) and, on the other hand, it can involve a reduction in investment in assets, which leads to a reduction in fixed costs (Bettis et al. 1992; Gilley and Rasheed 2000). However, as we are focusing on services activities, this reduction may not be quite so obvious as the reduction in personnel costs. The result of all this should be a decrease in the cost of the service and, consequently, a reduction in total costs for the outsourcing company.

Bearing all the above in mind, the greater the depth in a company’s services outsourcing strategy (i.e., the more a company tends to outsource individual services totally), the fewer the personnel and assets it will need to provide the services. Similarly, there will be a greater possibility of taking advantage of the external supplier’s smaller cost structure. The foregoing arguments lead us to propose that following a deeper outsourcing strategy can be expected to lead to cost savings for a firm. Therefore, the first research hypothesis is formulated as follows:

H1 Companies that tend to outsource their services totally present fewer costs than those that do so partially.

Secondly, following to Boyson et al. (1999) outsourcing generates improvements on the productivity level. These improvements can be explained by the Resource-Based View (Barney 1991), according to which the decision is taken to outsource when it enables companies to target their resources towards the activities that generate the greatest added value. Moreover, transferring a service to a specialized company also means that said company not only supplies the service more economically, but also with higher quality and more reliably (Perry 1997). Together with the above-mentioned reduction in requirements for personnel and assets, this should be reflected in a more efficient use of assets and also on the human resources level (Brown and Wilson 2005).

It is therefore reasonable to think that companies that follow a strategy of greater depth in services outsourcing will achieve a tighter focus on the higher added value activities. This, together with a greater reduction in the need for personnel and assets, should result in better productivity. For all these reasons, the second research hypothesis is formulated as follows:

H2 Companies that tend to outsource their services totally present greater productivity than those that do so partially.

According to the various aforementioned theories, the improvements that a deeper services outsourcing strategy is supposed to produce in both costs and productivity should have a positive effect on profitability. Moreover, other operating advantages of services outsourcing are expected that should likewise have positive effects on improvements to profitability. As mentioned previously, these are: improved flexibility, improved customer service, enabling firms to focus on their core business, and enabling access to new markets and the latest technologies and expertise. Given that the firms that can most exploit these advantages are those that commit to outsourcing individual services totally, the third research hypothesis is therefore formulated as follows:

H3 Companies that tend to outsource their services totally present a better profitability than those that do so partially.

4 Methodology

4.1 Sample selection

The data used were taken from the Ministry of Industry, Energy and Tourism Survey on Business Strategies (Encuesta Sobre Estrategias Empresariales, ESEE). This survey provides information on over 5,000 companies with more than 10 employees in the Spanish manufacturing sector.

As stated in the Introduction section, following Gilley and Rasheed (2000) outsourcing strategies are configured by two dimensions: the number of services that are outsourced (breadth) and whether each of these is outsourced totally or partially (depth), and it is this second dimension that the present study analyzes. As such, determining the depth of services outsourcing strategy is a complex task. Research in the field is made difficult by the fact that this is a variable that cannot be obtained from accounting data (such as a company’s sales figures or the volume of assets), which would precisely identify the depth of each company’s services outsourcing strategy. For this reason, a survey method such as that conducted for the database used in this research has to be used to ask direct questions about which activities are outsourced partially and/or totally, and which are not. To be specific, there are two items in the database that enable the depth of the surveyed companies’ services outsourcing to be determined: 1) the number of services that are totally outsourced by the firms, and 2) the number of services that are partially outsourced. It should be noted that the ESEE distinguishes among the following services: legal advice; tax and economic-financial advice; audits; administration; personnel selection and training; IT programming; implementation of IT packages; courier services; machinery hire; surveillance and security; cleaning; and packaging, packing and labeling.

These two items will help group the companies in the sample according to the depth of their outsourcing strategies. In order to homogenize the sample with regard to the breadth of the outsourcing strategy, a strategic dimension not considered in this study, we filtered the sample to obtain firms which stated that they used at least ten of the fourteen services covered in the database used. Apart from this, not all the companies in the database indicated whether they outsourced and some even stated that they did not outsource. All of these were omitted. The final sample comprised 1,522 companies. Our study focuses on the year 2010, as this is the last year for which there is information about services outsourcing (this information is only provided every four years).

To define groups of companies according to the depth of their services outsourcing strategies, the sample was divided into clusters based on the two above-mentioned factors (number of services that are totally outsourced and number of services that are partially outsourced by the firms). This enabled homogeneous groups of companies to be formed with similar companies in the groups but different from the other groups (Frohlich and Dixon 2001). For this, a two-stage cluster analysis methodology was followed. This consisted of first establishing the appropriate number of groups by using a hierarchical cluster analysis of a sub-sample of cases. The K-means algorithm was subsequently used to classify each of the cases into the groups that had been established. Three groups of homogeneous companies were obtained from this two-stage cluster analysis. The methodology used is indicated for exploiting the advantages of both methods (hierarchical and K-means) and for samples of over 200 cases (Ketchen and Shook 1996).

Table 2 shows the number of companies in the sample (in absolute and relative values) that make up each of the groups and their distribution by sector. The centers of the clusters identified and the mean number of totally and partially outsourced services in each strategy group are also shown. The three groups can all be seen to have a number of services that are outsourced quite near to the mean. The differences among the groups therefore lie in the number of services that the different groups outsource totally compared to the number of services that they outsource partially. The three groups identified are as follows:

  1. Group 1 (Total)

    This is the group with the highest number of companies (687) and includes the firms in which services that are totally outsourced (7) outnumber services that are partially outsourced (2). Therefore, this is the group of companies that most clearly commit to a deeper services outsourcing strategy.

  2. Group 2 (Partial)

    This group of companies (421) tends to outsource more services partially (6) than totally (3). This is why it has been identified as the partial services outsourcing group.

  3. Group 3 (Neutral)

    This group of companies (414) does not show a clear trend regarding either partially outsourced services (3) or totally outsourced services (4).

Table 2 Sample companies by depth of services outsourcing strategy
Full size table

4.2 Financial variables

For financial impact to be measured according to the depth of the services outsourcing strategy, first a series of cost, productivity and profitability indicators need to be defined. For this, the indicators identified in the previously-conducted literature review are used (see Table 1). However, not all of these can be considered in the current study as the information is not available in the database used. The indicators used are given in Table 3.

Table 3 Financial variables
Full size table

Total costs and labor costs divided by company total sales have been used as cost indicators. For productivity, sales divided by company total assets and by average number of company employees have been used. And finally, four ratios have been used to measure profitability: return on value added, return per employee, return on assets and profit margin. Data were not available for operating cost, general cost and inventory turnover, which Table 1 shows were some of the indicators selected by Jiang et al. (2006).

4.3 Data analysis

This study aims to find any differences in a series of financial variables between different groups of companies classified according to the depth of their services outsourcing strategies. The most suitable methodology for this is the statistical technique known as analysis of variance (ANOVA) for a single factor. Other regression techniques are not viable, as the depth of the services outsourcing strategy has not been defined as a linear variable. The ANOVA test compares the mean values of the financial variables of companies with different depths of services outsourcing strategy while taking into account the variability of the observations within each group. This method continues to be widely used at the current time (e.g., Kim 2013; Phan and Matsui 2010). The data were analyzed using SPSS software v.21.

As is known, before an ANOVA analysis can be applied, a number of tests have to be performed to check whether the conditions required for this methodology to be conducted are complied with. In the sample used here, the condition of normality of the dependent variables was not met. Despite this, the large size of the sample meant that the ANOVA technique could still be applied. The Levene Test was performed with respect to the condition of homoscedasticity of the variances for the services outsourcing factor. Depending on the result, either the classic ANOVA F statistic or the Welch Test was chosen when no equality of variances was assumed. Finally, the DMS test was chosen for the post hoc comparisons that enable groups between which significant differences exist to be observed.

5 Results and discussion

This section firstly presents the descriptive statistical analysis of the financial variables according to the depth of services outsourcing strategy (Table 4). Subsequently, the variance analysis test is applied to determine whether any differences found in the means comparison are statistically significant (Table 5). All these results are analyzed jointly for each group of financial indicators.

Table 4 Financial variables descriptive statistics
Full size table
Table 5 ANOVA results
Full size table

5.1 Depth of services outsourcing strategy and costs

According to the ANOVA Test, differences are statistically significant for ratio of labor costs but not for ratio of total cost (Table 5). Moreover, the descriptive statistics show that the lowest ratio of labor costs can be found in group 1 (Total). Corresponding post hoc analyses were carried out to identify groups with significant differences in their labor cost ratios. These analyses indicate that there are only significant differences between group 1 (Total) and the other two groups. Therefore, companies that outsource services totally present lower labor costs.

These results partially confirm Hypothesis H1 in the sense that labor costs can be stated to be significantly different depending on the depth of the services outsourcing strategy, but this is not the case for general cost. The reason for this is that certain monitoring costs will be higher when services are outsourced totally, and this results in the effect of the reduction in labor costs being diluted on the total costs level.

5.2 Depth of services outsourcing strategy and productivity

The ANOVA analysis of the indicators used to measure productivity (Table 5) shows that the only differences that depend on the depth of services outsourcing strategy are in employee productivity. Moreover, the mean values of the different groups (see Table 4) show that the best employee productivity appears in the group with greater total services outsourcing. Specifically, the post hoc analysis determines that these differences are significant between group 1 (Total) and the other two groups.

Hypothesis H2 is therefore partially confirmed, as although employee productivity is affected by the depth of the services outsourcing strategy, the same cannot be said for asset turnover. One possible explanation is that the depth of the outsourcing strategy does not have a significant effect on the volume of assets required.

5.3 Depth of services outsourcing strategy and profitability

The ANOVA Test (Table 5) shows that the only significant difference at the profitability level is in the return per employee ratio. The post hoc analysis shows that these differences occur between groups 1 (Total) and 2 (Partial). In addition, the mean values of the return per employee ratio (Table 4) show once more that it is group 1 (Total) that presents the best value. This result partially confirms Hypothesis H3.

6 Conclusions

This article offers new empirical evidence of the financial consequences of outsourcing services for manufacturing companies considering a new approach: the depth of the outsourcing strategy. Hypotheses were tested in this empirical research using a sample of Spanish manufacturing companies and the analysis of variance statistical technique to verify whether differences could be observed in variables relating to cost, productivity and profitability, depending on the depth of their services outsourcing strategy.

The results of the analysis show that, depending on the depth of the services outsourcing strategy, labor cost ratio, employee productivity and return per employee differ significantly. This indicates that the need for personnel is the main variable affected by services outsourcing. In addition, the companies that tend to outsource services totally present the most favorable values for each of these three indicators; on the one hand, according to Transaction Cost Economics Theory, the lesser need for employees drives labor cost reduction, and on the other hand, in keeping with the Resource-Based View, the more efficient use of human resources leads to greater productivity on the employee level. Ultimately, better return per employee is produced as a result of these two improvements. However, companies that decide to follow a partial or neutral services outsourcing strategy find it more difficult to achieve these improvements. Therefore, it can be stated that it is better for individual services to be totally outsourced to obtain better financial performance.

On the other hand, significantly better results have not been observed for ratio of total cost, asset turnover and return on assets depending on the depth of the services outsourcing strategy. Firstly, the reasons for ratio of total cost may lie in greater monitoring expenses that neutralize reduced costs in personnel. Secondly, the reason for the other two asset volume-related ratios is that services outsourcing will affect competencies in services, but not necessarily materials, and the assets required do not vary substantially whether services are outsourced partially or totally.

To conclude, these results confirm the theory that services outsourcing contributes to better human resources performance (Brown and Wilson 2005), although this depends on the depth of the company’s services outsourcing strategy. These findings imply that when companies outsource services they should consider that the depth of the outsourcing services strategy followed has an impact on financial indicators.

The present research requires more detailed future investigation into the effects that the services outsourcing strategy (in its two dimensions of depth and breadth) has on financial performance, in order to better identify any associated advantages and disadvantages.