UEF//eRepository
DSpace https://erepo.uef.fi
Rinnakkaistallenteet Yhteiskuntatieteiden ja kauppatieteiden tiedekunta
2018
E-procurement and SME involvement in public procurement of innovations:
an exploratory study
Saastamoinen, Jani
Inderscience Publishers
Tieteelliset aikakauslehtiartikkelit
© Inderscience Enterprises Ltd.
All rights reserved
http://dx.doi.org/10.1504/IJPM.2018.10012141
https://erepo.uef.fi/handle/123456789/6978
Downloaded from University of Eastern Finland's eRepository
1
E-procurement and SME involvement in public procurement of innovations: An exploratory study
Jani Saastamoinena*, Timo Tammib and Helen Reijonenc Abstract
Current public procurement policies in the European Union (EU) seek to improve small and medium- sized enterprises’ (SMEs) access to public procurement, and emphasize public procurement of innovations (PPI) and aim to promote a switch to e-procurement. However, many information and communication related barriers impede SME access to public sector contracts, which could be mitigated by knowledge management (KM). While e-procurement is expected to increase the involvement of SMEs in public procurement, it is not well-understood how or which KM properties of an e-procurement service will best facilitate PPI. This paper presents an exploratory inquiry of how suppliers perceive the properties of an e-procurement platform, and how these properties are associated with their involvement in PPI. The results show that while information exchange properties were not estimated to be as important as other properties they are nevertheless more important in the sense of being positively associated with participating in PPI.
a University of Eastern Finland Business School. Address: P.O. Box 111 FI-81101 Finland; Tel. +358 50 442 3463; e-mail: jani.saastamoinen@uef.fi.
b University of Eastern Finland Business School. Address: P.O. Box 111 FI-81101 Finland; Tel. +358 50 442 3995; e-mail: timo.tammi@uef.fi.
* Corresponding author.
c University of Eastern Finland Business School. Address: P.O. Box 111 FI-81101 Finland; Tel. +358 50 435 2408; e-mail: helen.reijonen@uef.fi.
Keywords: Public procurement, public innovation procurement, electronic procurement, e- procurement, SMEs, innovations, knowledge management, suppliers, public sector, contracting authorities, information exchange
Biographical notes: This paper is a revised and expanded version of a paper entitled ‘The role of electronic procurement system in facilitating SMEs’ involvement in public innovation procurement’, presented at the 39th ISBE conference in Paris, France, in 2016.
Author biographies:
Jani Saastamoinen, Ph.D., is University Lecturer, Business School, University of Eastern Finland.
His research interests include public procurement, industrial organization and the analysis of financial decision making.
Helen Reijonen, Ph.D., is a University Lecturer, Business School, University of Eastern Finland. Her research interests include strategic orientations, public procurement and small firm and
entrepreneurial marketing.
Timo Tammi, Ph.D., is University Lecturer, Business School, University of Eastern Finland Business School. His teaching and research interests are in behavioral economics, public procurement and decision-making.
Acknowledgement: This study was financed by the Finnish Funding Agency for Innovation and Technology (TEKES) research grant #40055/14.
2
1. Introduction
Securing value for money is the main goal of public procurement (McKevitt and Davis, 2016), and consequently, the European Union (EU) has introduced three policies to support this goal. First, improving the access of small and medium-sized enterprises (SMEs) to public sector tenders would increase competition in the supplier market and support the single European market for goods and services. This, in turn, increases value for money in public sector purchases and contributes to job creation, economic growth and innovation (European Union, 2008). For these reasons, the EU’s new procurement directives seek to reduce the need for extensive documentation in the tendering process, and this has the potential to increase SMEs’ involvement in public procurement (European Commission, 2016a). Second, public procurement of innovative solutions (PPI) may occur when a public organization wishes to purchase something which does not exist yet, but which can be developed with minimal R&D (Edquist and Zabala-Iturriagagoitia, 2012). In this way, PPI can be seen as a means to achieve better value for money (Arlbjørn and Freytag, 2012). PPI has been highlighted as a mechanism which boosts supplier innovation in the community (Rolfstam, 2009), and SMEs are perceived as the drivers of innovation in this setting (European Commission, 2016a). Third, electronic procurement (e-procurement) is being phased in as a mandatory procurement mechanism for all types of procurement by 2018 and this initiative is expected to simplify the entire procurement cycle for all parties involved (European Commission, 2016b).
These policies are a response to the fact that public procurement, and PPI in particular, remain an untapped source of business for SMEs (GHK, 2010; Nicholas and Fruhmann, 2014; PwC, 2014), because SMEs face several barriers in tendering with the public sector (e.g. Saastamoinen et al., 2017).
Interestingly, many barriers appear to be related to information and communication or, more precisely, a lack thereof. For instance, barriers identified in the literature include a lack of awareness of procurement opportunities (Loader, 2005), a lack of communication in the form of feedback (Loader, 2015), a lack of electronic purchasing systems (Karjalainen and Kemppainen, 2008), and bureaucratic procurement processes (Peck and Cabras, 2011). Furthermore, SME suppliers have criticized the contracting authorities’ lack of market knowledge (Loader, 2015), which may hinder harnessing the innovation potential of SMEs in public procurement. With regard to PPI, the barriers identified are related to the procurement process, such as the aforementioned lack of feedback and communication (Uyarra et al., 2014).
It could also be argued that the arms-length nature of public procurement limits interaction between procurement contracting authorities and suppliers leading to information and communication related problems. This is a cause for concern because the emergence of innovations requires interactive
3
learning between organizations, which is “to a large extent the same as communication and cooperation outside the market” (Edquist and Zabala-Iturriagagoitia, 2012, p. 1757). Thus, PPI requires more interaction between the procurer and suppliers than regular procurement because “information about procurers’ unmet needs must be communicated to potential suppliers” (Valovirta, 2015, p. 72).
However, the strategic use of e-procurement, whether using it for collaboration or supplier management, has not manifested in practice (McCue and Roman, 2012). There is even evidence that e-procurement may hinder purchasing from SMEs (Walker and Brammer, 2012). Therefore, how communication between the buyer and suppliers is established might prove essential in harnessing SMEs’ innovation potential in the domain of public procurement. For instance, it is not clear whether procurers are aware of innovative solutions that SMEs could provide, and whether SME suppliers are aware of the public sector’s demand for their solutions. Yet, e-procurement has the capacity to transform bureaucratic ways of contracting into more effective forms of organization in which communication is informal and jobs become cross-functional and enriched with content (Gardenal, 2013).
However, information and communication barriers could be removed, or at least lowered, with efficient and effective knowledge management (KM). The use of different information and communication technology (ICT) could facilitate effective KM and make its benefits more accessible for SMEs. However, SMEs tend to be informal in their KM (Nunes et al., 2006), although studies have shown that KM is often associated with achieving competitive advantage (López-Nicolás and Meroño- Cerdán, 2011) and greater innovation (Nunes et al., 2006). As a result, e-procurement services, which allow users to manage and retrieve information related to procurement contracts and communicate with procurers, as a means of KM may enhance SMEs’ participation and opportunities for success in regular and innovation procurement. Electronic procurement has been regarded as means to streamline the procurement process (Panayiotou, 2004), which may reduce bureaucracy related to public sector tendering. In addition, the sort of ICT required in e-procurement provides more intense inter- organizational communication channels, which in turn improve SMEs’ innovation capability (Corso et al., 2001; Lin 2007). It has also been suggested that SMEs which are active in public procurement tend to be more entrepreneurial, or more specifically, they are often more innovative, proactive (Reijonen et al., 2016) and market-oriented than their less active peers (Tammi et al., 2014). Thus, these firms are more likely to be a source of innovation because they are more focused on collecting new information, interpreting market signals and bringing innovative solutions to the market.
Consequently, the proliferation of e-procurement systems has the potential to improve the market access of SMEs and at the same time improve the communication between the supplier market and the procurer, which may also lead to further supplier innovations. Indeed, the procurer’s market knowledge
4
has been regarded as an essential ingredient for innovation-friendly public procurement (Knutsson and Thomasson, 2014).
However, very little is known about how suppliers perceive the importance of the KM properties of e- procurement services in facilitating SME innovations in public procurement. We address this gap in the literature by providing an exploratory study of the KM properties of an e-procurement service which are perceived as being important in a systematic manner. We further examine how these elements are associated with the SMEs’ involvement in PPI. We use survey data from a randomly selected sample of registered users of the most widely-used electronic public procurement service in Finland. The survey instrument explores users’ perceptions of the e-procurement service and the respondent’s involvement in public procurement including innovation procurement. The main contribution of this paper is to show that while suppliers perceive the properties which make the service easy to use and lower administrative costs as being more important than the properties that improve information exchange, the latter properties appear to be associated with being involved in PPI. These findings imply that while suppliers may not regard inter-organizational information exchange and communication as important as some of the more ‘practical’ aspects of e-procurement, their development should not be disregarded because the current policy emphasizes encouraging supplier innovations in public procurement.
The paper proceeds as follows. Section 2 provides a short literature review and lays out the research hypotheses. In section 3 we describe the data and present the results of the data-analysis. Section 4 presents the results. Section 5 discusses the research results and concludes the study by suggesting managerial and policy implications.
2. Review of relevant literature and research hypotheses 2.1. Knowledge management, ICT tools and innovation
Knowledge management refers to creating and mobilizing knowledge for certain purposes, such as for creating a competitive advantage (Quintas et al., 1997). In today’s markets, which are often characterized as rapidly changing and intensely competitive, firms that create new knowledge efficiently and effectively are in a better position to achieve a competitive advantage (López-Nicolás and Meroño-Cerdán, 2011) by being better able to meet their customers’ changing needs, for example.
Du Plessis (2007) defines KM as “a planned, structured approach to managing the creation, sharing, harvesting and leveraging of knowledge as an organizational asset, to enhance a company’s ability, speed and effectiveness in delivering products or services for the benefit of clients”. This implies that KM is linked to innovation (see also Darroch and McNaughton, 2002).
5
Innovation initiatives often depend on a person’s knowledge, skill and experience in the value creation process (Wang and Wang, 2012), and innovations are typically based on knowledge that has long been known but not applied to the present situation (Quintas et al., 1997). Consequently, the more efficient and effective use of knowledge and information could lead to improved innovation performance. Prior studies have shown that practices of sharing explicit and tacit knowledge facilitate innovation and performance – explicit knowledge contributes to innovation speed and financial performance, while tacit knowledge contributes to innovation quality and operational performance (Wang and Wang, 2012). Furthermore, there are findings indicating that KM strategies which refer to codification and personalization affect innovation and organizational performance directly and indirectly through an improvement in innovation capability (López-Nicolás and Meroño- Cerdán, 2011). Thus, it seems KM plays several key roles in innovation: first, it enables sharing and codification of tacit knowledge;
second, it enhances the recombination of existing knowledge in new ways and third, it enables collaboration (du Plessis, 2007).
At the same time, it has been recognized that SMEs tend to be informal in their knowledge management, although they acknowledge that capturing, storing, sharing and disseminating knowledge can lead e.g. to greater innovation (Nunes et al., 2006). The use of different information and communication technologies could help firms to efficiently manage knowledge flows within and between organizations (Gressgård et al., 2014). ICT addresses the information needs of firms by facilitating their decision making and enhancing knowledge acquisition and transfer (García-Álvarez, 2015).
Prior research has identified several factors that can affect the use of technological tools and the effectiveness of this use on innovation activities. For example, Jarvenpaa and Staples (2000) argue that the use of e-systems for communicating and sharing information is strongly related to beliefs as to whether the system can provide valuable information in an effective manner. Gressgård et al. (2014) point further that the use of ICT-tools can improve the efficiency of knowledge processes, but it requires that careful attention is paid to accessibility and the reliability of the sources of knowledge.
On a more general level, Kim and Lee (2006) state that both the usage of IT applications and the user- friendliness of IT systems enhance employees’ knowledge sharing capabilities and that this also includes the acquisition of knowledge (Kim & Lee 2006). Finally, it is important to note that while KM systems do not alone have the qualities needed to create competitive advantage they contribute to it when bundled with other company resources and core competencies (du Plessis, 2007).
Consequently, the introduction of ICT can result in both knowledge creation and innovation processes (García-Álvarez, 2015).
6
Prior research has also examined collaborative technologies which are used to provide information online to communicate and exchange information, as well as to automate business processes (Merono- Cerdán et al., 2010). Thus, they enable activities such as accessing, searching, sharing, storing and publishing information in a computer network internally within an organization, as well as external to it (Jarvenpaa and Staples, 2010). Based on the prior literature, Merono-Cerdán et al. (2010) identify that these collaborative technologies support innovation processes by enhancing the efficient storage and retrieval of codified knowledge, and by bringing people together to innovate, as well as by enabling the formation of virtual teams and favourable climates for innovation. In their study, they found further that document management systems for storing information and providing easy access to it with workflows that include a degree of process automation (e.g. passing documents, information or tasks from an actor to another for action) were found to be the key features of collaborative technologies that contributed the most to innovation in SMEs (Merono-Cerdán et al., 2010).
2.2. PPI and e-procurement
Over the recent years, many steps have been taken to facilitate e-procurement in public sector tendering as public procurers have been early adopters of e-procurement (Rao and Tummala, 2007). As part of a wider thrust towards electronic government, significant investments have been made in information communication technology (ICT) in order to digitalize public procurement (McCue and Roman, 2012).
However, e-procurement is more than digitizing different stages of the procurement process as implementing it requires “new technological, organizational and legal competencies for dealing with the procurement process, managing information and interacting with economic operators and e- procurement platforms (Soares and Carvalho, 2017, p. 364). Here we adopt the following definition by Croom and Brandon-Jones (2007, 295; for a recent account of definitions, see Fernandes and Vieira 2015): “Electronic procurement refers to the use of integrated (commonly web-based) communication systems for the conduct of part or all of the purchasing process; a process that may incorporate stages from the initial need identification by users, through search, sourcing, negotiation, ordering, receipt and post-purchase review.”
The current policy is supportive towards e-procurement; in the European Union, e-submission is intended to be mandatory for all procurement procedures and all contracting authorities by October 2018. The European Commission characterizes the ‘reform’ as beneficial due to savings of resources, increased transparency, having a positive influence on innovation, and easier ability for SMEs and other firms to participate in public procurement (European Commission, 2016c). This reform is involved with two other targets concerning public procurement in the EU – namely, improving SMEs’
access to procurement by lowering the extra administrative burdens of participating in public
7
procurement and encouraging public sector authorities to use and develop procedures for carrying out public procurement for innovations. The goal of PPI projects is to be mutually beneficial to the parties involved: the public sector receives innovative solutions and value for money, while private sector suppliers are presented with a market opportunity (Arlbjørn and Freytag, 2012). Since public procurement for innovations departs in many relevant aspects from a regular public procurement, there is a need to study the special factors related to e-procurement in the context of public innovation procurement.
Public procurement for innovation is “a public agency engaged together with one or several private firms or other organizations in activities that may lead to or promote innovation of some kind”
(Rolfstam et al., 2005). Edquist and Zabala-Iturriagagoitia (2012) argue that “regular procurement has nothing to do with innovation” because it is not a policy instrument. It must be noted; however, that the purchasing power of the public sector can be used to enlarge markets, and this creates incentives for suppliers to come up with innovations (Uyarra and Flanagan, 2010). Compared to regular public procurement, PPI changes the nature of public procurement considerably creating an arena of interaction between the public sector and potential and actual private sector suppliers. PPI involves complexities related to the nature of innovation, information sharing, co-operation and the manifestations of societal problems and the needs of the end users (Rolfstam et al., 2005; Edquist and Zabala-Iturriagagoitia, 2012). Edler and Georghiou (2007) contend that there are characteristics in PPI which cannot be entirely understood from the traditional market failure (mostly concerning information asymmetries) approach nor from the somewhat more recent system failure (inoperative interaction) approach. In addition, attention has been paid to different forms of innovation. In general, incremental innovations are less complex and not as risky as radical innovations because radical innovations demand more effort in searching and analysing customer and competitor information than incremental innovations do. As noted by Lember et al. (2014), one form of incremental innovation is the case when an existing product or service is introduced into a new region or country.
From this perspective, public procurement practices in general and e-procurement systems in particular have a high degree of influence on the success or failure of innovation procurement. Previous research, though dealing mainly with e-procurement in general and thus being focused on regular forms of public procurement, has identified the most important benefits and challenges of e-procurement. On the positive side there are cost reduction, time savings, less paperwork and bureaucracy, transparency, and access to information, for example (Engström et al., 2009; Costa et al., 2013). As such, e-procurement can be used to achieve perquisite conditions for competition, namely access to information, lower barriers to entry and an increased number of competitors in the market (Carayannis and Popescu,
8
2005). The negative aspects, respectively, consist of challenges related to issues such as adequate planning, time considerations, communication issues, data quality and technical issues (ibid.).
Fernandes and Vieira (2015) sum up the recent research into the following ‘public procurement adoption drivers’: cost reductions, contracting time reduction, increased transparency (and the sharing of information), increased competitiveness, improvements in proposals, and higher information systems integration. In addition, better access to procurement contracts through e-procurement, which increases competition among suppliers, may provide more or fewer incentives to innovate depending on the market structure of a given industry as suggested by Aghion et al. (2005). Respectively, the main ‘public procurement adoption inhibitors’ are related to the security of the information exchanged, variation in procurement procedures and between information systems, slowness of revising organizational processes, resistance to change and technological progress, as well as the transformation of new legal frameworks into established interpretations and practices (Fernandes and Vieira, 2015).
Which questions, then, appear to be the most critical concerning procurement practices in the context of public procurement of innovations? Edler et al. (2015) argue on the basis of a survey of 800 private firms and third sector actors supplying the public sector in the UK, however, that not all procurement practices related to the efficiency of the process are perceived as encouraging innovation. Although qualities such as open competitive tendering, framework agreements and electronic submissions of tenders were perceived as encouraging innovations, profit-sharing arrangements in incentive contracts, private finance initiatives and e-auctions were not. Instead, respondents preferred pre-procurement communication and early interaction with the procuring entity, outcome-based specifications and an emphasis on sustainability criteria. Indeed, Georghiou et al. (2014) point that the communication of needs between buyers and suppliers is “at the core of innovation procurement policy.” Further, a systemic approach to innovation would emphasise user-producer interaction in the production of innovations (Edler and Uyarra, 2013). Consequently, enabling the exchange of buyer-supplier information could be critical because Lewis and Roehrich (2009) argue that both the buyer and the supplier benefit from greater interaction in the procurement of complex performance, which is often a characteristic of innovation procurement.
Based on the above discussion concerning knowledge management, public procurement of innovation and e-procurement, we aim to find out how an e-procurement system can serve SMEs as an effective KM tool that enables their participation in public innovation procurement. Thus, we examine 1) which properties of the e-procurement system suppliers perceive as important and 2) how these properties are associated with their involvement in public innovation procurement.
9
3. Data and methods 3.1 Survey design
This study uses survey data collected from Finnish firms. The survey instrument, which is reported in Table 1 (see also Footnote 2), was developed using Skok et al. (2001) as a baseline item battery. This battery was then amended with statements designed to measure properties of an e-procurement service.
To this end, the survey instrument was carefully validated by using developers of a Finnish e- procurement service as experts in assessing how well the item battery reflected the key issues in e- procurement. In addition, the survey instrument included questions about the respondents’
backgrounds (e.g. firm size measured in the number of employees, firm age, involvement in public innovation procurement, experience of public procurement and electronic procurement).
An electronic survey questionnaire was sent to a randomly selected sample of firms which were registered users of the most widely-used electronic public procurement service. The service provider’s market share in Finland is almost 90 % comprising major municipalities and contracting entities as its customers including those responsible for tenders of the national government and the federation of municipalities. This ensures that most public tenders are carried out through this particular e- procurement service. Consequently, the sample provides a reliable cross-section of firms that are involved in public procurement in Finland. 253 responses were received covering the properties of the e-procurement system part of the survey. The responses were analysed using principal component factor analysis (PCFA), t-tests and logistic regression.
3.2. Constructs for perceptions of e-procurement service
The statistical analysis of the data proceeded as follows. We ran a PCFA on a set of items which measured the respondents’ perceptions of the importance of the properties of the electronic procurement service. The factor solution using Varimax rotation is reported in Table 1. Four items were dropped from the final solution1. The PCFA produced four principal component factors from which the Bartlett scores were used as the variables of interest (the variable names are in parentheses in Table 1) in a logistic regression analysis.
The principal component factors were labelled as: Data management referring to improvements in the ability to manage procurement data, Ease of use referring to the general user experience of the e- procurement service, Administrative burden referring to improvements that lowered administrative costs pertaining to the procurement process and Information exchange referring to improvements in
1 The following items were dropped: “There is enough practical training/guidance for the use of the service.”, “The service increases productivity.”, “The service makes working faster.” and “The service is helpful in creating a more positive attitude towards invitations to tender”.
10
the facilitation of information exchange among procurers and suppliers. The Cronbach’s alphas of the factors ranged from 0,879 to 0,957 which implies that the constructs have a very good degree of internal consistency.
Table 1. Factor solution for perceptions of e-procurement service.
Information exchange (alpha = 0.957, lambda = 15.878) [INFO_EXC]
Construct average score = 3.422. Average
score Factor loading The service is helpful in submitting more competitive bids. 3.834 0.561
The service is helpful in innovation procurement. 3.565 0.724
The service is helpful in the development of the buyer-seller-relationship 3.518 0.759 The service is helpful in finding new business opportunities. 3.711 0.667 The service is helpful in establishing contacts between different actors. 3.431 0.723 The service is helpful in establishing contacts between firms. 3.308 0.810
The service is helpful in establishing networks. 3.225 0.857
The service is helpful in our organization’s strategic discussion regarding the direction of our
innovations. 3.158 0.857
The service is helpful in bringing forth ideas for the needs of the public sector. 3.265 0.863 The service is helpful in establishing a joint view shared by firms and public sector customers
regarding a product or service procurement requiring an innovative solution.
3.261 0.868 The service is helpful in establishing negotiations regarding e.g. quality standards and
customization opportunities. 3.368 0.753
Ease of use (alpha = 0.944, lambda = 3.814) [EASE_USE]
Construct average score = 4.049.
Average score
Factor loading
The service support responds quickly to emerging problems. 3.818 0.667
The service is easy to use. 4.174 0.822
The user interface (layout) of the service is clear. 4.130 0.822
The service is consistent to use. 4.166 0.837
The language used is clear and easy to understand. 4.245 0.763
User input is taken into account in the design/development of the service. 3.802 0.679
Users trust the service. 4.115 0.752
Users have a clear understanding about the service and its capabilities. 3.941 0.727 Administrative burden (alpha = 0.934, lambda = 1.622) [ADM_BURDEN]
Construct average score = 3.967. Average
score Factor loading The service makes finding and processing information and providing answers to questions
faster.
3.957 0.510
The service lowers administrative costs. 3.822 0.615
The service reduces bureaucracy. 3.921 0.661
The service enhances participating in invitations to tender. 4.004 0.683
The service is cost efficient. 3.937 0.658
The service is helpful in observing the requirements of the procurement law and directives 4.047 0.738 The service is helpful in observing required procurement procedures. 4.083 0.726 Data management (alpha = 0.879, lambda = 1.287) [DATA_MGMT]
Construct average score = 3.897. Average
score Factor loading
The service reduces mistakes which would occur otherwise. 3.957 0.786
The service provides useful information for decision making. 3.818 0.796
The information provided by the service is up-to-date. 4.087 0.777
The service assists in tracking the fulfilment of the procurement contract in such a manner that
it provides information about e.g. orders, deliveries and possible reclamations. 3.814 0.731 The service saves information about the fulfilment of past procurement contracts, and this
database can be utilized in future procurements. 3.889 0.594
Notes: Likelihood ratio test: χ2(465) = 7816.57 (p-value < 0.0001); Kaiser-Meyer-Olkin: KMO = 0.947. Rotation:
Varimax. Survey instrument answer scale (1 to 5): 1 = No importance … 5 = Very important.
3.3 Variables used in the analyses
11
The variables used in the statistical analyses were based on the self-reported answers to the survey questionnaire. A brief description of these variables can be found in Table 2. The dependent variables were binary variables for different types of supplier innovations originating from the demand of public- sector customers. They take the value of 1 if a supplier reports having developed an innovative product or service for a public-sector customer, and 0 otherwise. A similar operationalization was adopted in Aschhoff and Sofka (2009) and Uyarra et al. (2014). There were four dependent variables which were:
developing a new-to-the-market product/service for the public sector customer (NEW_PROD), a significantly improved (IMP_PROD) product/service for the public sector customer, an improved production process (IMP_PROC) for the public sector customer, or a new/significantly improved product/service for the private sector customer based on an innovation developed for the public sector customer (NEW_PVT), which is an indicator of further commercialization of a product/service developed in the PPI process.
The independent variables in the logistic regression analyses can be divided into focus variables and control variables. The focus variables are INFO_EXC, EASE_USE, ADM_BURDEN and DATA_MGMT which are the normalized Bartlett scores obtained from the PCFA. Their operationalization has been discussed in Section 3.1. We use several control variables based on the research into public procurement. First, it has been shown that firm size (SIZE), measured by the number of employees, is an important predictor of a firm’s success and participation in public procurement (e.g. Pickernell et al., 2011; Flynn et al., 2015; Reijonen et al., 2016). Second, recent studies also show that firm age is positively associated with the likelihood of being involved in public procurement (e.g. Pickernell et al., 2013; Reijonen et al., 2016). Therefore, we controlled for firm age (AGE) which is measured in the number of years from incorporation. Third, being a supplier to the public sector may require specific skills, which may be unobservable, and firms may exhibit learning through experience from public procurement. For instance, Flynn and Davis (2016) found that suppliers with sufficient procedural capability, which “embodies a firm’s ability to deal with the administrative and technical demands of the tendering process (p. 7)”, were more likely to succeed in securing public sector contracts. As a consequence, we used the number of public sector contracts won over the past five years (CONTRACTS) as a proxy for their potential influence. Fourth, having experience from e-procurement may be associated with being a supplier to the public sector (Karjalainen and Kemppainen, 2008). For this reason, we control for e-procurement experience (E- PROC_EXP) which is the number of years of the reported use of e-procurement. Finally, it was important to control for industry differences because the importance of the public-sector demand varies across industries (Edquist and Zabala-Iturriagagoitia, 2012). Thus, we included control variables for
12
broad industry classifications, which were: industrial production and manufacturing (INDUST), construction (CONST), wholesale and retail trade (TRADE), knowledge- and information-based services (KIBS), healthcare (HEALTH) with other (OTHER) for industries other than those identified earlier.
4. Results
4.1 Descriptive statistics
The descriptive statistics of the variables used in the regression analysis are shown in Table 2. The responses (Panel A) indicated that an improved production process is the most common (40 %) outcome of public procurement innovations, whereas the least common outcome was a product/service that could be transferred to the private sector (19 %). However, both new-to-the-market products (29
%) and significantly improved products (39 %) were also frequent outcomes.
The control variables are reported in Panel C. While the average firm had 677 employees, most of the responding firms were SMEs as firms with more than 250 employees, which are large firms by this size measure, accounted for 13 % of the observations. The average firm was incorporated 25 years ago. The use of e-procurement is a relatively recent phenomenon and this can be seen in the responses indicating that the average length of experience of e-procurement was only 4 years. However, the average number of public sector contracts won over the past five years was 55, which indicates that the firms in the sample were active in public procurement.
The Pearson correlations matrix of the variables used in the statistical analysis are reported in Table 3. It is evident that the variables measuring different types of innovations are highly correlated with each other suggesting that firms engaging in one type of innovation also produced other innovation types. Regarding independent variables, the variables controlling for experience E-PROC_EXP and CONTRACTS correlate with the dependent variables to a degree. However, correlations between independent variables seem to be fairly low which suggests that multicollinearity is not an issue in regression analysis. Notable exceptions are firm size and age and the measures of experience, which are highly correlated with each other.
13
Table 2. Variables used in the statistical analyses and their descriptive statistics.
Panel A: Dependent variables
Variable Description Obs
. Mean S.D. Min. Max.
NEW_PROD Takes the value 1 if a firm has developed a new-to-the market product/service for
the public-sector customer.
253 0.289 0.454 0.000 1.000
IMP_PROD Takes the value 1 if a firm has developed a significantly improved product/service
for the public-sector customer.
253 0.391 0.489 0.000 1.000
IMP_PROC Takes the value 1 if a firm has developed a significantly improved production process for the public-sector customer.
253 0.399 0.491 0.000 1.000
NEW_PVT Takes the value 1 if a firm has developed a new or significantly improved product/service), which is based on an
innovation developed for the public- sector customer, for the private sector
customer.
253 0.190 0.393 0.000 1.000
Panel B: Focus variables
Variable Description Obs
. Mean S.D. Min. Max.
INFO_EXC A Bartlett score from PCFAC. 253 0.000 1.001 -2.813 1.983 EASE_USE A Bartlett score from PCFAC. 253 0.000 1.007 -3.823 2.359 ADM_BURDEN A Bartlett score from PCFA C. 253 0.000 1.003 -4.500 2.521 DATA_MGMT A Bartlett score from PCFA C. 253 0.000 1.005 -4.737 2.569 INFO_EXC A Bartlett score from PCFAC. 253 0.000 1.001 -2.813 1.983
Panel C: Control variables
Variable Description Obs
. Mean S.D. Min. Max.
INDUST Takes the value 1 if a firm operates in the
industrial goods/services sector. 253 0.067 0.251 0.000 1.000 CONST Takes the value 1 if a firm operates in the
construction sector. 253 0.087 0.282 0.000 1.000 TRADE Takes the value 1 if a firm operates in the
retail and wholesale trade sector. 253 0.178 0.383 0.000 1.000 KIBS Takes the value 1 if a firm operates in the
knowledge- and information-based sector. 253 0.087 0.282 0.000 1.000 HEALTH Takes the value 1 if a firm operates in the
healthcare and social services sector. 253 0.178 0.383 0.000 1.000 OTHER Takes the value 1 if a firm operates in
another sector than those mentioned above.
253 0.166 0.373 0.000 1.000
AGE Firm age measured in years since incorporation.
253 25.474 27.981 0.000 156.000 SIZE Firm size measured in the number of
employees. 253 676.986 6386.940 0.000 99999.000 E-PROC_EXP Experience from using e-procurement
measured in years.
245 3.873 3.055 0.000 19.000 CONTRACTS The number of public sector contracts
secured over the past five years. 244 55.439 513.906 0.000 8000.000 Notes: C Please, see Table 1 for further details on the construct measurement and properties.
14
Table 3. Pearson correlations matrix.
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17)
NEW_PROD (1) 1.00
IMP_PROD (2) 0.69 1.00
IMP_PROC (3) 0.47 0.53 1.00
NEW_PVT (4) 0.68 0.53 0.31 1.00
INFO_EXC (5) 0.13 0.09 -0.02 0.10 1.00
EASE_USE (6) 0.04 0.09 0.04 -0.07 0.01 1.00
ADM_BURDEN (7) 0.09 0.10 0.05 0.02 0.01 0.00 1.00
DATA_MGMT (8) 0.04 0.04 0.02 0.06 0.01 -0.02 0.00 1.00
AGE (9) 0.03 0.02 0.04 0.01 -0.02 0.05 0.03 0.08 1.00
SIZE (10) 0.11 0.08 0.07 -0.01 -0.06 -0.04 0.11 0.01 0.16 1.00
E-PROC_EXPE (11) 0.14 0.26 0.24 0.20 -0.05 0.00 0.05 -0.11 0.20 -0.02 1.00
CONTRACTS (12) -0.03 -0.05 0.09 -0.02 -0.11 0.02 0.03 0.03 0.00 0.01 0.02 1.00
INDUST (13) -0.02 0.00 0.03 0.01 -0.01 -0.10 0.10 0.08 0.20 -0.02 0.04 -0.01 1.00
CONST (14) -0.08 -0.02 0.06 -0.04 -0.12 0.01 -0.09 -0.07 -0.05 -0.02 0.07 0.21 -0.08 1.00
TRADE (15) 0.04 0.05 -0.11 0.06 0.03 -0.03 -0.12 -0.02 0.14 -0.04 0.18 -0.02 -0.12 -0.15 1.00
KIBS (16) 0.16 0.17 0.14 0.11 0.04 -0.01 0.08 -0.05 -0.07 0.22 -0.03 -0.02 -0.08 -0.10 -0.14 1.00
HEALTH (17) -0.10 -0.12 -0.15 -0.14 -0.08 0.06 0.11 0.16 -0.10 -0.03 -0.18 -0.04 -0.12 -0.15 -0.21 -0.14 1.00
OTHER (18) -0.01 -0.03 0.10 0.08 0.04 0.01 0.01 0.03 -0.08 -0.01 0.04 -0.02 -0.12 -0.14 -0.21 -0.14 -0.20
15
4.2 The importance of e-procurement service properties
The importance of the properties of an e-procurement service were assessed using average scores from the survey instrument (see Table 1). Table 4 reports the average scores for each construct measured on a Likert scale ranging from 1 = No importance to 5 = Very important. We also carried out t-tests across construct pairs to determine which of them the respondents perceived to be of higher importance.
Table 4. Average scores for constructs and t-tests for construct pairs.
Pair Means Difference T-value
INFO_EXC vs.
EASE_USE 3.422 – 4.049
(0.073) (0.062) -0.627
(0.096) -6.560***
INFO_EXC vs.
ADM_BURDEN 3.422 – 3.967
(0.073) (0.064) -0.545
(0.097) -5.612***
INFO_EXC vs
DATA_MGMT 3.422 – 3.897
(0.073) (0.065) -0.475
(0.097) -4.875***
EASE_USE vs
ADM_BURDEN 4.049 – 3.967
(0.062) (0.064) 0.082
(0.089) 0.914
EASE_USE vs DATA_MGMT
4.049 – 3.897 (0.062) (0.065)
0.152 (0.090)
1.692*
ADM_BURD vs
DATA MGMT 3.967 – 3.897
(0.064) (0.065) 0.070
(0.091) 0.767
Notes: Standard errors in parentheses. Statistical significance (two-tailed): *** p-value <
0.01; * p-value < 0.01;
Judging by the raw average scores, the respondents perceived INFO_EXC (3.422) to be the least important and EASE_USE (4.049) to be the most important property of an e-procurement service. The ADM_BURDEN (3.967) and DATA_MGMT (3.897) items were perceived to be slightly less important properties than the ease of use. T-tests indicated that the respondents did not value INFO_EXC as much as other three properties (p < 0.01 in each case). There was also a marginally statistically significant (p < 0.1) difference between the EASE_USE and DATA_MGMT items in favour of the former. Other pairs, however, were not statistically significant. Based on these results, it appears that the respondents placed the highest value on those properties which make an e-procurement service easy to use and reduce the administrative burden. By contrast, the properties that improve information exchange in the context of public procurement were perceived to be less valuable to the respondents.
4.3 The importance of e-procurement service properties in public innovation procurement The results of the logistic regression analysis are reported in Table 5. Logarithmic transformations were used for the continuous independent variables. There are four models labelled “New-to-the- market products/services”, “Significantly improved products/services”, “Significantly improved production process” and “New/significantly improved products/services for the private sector” with NEW_PROD, IMP_PROD, IMP_PROC and NEW_PRI as the dependent variables for these models, respectively. All models were estimated with the same set of control variables described in the previous
16
section. The variance inflation factors (VIFs) ranged from 1.03 to 1.85 with an average of 1.29. This suggests that multicollinearity is unlikely to be problematic in the regression analyses.
The results show that only INFO_EXC was associated with the PPI in the cases of “New-to-the-market products/services” (p < 0.05), “Significantly improved products/services” (p < 0.1) and
“New/significantly improved products/services for the private sector” (p < 0.1). The results show that respondents who perceived high value in the information exchange facilitation properties of the e- procurement service were more likely to engage in producing innovative products/services to public sector customers, and more likely to transfer these to private sector customers. The other variables of interest were not statistically significant.
Several interesting insights can be inferred from the control variables. First, both CONTRACTS and SIZE are positively associated with new-to-the-market innovations and the further commercialization of innovations developed for the public-sector customer. This suggests that a supplier’s experience and procedural capability play a role in innovation procurement but are not crucial factors in being involved in innovation procurement. Interestingly, a limited association between firm size and innovation type implies that SMEs do not appear to be entirely disadvantaged in PPI. Second, firm age is statistically significant but negative in the cases of new-to-the-market products/services and new/significantly improved products/services for the private sector, which hints at start-ups or young firms being more active in innovation procurement which seeks to introduce more ‘radical’ innovations to the market.
Finally, the experience of e-procurement was positively associated with all innovation types, which may be a reflection of innovation capability (Pickernell et al., 2011).
17
Table 5. Logistic regression models.
Model “New-to-the-market
products/services”
“Significantly improved products/services”
“Significantly improved production process”
“New/significantly improved products/services for the private
sector”
D.V. NEW_PROD IMP_PROD IMP_PROC NEW_PRI
Coef. S.E. p-value Coef. S.E. p-value Coef. S.E. p-value Coef. S.E. p-value
INFO_EXC 0.350** 0.163 0.032 0.273* 0.150 0.069 0.020 0.148 0.893 0.346* 0.199 0.082
EASE_USE 0.067 0.167 0.690 0.204 0.152 0.178 0.057 0.144 0.692 -0.266 0.190 0.162
ADM_BURDEN 0.135 0.167 0.419 0.153 0.150 0.310 0.025 0.143 0.863 -0.073 0.192 0.705
DATA_MGMT 0.139 0.160 0.385 0.189 0.150 0.207 0.139 0.147 0.341 0.248 0.187 0.186
Ln(AGE) -0.446** 0.196 0.023 -0.282 0.181 0.119 -0.195 0.179 0.275 -0.680*** 0.239 0.004
Ln(SIZE) 0.172* 0.096 0.073 0.015 0.092 0.872 0.072 0.092 0.430 0.185* 0.106 0.081
Ln(E-PROC_EXP) 0.564* 0.291 0.053 1.043*** 0.288 0.000 0.877*** 0.274 0.001 1.041*** 0.364 0.004
Ln(CONTRACTS) 0.260** 0.119 0.029 0.150 0.114 0.186 0.148 0.115 0.195 0.239* 0.133 0.073
INDUST -0.114 0.677 0.866 0.130 0.631 0.837 0.066 0.615 0.915 0.067 0.786 0.932
CONST -1.001 0.666 0.133 -0.123 0.544 0.822 0.031 0.514 0.952 -0.922 0.759 0.225
TRADE 0.194 0.428 0.650 0.237 0.409 0.563 -0.883** 0.427 0.039 0.232 0.477 0.627
KIBS 0.910* 0.527 0.084 1.352** 0.553 0.014 0.802 0.527 0.128 0.714 0.583 0.221
HEALTH -0.480 0.492 0.329 -0.411 0.447 0.358 -0.812* 0.450 0.071 -1.097 0.693 0.113
CONSTANT -1.444*** 0.546 0.008 -1.565*** 0.525 0.003 -1.393*** 0.510 0.006 -2.135*** 0.662 0.001
Obs. 237 237 237 237
LR Chi2 34.76*** 0.001 39.03*** 0.000 34.43*** 0.001 34.07*** 0.001
Pseudo R2 0.121 0.123 0.108 0.148
Notes: OTHER used as the reference category. Statistical significance *** p-value < 0.01; ** p-value < 0.05; * p-value < 0.1. Coef.: estimated coefficient; D.V.: dependent variable; S.E.: standard error.
18
5. CONCLUSION 5.1 Discussion
This paper provided an exploratory study of which properties of the e-procurement system SMEs regard as important and how the properties are associated with the SMEs’ involvement in the public procurement of innovations. Our findings suggest that suppliers place most value on the properties which make the e-service easy to use and lower administrative costs, whereas the information exchange properties, in general, are not evaluated as important. Kim and Lee (2006) found that the user-friendliness of IT systems was significant in affecting employees’ knowledge-sharing capabilities both in private and public organizations. Thus, from the KM perspective, ease of use could be considered a basic property requirement which affects the overall use of the e-system and also its effectiveness and efficiency as a KM-tool. Consequently, those properties that lessen the work burden, which is a frequent concern in public procurement (e.g. Loader, 2013), are highly valued by SMEs.
Our findings also suggest that the information exchange function is important when innovations produced/demanded belong to the ‘radical side’ of the innovation spectrum rather than improving production processes of the existing offerings. This is, perhaps, understandable in terms of the differences in the very nature of the varying forms of innovations: because developing products or services which are new to the market or the customer require more communication between the supplier and the buyer compared to focusing on innovations in production processes which are more of a firm’s internal matter. These findings suggest that information exchange pertaining to innovations is important, or perhaps, “part of the strategic architecture of a firm and provides support to outcomes such as innovation” (Darroch and McNaughton, 2002, p. 219).
Our results qualify the earlier findings of the role and significance of e-procurement in regular public procurement and in public procurement for innovation. Thus, compared to regular public procurement where e-procurement has been found to be associated with better possibilities in data management, ease of use and lower administrative costs, the most important property of e-procurement in public procurement of innovations appears to be improved information exchange. This is obviously related to the nature of public procurement of innovations: it involves complexities which are associated with the nature of innovation, such as information sharing, co-operation and the manifestations of societal problems and the needs of the end users (Rolfstam et al., 2005; Edquist and Zabala-Iturriagagoitia, 2012). However, this result can be considered to qualify wider ramifications because the maximum value of procurement can be achieved through the “effective coordination of sourcing, purchasing or distribution” (Hong and Kwon, 2012), which, obviously, requires efficient means of communication.
19
5.2 Managerial and policy implications
The main result of this study can be addressed to the developers of e-procurement systems – viz. firms who work for improving e-procurement systems and public sector actors who specify their needs concerning e-procurement and who set and implement rules and conventions of both regular public procurement and innovation procurement. The most important managerial implication emanating from this study is that providers of e-procurement services should focus on how they facilitate information exchange between potential suppliers, the procurer and end-users because this improves interaction between suppliers and public sector customers. Because public procurement is characterized as bureaucratic, rigid and requiring constant renegotiation and proactive conflict management (Costa et al., 2013), this could improve the general functionality of the procurement process. Moreover, the ability to communicate and influence the buyer’s supply preferences has been shown to be focal in enhancing SMEs’ opportunities to succeed in invitations to tender (Flynn and Davis, 2016). Further, an in-advance ‘market-sounding’ process before an invitation tender is published has been considered to be an effective tool in refining the outcome of innovation procurement (Georghiou et al., 2014).
Therefore, improvements in these properties of the e-procurement system could increase the effectiveness of public innovation procurement because information flows between the market and the customer increase the awareness of the needs of the public-sector customer as well as of the solutions available in the market.
Enhancing the information sharing functions of e-procurement systems could also be instrumental in attracting SMEs to provide innovative products and services to local-level as opposed to national-level public sector customers who are more often their customer type in the public sector (Tammi et al., 2016). Further, as identifying the high-growth firms (‘gazelles’) from the rest of the SME population could be a sound policy (Nightingale and Coad, 2013), and improving e-procurement systems could also assist in this process. Thus, policy makers should be aware that the innovation potential of SMEs may be inadequately unleashed, especially at the local level, if e-procurement services cannot facilitate information exchange between suppliers and customers properly. Walker and Harland (2008) suggest that “e-procurement should be considered in the context of other policy objectives”, so improving the properties of e-procurement services could provide indirect support to these objectives. It could also be argued that these properties improve strategic collaboration opportunities between buyers and suppliers, which have been highlighted as critical building blocks in long-term business relationships (Sillanpää et al., 2015).
Prior studies suggest that SMEs have weaker relational capabilities than procedural capabilities with regard to public procurement (Flynn and Davis, 2016). To strengthen these relational capabilities,
20
efforts from both the supplier and the public procurer sides are needed. For example, SMEs could focus more on finding ways to promote themselves as trustworthy and skilled suppliers as well as creating demand for new products and services. Public procurers, on the other hand, could pay attention to opening new avenues for more regular and profound interaction. As suggested in prior literature, one way of enhancing relational capabilities is through training programmes and workshops (Flynn and Davis, 2016). Another efficient way could be the development of e-procurement systems. At its best, an e-procurement system could offer a platform that enhances open discussion and information exchange that is timely, readily available to all interested parties and free of time and place constraints.
This kind of platform promotes networking, not only between SMEs and the public buyer, but also among SMEs and third parties so that they can combine and complement resources in order to be able to better meet the public buyer preferences. This could also lead to more efficient development and commercialization of innovations. Finally, some kind of cultural change among public procurers is needed to realize the more active role of the public procurers in promoting the exchange of information in the ways described above. This suggestion echoes Loader’s (2013) notion that even though tough improvements in public procurement have been suggested, the development has been slow due to a lack of clear priorities and objectives and a culture which functions disadvantageously to SMEs.
Obviously, the situation is largely the same with improvements concerning the development of e- procurement.
5.3 Limitations and future research
As is the case with all studies, this paper has some limitations. First, the sample is from a single country.
Consequently, the conclusions reached in this paper may not be applicable to other countries because of the many differences in rules, cultures and other matters in how public procurement and innovation procurement are carried out. Second, our analysis is based on data gathered from the users of a single e-procurement system. Although the system in question is the most widely-used by the Finnish contracting authorities, another platform might have provided different results. Third, the data set may not be a representative sample from the firm population because a decision to register with an e- procurement system may be biased towards larger firms as suggested by the SMEs’ participation rate in public procurement (GHK, 2010; Nicholas and Fruhmann, 2014; PwC, 2014).
Based on the results of this study, some potential directions for future research can be proposed. For instance, the survey instrument used in this paper should be tested and further developed with data from other countries and the results should be checked against differences in platforms, rules and cultures. In addition, further analysis of the significance and functioning of the various properties of e-procurement systems would help to better understand the need to improve both the usability and
