Hypothesis 3: Interorganizational trust will result in a less detailed alliance contract.

2.4 Control variables

We included four control variables in our analysis concerning the antecedents of alliance contract details. We include focal firm size as a variable in our model to control for extraneous factors such as bargaining power and resource base (Arino & Reuer 2004).

These factors may influence the governance because large firms have more legal resources, experience and staff, and may be more successful in directly extracting hostages than smaller firms. Secondly, we assess the strategic importance of the interfirm project (Reuer et al. 2003). Companies will design more detailed contracts for alliances that involve strategically important projects because they are more exposed to the hazards of the interfirm alliance.

A similar argument applies to the complexity of interfirm projects. IT literature suggests that contracts are more detailed according to an increasing complexity of projects because of a need to avoid ambiguity and a need for coordination among alliance partners (Anderson & Dekker 2005).

Our final control variable is risk avoidance, that is, the willingness to take risks differs among firms and this is reflected by the level of detail of formal contracts (Nooteboom 1996).

3. METHODS

3.1 Data collection and sample

This study focuses on business relationships between two or more firms and/or research institutes that operate in high-tech industries (biotechnology, new material development, information technology, maritime technologies and environmental technology). The lifecycle of R&D in these industries is usually very short. Much of the new technological knowledge quickly becomes outdated, often even before it has been incorporated in new products and/or services. Hence, in the high-tech industries in particular, we find many collaborative efforts between firms, including rival firms. Furthermore, given environmental uncertainty, we expect contracts to operate in this context.

Our research proceeded in three stages. In the preparatory phase of the fieldwork, we conducted 25 semi-structured interviews with consultants of the Dutch Ministry of Economic Affairs that were involved in policy programmes to stimulate interfirm collaboration on innovation. Additionally, the consultants selected 20 cases (ten successful and ten less successful ones) that we studied in great detail to obtain in-depth knowledge of the high-tech collaboration.

Case research is suitable for exploratory research where understanding is the primary objective (Yin 2003). The 20 cases dealt with collaborative innovation and hence involved complex transactions for which close collaboration between partners was necessary over a considerable period of time. The cases involved legally independent partners that shared costs and benefits more or less evenly. All cases entailed uncertainty and/or complexity, and specific assets, and hence risks of dependence, opportunism and 'hold-up'.

Under strict confidentiality, we received full access to all documents of the cases--including the interfirm contracts but also project plans, annual reports of the companies involved, personal notes and letters, and half-yearly progress reports--that were available at the Ministry. Among other things, this allowed us to examine the content of the contracts with respect to the clauses that were laid down in the contract and the exact content of each clause. Also, clippings from newspapers and trade magazines concerning the collaborations were collected.

To enable comparison between the cases and to ensure the quality of the case analysis, a case protocol was written (Yin 2003), to describe the alliance's history, development and outcome. The interviews with the consultants were transcribed into interview reports and send back for verification and agreement. Hence, all this allowed us to reconstruct the development of high-tech alliances and to check the data from the interviews with the secondary sources. We used this information to design our survey. The survey was field-tested using a sample of ten companies involved in R&D alliances. This resulted in a number of modifications to the questionnaire.

In the second stage, a research team conducted telephone interviews with 572 business managers of interfirm R&D collaboration. Prior to these interviews, all managers received an explanatory letter inviting them to participate. We briefed the team on the features of R&D, high-tech industries and interfirm relationships. The team made three attempts to identify and interview the selected respondents. The case firms were identified from a database of Dutch interfirm high-technology alliances published by the Ministry of Economics Affairs. This enabled us to identify the business managers who were responsible for interfacing with the partner firms. They were considered to be the most knowledgeable informants about the interfirm relationships. During the interview main topics such as the history and purpose of the alliance as well as contracts, investments and industry dynamics were discussed.

One of the first questions required the respondents to identify the business partner in the alliance in question. We used this information to cross-validate the information from the database.

Because high-tech alliances are typically concerned with specific projects and goals, we also asked the respondents to identify one project that was the most important to the interfirm alliance. By focusing on interfirm collaboration within one sector (high-tech industries), we reduced the range of extraneous variations such as the level of uncertainty or competition that might influence the constructs of interest. Some open questions were added to enliven the interview and to enable the respondents to tell their own story to some extent. In total 50 main questions (often divided in several sub questions) were asked. An outcome of this was that the interviews that were designed to take half an hour would sometimes take up to one hour depending on the respondent.

We obtained 391 useable responses, giving an effective response rate of 68.5 percent.

This rate is considerably higher than those observed in prior studies on interfirm relationships that usually is in the 10 to 33 percent range (Parkhe 1993; Poppo & Zenger 2002; Subramani & Venkatraman 2003). It was also satisfactory considering this studies' requirement for direct senior management involvement and the confidentiality of some of the requested information. Although the high level of response from knowledgeable executives that were closely involved in the management of the high-tech collaboration was encouraging, it does not directly address the potential issues of consistency motives and social desirability (Podsakoff & Organ 1986). When self-reported two or more variables are collected from the same source at one time, correlations among them may be systemically contaminated.

However, for the aim of this study, reliance on key informants such as our respondents seems to be the only realistic and feasible way to obtain the required information (cf. Huber & Power 1985). We used the following actions to address possible concerns of validity in stage three of our research.

Secondary data

Available data can be tested for convergence by triangulation with secondary data (Keats & Hitt 1988). We compared the outcomes of the self-reported data in the questionnaire with the archival data on the 20 cooperative projects that we studied the first phase of the data-collection. The congruence of the data from the questionnaires and case studies supports the accuracy of the reported data.

Questionnaire structure

Via the sequence of our questions we aimed to minimize the effects of consistency artefacts. Whereas Salancik and Pfeffer (1977) suggest letting the independent variable follow, rather then precede, the independent variables, Podsakoff and Organ (1986) argue that correlations will be similar using either method. In our opinion, a lifecycle approach would best serve an accurate reflection of the interfirm collaboration.

Hence, for the purpose of this study, we structured the questions in the survey from past interactions through partner selection, contract negotiations, contract execution and outcomes of the interfirm collaboration.

Non-response analysis.

The non-response is low (31.5 percent) especially considering that only 10.5 percent actually refused to be interviewed. 20.1 Percent could not be contacted within the 3 attempts that the interviewers used to try to get in touch with the respondent.

To investigate whether the non-response incurs a bias, the non-cooperating respondents (10.5 percent) were asked for their reasons not to participate. The reasons for refusal were on the one hand a lack of time and interest, and on the other hand, irritation because they had recently cooperated in another survey. Although these reasons can hide their true motive for not participating in the survey (such as an unsuccessful cooperation), the low non-response and the reasons for not participating do not raise serious doubts on the implications of non-response.

3.2 Measures

Table 1 provides an overview of the items that we used to measure the constructs of our theoretical model.

The dependent construct is contract detail. The business relations that we analyse are characterised by high uncertainty or complexity, entail substantial alliance-specific investments, and require intensive knowledge transfer. Empirical studies suggest that for these circumstances, alliance contracts typically include clauses safeguarding (intellectual) property rights and spillover (ownership of knowledge, products or methods, pledge of secrecy, sanctions on spillover); clauses determining the management of complex relationships (relationship duration, project management, investments, communication); and clauses relating to future contingencies (environmental uncertainty, relationship adjustments, liability 'in the event of') (Klein Woolthuis et al. 2005).

Print Get the Mag Weekly Updates