Contract Design Choices and the Balance of Ex-Ante and Ex-Post Transaction Costs in Software Development Outsourcing

0

Author Information : Michel Benaroch (Whitman School of Management, Syracuse University)
Yossi Lichtenstein (Sir John Cass Business School, City University London)
Lior Fink (Ben-Gurion University of the Negev)

Year of Publication : MIS Quarterly (2015)

Summary of Findings : In high-technology contexts, and particularly software development outsourcing, contract design ought to be sensitive to nuanced interactions between transaction and relational attributes, the balance of ex-ante contracting costs and ex-post hazard costs, the choice of contract type, and the choice of extensiveness of specific contract functions.

Research Questions : 1. What implications do the attributes of a transaction have on ex-ante as well as ex-post transaction costs?

2. How does the balance between ex-ante and ex-post transaction costs interacts with choices on contract type (Fixed-Price [FP] vs. Time-and-Material [T&M]) and the extensiveness of specific contract functions (safeguarding, coordination and adaptability)?

What we know : Extant research has focused on separate elements of the overall contract design problem, and there is a need to integrate several of those elements and their nuanced interactions.

o Ex-ante and ex-post transaction costs. Research is predominantly concerned with how contract extensiveness is affected by transaction attributes that give rise to ex-post hazard costs (of opportunism, inefficiencies, and disputes), downplaying the ex-ante transaction costs of “planning” the exchange and writing the contract; the latter costs are significant in high-technology contexts and software development outsourcing.

o Contract type. Prior research has focused on either contract type or contract extensiveness although ex-ante contracting costs also influence contract extensiveness through the choice of contract type; high ex-ante costs facilitate the use of T&M contracts rather than FP contracts, as the former need not be as detailed or extensive.

o Contract functions. Past studies have focused on the aggregate extensiveness of a contract rather than on the extensiveness of specific contract functions. Moreover, empirical work has focused on the safeguarding function although transaction cost economics also recognizes the importance of the coordination and adaptability functions.

We are first to address, in a single study, the interplay between transaction and relational attributes, the balance of ex-ante and ex-post transaction costs, the contract type choice, and the choice of extensiveness of specific contract functions.

Novel Findings : Our analysis of 210 software development outsourcing contracts that a large European bank signed with multiple vendors shows that:

Choices in both contract type and the extensiveness of specific contract functions balance the implications of transaction and relational attributes on specific ex-ante and ex-post transaction costs – these contract design choices are alternative devices that complement and substitute for one another in their ability to economize on specific transaction costs.

Explanatory power increases when analyzing the extensiveness of individual contract functions rather than the aggregate contract extensiveness, highlighting subtle competing influences that are otherwise masked by an aggregate measure.

A preference for a T&M contract counteracts the effect of certain transaction attributes on contract extensiveness, and even cancels it out in the case of transaction uncertainty.

Implications for Practice : Managers should discuss and assess relatively subtle attributes of the software development outsourcing project and the relationship in negotiating and contracting with their vendors, rather than use simple measures such as the cost or duration of a project to guide decisions on contract extensiveness.

Different software development outsourcing projects imply the inclusion of different kinds of contract provisions. Specifically, managers should first carefully decide on whether to use an FP contract or a T&M contract given the degree of transaction complexity and uncertainty, and the associated ex-ante contracting costs. Managers should then factor this choice into decisions on whether or not to include specific contractual provisions.

Any advice to write more extensive contracts must also identify the specific contract functions that need to be extensive.

Implications on Research: Contracting research ought to explicitly account for how ex-ante and ex-post transaction costs come into play in formulating hypotheses about contract design choices.

Our research model provides the basis for theorizing and demonstrating the importance of considering the extensiveness of specific contract functions rather just aggregate extensiveness – studying aggregate contract extensiveness alone might conceal granular variations and even opposing influences on the extensiveness of specific contract functions.

Contract type must be considered as a design choice that complements and substitutes for contract extensiveness, that is, a design choice that influences contract extensiveness while also being influenced by the same transaction and relational attributes influencing contract extensiveness.

Full Citations : Benaroch, M., Lichtenstein, Y. and Fink, L., “Contract Design Choices and the Balance of Ex-Ante and Ex-Post Transaction Costs in Software Development Outsourcing,” forthcoming in MIS Quarterly.

Abstract : This paper examines multiple contract design choices in the context of transaction and relational attributes and consequent ex-ante and ex-post transaction costs. It focuses on two understudied themes in the IT outsourcing literature. First, while the literature is predominantly concerned with opportunism and consequent ex-post hazard costs that contracts can safeguard against, parties to a contract also economize on ex-ante transaction costs by their choice of contract type and contract extensiveness. Second, the literature studies the aggregate extensiveness of contracts rather than of distinct contract functions – safeguarding, coordination, and adaptability. Against this backdrop, our research model portrays a nuanced picture that is anchored in the following theoretical interpretation: transaction and relational attributes have implications on specific ex-ante and ex-post transaction costs, and these implications can be balanced by respective choices in both contract type and the extensiveness of specific contract functions. These two contract design choices complement and substitute for each other in their ability to economize on specific transaction costs. Our analysis of 210 software development outsourcing contracts finds that explanatory power increases when analyzing the extensiveness of individual contract functions rather than the aggregate contract extensiveness, highlighting subtle competing influences that are otherwise masked by an aggregate measure. Our analysis also shows that a preference for Time-and-Material contracts counteracts the effect of certain transaction attributes on contract extensiveness, and even cancels it out in the case of transaction uncertainty.

Click here to access Full Paper

Benaroch model

Michel Benaroch

Michel Benaroch

Professor Benaroch is associate dean for research and professor of MIS at the Whitman School of Management. Professor Benaroch’s research addresses issues concerning the economics of IT investment, IT investment risk, ontology-centered knowledge representation, and artificial intelligence applications in finance. He has published extensively in information systems and computer science journals, including MIS Quarterly, Information Systems Research, Journal of MIS, IEEE Transaction on Software Engineering, International Journal of Accounting Information Systems, and International Journal of Human-Computer Interaction. He was ranked #26 (out of top-100 researchers worldwide) who published in top Information Systems journals (MISQ, ISR, JMIS, and JAIS) during 1999-2011.
Michel Benaroch
Share.

Leave A Reply