466. Transaction costs

Transaction costs are the costs of markets: costs of contact, contract and control. Costs of contact lie in searching for buyers/suppliers, evaluation of the quality of their products, their reliability, etc. Costs of contract lie in coming to some form of governance, in reaching an agreement, negotiation, contracting, and costs of organization (division of labour, collaboration, and exchange). Costs of control lie in monitoring performance, adjustment of agreements, haggling, conflict resolution, litigation, if it comes to that, and costs of separation.

Governance needs to deal with relational risks. A central notion here is that of relation-specific investments: investments whose value is partly or wholly limited to the relation, and is largely worthless outside the relationship, and therefore entail switching costs: costs of switching to another buyer or supplier. The possible loss of such specific investment constitutes a risk, in that the partner can threaten to break up the relationship at the cost of that loss. As a result, such investments cause dependence that can be used as leverage in the distribution of jointly produced added value. A condition for making a specific investments is that one can expect that the relationship will last long enough, or will yield a sufficient volume of transactions, to recoup the investment.

Why incur such problems and not avoid specific investments? Often, most value is added, and most profit made, with specialized, differentiated products that distinguish themselves from the competition. That is mostly achieved by combining complementary resources from different firms (knowledge, technology, skills, market access, …) into novel, unique combinations of features. However, that requires relation-specific investments. So, one incurs the problems of dependence to utilize an opportunity for profit. As a result, collaboration is as much part of markets as competition is.  

There is a dominant rhetoric of maximum flexibility to ensure maximum efficiency, in hiring and firing workers, in buying and selling parts of firms, and in beginning and ending relationships of collaboration. The logic of specific investments goes against that. Instead of maximum flexibility one should go for optimal flexibility, with relationships that are durable enough to elicit specific investments but not so fixed that they yield rigidities in the inability to adjust to novel conditions or opportunities.

There are also implications for choosing between on the one hand markets and on the other hand integration within a large organization. When the costs of the market exceed a certain limit, its motivational advantages, in partners being independently responsible for their own survival, becomes less than the costs involved in exchange and governance, and it is more efficient to integrate.

The standard theory proposes that the greater the uncertainty the higher relational risks and hence the greater the preference for integration. However, that is not quite right. It is useful to distinguish between behavioural and technical/commercial uncertainty. The claim applies for behavioural uncertainty, as when motives and capabilities are difficult to judge or when there is a low general level of trust. Under the technological and commercial uncertainty of innovation, on the other hand, outside, independent partners contribute more to the necessary variety of knowledge and the flexibility and scope of a variety of relationships. So there the prediction is the reverse: more uncertainty pleads for less integration.