Incentives: How they Function

Dec 8, 2016 — Neil Boyle


“The problem with practitioners and researchers is that practitioners practice but don’t read, and researchers read but don’t practice” (John Briscoe, former head of the central Urban and Water unit of the World Bank, 2014)

The incentives of PPP infrastructure projects start within the transaction— in the economic exchange that occurs beween buyers and suppliers.  The expected performance of the parties generally serves to trigger motivation.  Motivation for these incentives come from three sources that are called transaction attributes: asset specificity—the special term economists use to describe the level of specialization (k, from zero/negligible to high) that is exchanged between buyers and suppliers of a transaction; uncertainty  (from zero/negligible to high) another transaction attrubute with a unique adaptive need for additional information (the ambiguity associated with makiing a choice), a cost that is associated with the absence of information; and frequency—a third transaction attribute that refers to the number of times the transaction is repeated and the reputation effects that emerge as a result especially when communication among parties is facilitated as in a cooperative or suppliers association (from negligible to high). Among the three, asset specificity is the central attribute for discussing incentives.

Specialization is needed to solve the problem of complex production.  Complex production is another term economist use to describe transactions that involve incomplete contract, specialized assets, uncertainty, and together a bilateral dependency sets in between the buyer and the supplier. Most infrastructure investments by definition involve complex production due to the specialized assets that are within the exchange process.  A specialized exchange asset is what gives value to the transaction that attracts corruption in one form or another.  This may not include collusion among bidders at the bidding stage of a project because collusion can be for any reason, while for the specialized version of corruption there is a specific value that can be identified.

Curiously enough, Adam Smith was the first person to grapple with the role specialization played in the organization of production.  His study of the economic organization of 18th century Scotland’s pin industry led to solving the dismal production of a few number of pins per day to several thousands per day by virtue of understanding the way the industry was organized. By understanding the industry, he managed to observe its market price variation.  Smith accomplished this by discovering the relationships between and among the institutional concepts of the division of labor, specialization, economies of scale, assembly line production, and self-interest rather than individual fiat by monarch rule.  On this last point, Smith found that individuals acted on the basis of intrinsic self-motivation rather than on what the reigning monarch said was to be produced.  Asset specificity, however, is a double edged sword—the higher the specialization the more nuanced are the incentives, and the more nuanced the incentives—the more opportunism sets-into the transaction that tends to disrupt trouble-free operations.

As a result, this form of reciprocal relations between asset specificity and opportunism creates a need for cooperation from the counterparty.  Otherwise, in the presence of opportunities for private gain, the guileful self-interest of the counterparty is activated. Because guile is a universal human behavior, this works for both the buyer and supplier; guile is a threat to both parties and hence to the success of the transaction.

These special purpose transactions are almost always governed by hybrid (aka long-term contract) at mid=levels, of asset specificity, and at very high levels of asset specificity, hierarchy (aka the firm) governance structures. General purpose (generic) transactions are governed by market governance structures, the third of a generic (underlined) set of governance strucetures (i.e., markets. Hybrids, and hierarchies)   Market intermediation excludes guile from taking root for two reasons. The intense incentives of the market override most mediation hurdles, and thick markets provide ample and ready exit strategies for trading with others. The supplier’s incentive would have been to find the best competitive price for the good or service among the thick market of numerous traders.  Moreover, if generic assets were used, the value of continuing with the transaction once the bill is paid diminishes to zero because general purpose assets do not need to be continued—construction, equipment installation, commissioning, and operating the asset are typically excluded from market trading.  Once nuts and bolts are fastened and the bill is paid, the technology operates on its own; this is generally quick, if not instantaneous in spot markets.  Maintenance of generic assets through some form of warranty may be needed.

In both specialized and generic governance cases, the business purpose of both parties is to jointly increase the net present value of the transaction so that the project company benefits financially, and hence in the case of PPPs, both partners gain according to pre-agreed shares, that is, there are no bargaining costs.

But there are two additional reasons why incentives are unique in PPP infrastructure projects.  The first is PPP infrastructure assets have negligible productive value outside of the transaction for which they were intended; they are non-redeployable durable assets.  In financial terms, they are sunk costs.  For the supplier, the asset must complete its work in its intended transaction or it is likely to be scrap, and in the event the project is terminated, its value would be near salvage levels. The implication is in a pragmatic sense infinite continuity. The second reason is PPP assets are unprotected from the guile of the counterparty so both parties require the cooperation of the counterparty to protect its asset.  This means the two parties to the contract are bilaterally dependent and autonomous at the same time; they are dependent and independent simultaneously.  Autonomous bilaterally dependent incentives have profound implications on PPP infrastructure systems where user charges are involved.  They are likely to antagonize the indigenous culture of consumer’s self-interest of affordability and equilibration of differences that are associated with high bargaining costs.

When contract incentive structures are characterized by autonomous bilateral trading and incomplete contract, disruptions will almost certainly occur and measures must be taken to mitigate them promptly.  Delay in the form of increased search costs (e.g., estimating the probability of every contingency) through “blackboard economics” (as mentioned by a conferee at the Ronald Coase Institute conference, Washington, DC, 2015), insertion of superior institutional measures or arrangements should be initiated promptly without delay. These measures are called “provisioning of contracts”, clauses that serve to reassign property rights of organizational business rules of counterparty bureaucratic structures, among other things, are thus obtained.

An important distinction should be made here.  The difficulty lies in the relationship and not in the individuals, per se.  Autonomous individuals can handle most disruptions (by definition).  However, in the autonomous bilateral interaction, and in the absence of a sequential decision-making apparatus, parties do not know or cannot rely on what the other is doing because information flow is erratic, especially when the stakes are not trivial.  This is the main characteristic of a complex contract.

We also learned from organization theorist/economist Herbert Simon (1963) and Oliver Williamson (1985) that promises were never self-enforcing, but because we glossed over this fact, it was easy to continue doing so rather than to find a solution to the problem such as “credible commitment”. (Credible commitment will be explained subsequently.) Instead, we tended to rely on “efficient risk bearing” to mitigate any risk we identified, which was convenient to do under the circumstances.  However, efficient risk bearing is problematic; it succumbs to the guile of self-interest.

Within all of the above, the twin assumptions of human behavior of bounded rationality (limited cognitive capacity) and opportunism (guileful self-interest) of agents who design and manage transactions are constant contextual conditions, the effects of which always require safeguarding the agreement in project design, implementation, and enforcement.  Similarly, politics must be taken into account in ex-ante contract design and ex-post implementation because property rights are defined through democratic politics.  One implication of this is that while changing property rights is well known among organizational theorists, another change process may be through the machinery of politics.

Production opportunities that interact with the transaction’s asset specificity and uncertainty create incentive structures.  For example, production price is a common incentive setter.  Production quantity and related contingencies are also incentive setters.  A cost plus contract sets up different incentives than say a lump sum contract.  The former gives budgetary discretion to the winning bidder, while the latter limits the winning bidder to execution within a specified budget.  The former is used in technologically complex projects where advanced technological engineering studies are incomplete and on-going.  The combination of price and contingency is akin to reducing the certainty of the established price.  The latter raises the question of contract agents’ staying-on-the contract curve (i.e., sticking to the agreement).  Other combinations of production opportunities and the attributes of transactions can also create incentive structures.  For example, the market buy-price for a farmer’s crop sets-up different incentive structures than an administratively-set buy-price for the same crop.

This article is filed in technicaltools. Start or participate in a discussion below or click here to see more articles in this category.


Please share your thoughts and comments about the article. We look forward to hearing from you. Fields marked with "*" are required: