The Price of Advice

Consulting is big business these days. Estimates put the turnover of management consulting firms in the United States alone at more than $120 billion. And, closer to home, roughly a third of every graduating MBA class at Kellogg joins a consulting firm for a median salary well into six figures. Since “consulting” can mean anything from providing information or advice professionally to firms and individual clients; working in fields such as information technology (IT), marketing, or human resources; or being a real estate agent, financial adviser, or lawyer, it is clear that consultants form a sizeable portion of the economy. Why consultants exist and, by direct extension, when clients retain their services, are questions that have been thoroughly analyzed. First of all, consultants possess specialized knowledge. For instance, financial advisers know significantly more about a variety of available investment vehicles than naïve investors. Second, consultants bring a fresh perspective to problems—management consulting firms pride themselves on their ability to deliver innovative ideas because of their outsider status.

The Value of Information
Despite the proliferation of consulting services, however, a formal theory that explains how consultants create value has been missing. Standard economic theory addresses how to make profits from providing goods or tangible services (manufacturing, labor). Economists soon realized that, similar to the way monopolist manufacturers make profits on the goods they provide, people who own information can command a “rent” for this information. In other words, they can command more money than the cost of the information. This field, known as information economics, grew over the 1970s and 1980s, and provided much insight into the organization of markets and various market mechanisms such as auctions. (The 2007 Nobel Memorial Prize was awarded to three early contributors to the foundations of mechanism design; see Professor Rakesh Vohra’s essay cited among the further readings.) However, these theories assumed that the person who owns the information also understands its value, an assumption that may not be suitable in this situation.

To understand why, think about a real estate agent showing houses to a prospective customer. The agent presumably knows about the state of the local market and what kinds of houses are available. But the agent does not know how the customer values this information. The customer may not reveal this for several reasons. For example, the client may not reveal his budget out of fear that the broker will then show him only the priciest houses in his range to maximize the commission, which is set as a fixed percentage of the price. Under the circumstances, it is not clear how much of the “information rent” the agent will be able to extract. Péter Eső, Assistant Professor of Managerial Economics and Decision Sciences at Kellogg, and co-author Balázs Szentes (University of Chicago) tackle this problem. Professor Eső explains, “The worst-case scenario is analogous to a client who needs to make a decision based on data in an encrypted file. The consultant is someone who has the key to decrypt the file, but no clue what the file itself contains, or how the client will use the data in question. We ask how the consultant can charge for his information in these circumstances.”

Under the right circumstances, having control over the information is all that matters in order to extract value.Eső and Szentes’s formal model involves a client who needs to decide whether or not to invest in a project. The client knows his costs (i.e., how much needs to be invested), but is unsure about the benefits that will accrue as a result. In particular, the client knows a portion of the profits, but is unsure about another component that may be positive or negative (i.e., a loss). He will invest in the project if he expects that total profits will exceed costs.

In what the authors consider the best-case scenario, a consultant would know the profitability of this component. Under these circumstances, the authors calculate the rent the consultant can extract. Intuition tells us that if the consultant does not know the profitability, he would make less. Consider, then, the opposite case—the consultant has information that will help the client determine profitability. The problem is that the consultant does not know how the client will use this information to determine profitability, or the profits that will result. For example, the consultant might be able to suggest a novel accounting framework to evaluate the project, but he has no idea what profits will result if the client uses this framework.

To analyze this model, the authors use an offshoot of game theory known as the Principal-Agent model, which involves a principal (in this setting, the consultant) and an agent (the client). The agent can take actions (invest or not invest) that will affect the principal (in this case, how much the consultant gets paid). The problem is how the principal should design a contract that spurs the agent to take actions that the principal prefers.

The Value of Controlling Information
The key finding of Eső and Szentes’s work is that an appropriate contract allows consultants to extract as much rent as if they were able to observe the profitability of the project themselves. It does not matter whether they understand the value of the information they have, or if they become better informed (as in Dana and Spier’s 1993 models of the attorney-client relationship). It only matters that they control access to it, meaning that clients can get relevant information by working with consultants. It is critical, however, that consultants be able to write and enforce a contract in which clients’ payments are contingent on the decisions they make after learning the information.

Let us imagine clients’ decisions in a world where consultants gave them information for free. If clients chose to invest with the information at hand, the implication would be that they had a net positive profit from the project, or a net loss if they chose not to. If the portion of the profit (or loss) that clients knew about were extremely large, the clients would almost always choose to invest (or not invest) regardless of the information. Therefore, consultants’ information would be worthless. The information is valuable when it influences the decisions clients would make, i.e., when the components they know about give them an intermediate level of profits. In this case, consultants can use client’s decisions to calculate the value of the information. The optimal contract would ask clients to pay just enough to not cause them to change their decisions from the ones they would have made if they could have viewed the information for free.

In brief, this work highlights the value of having information in the information age. Prior work in economics and management literature suggested that it was equally important to understand the value of the information one had in order to be able to extract value from it. Eső and Szentes challenge this conventional wisdom, showing that, under the right circumstances, having control over the information is all that matters in order to extract value. Their formal model shows that contracts in which consultants’ fees are contingent on clients’ actions, rather than the outcome of the actions (whether projects are profitable or not) are optimal. One can observe many examples of these contracts in fields such as real estate, law, IT consulting, and merger and acquisitions consulting.

Further reading:

Besanko, David, David Dranove, Mark Shanley, and Scott Schaefer (2006). Economics of Strategy, New York, NY: Wiley & Sons, 4th edition.

Dana, James D. and Kathryn E. Spier (1993). “Expertise and Contingent Fees: The Role of Asymmetric Information in Attorney Compensation,” Journal of Law, Economics, and Organization 9(2): 349–367.

Salanié, Bernard (2005). The Economics of Contracts: A Primer, Cambridge, MA: MIT Press, 2nd edition.

Vohra, Rakesh (2007). “Nobel Designs—Understanding Mechanism Design,” Kellogg Insight, November.