Faculty member in the Department of Managerial Economics & Decision Sciences until 2015
Anyone who has ever made a major purchase—a home, a car, a camera, a computer—knows the delicate dance performed between ourselves and the salesman. “Why should I buy the model you’re recommending?” we quiz this stranger asking us to fork over hundreds, even thousands of dollars based on his word. “Well…” begins the salesman, before spewing a torrent of facts, figures, and product reviews, all favoring the selected brand.
“Anything else I should know?” we ask ourselves, because no product or deal can be perfect. “By the way…” a car salesman might volunteer, seeming to read our thoughts. “I should tell you that Car and Driver criticized the on-center feel of this model’s electric power steering, but the problem was fixed this year.” Or in the case of a new computer, “I should tell you that this computer relies on graphics integrated into the processor rather than a standalone GPU.” You, the buyer, may have no idea what this statement means. But still, you begin feeling positive about your decision. After all, the salesman just listed a negative point, so what else could there be not to like?
Plenty, says Kellogg economist Wioletta Dziuda, who deconstructs strategic decision-making using mathematical models. Dziuda’s research focuses on consumer buying-selling scenarios, and lately, she says, she has been thinking a lot about the manipulation that occurs when salespeople and marketers are not entirely straightforward.
“What I was motivated by,” explains Dziuda, an assistant professor of managerial and decision sciences at the Kellogg School of Management, “was, first, my observation of what happens when the buyer interacts with the seller, or the renter interacts with the landlord. My starting observation was that usually not all information is revealed. But nevertheless, the expert does provide some information that seemingly goes against his or her interest.”
In an article published in the Journal of Economic Theory, Dziuda revealed a possible answer to this puzzle. Consider again the expert in the transaction—the product salesman or saleswoman, or in the rental-apartment scenario, the landlord or Realtor— who might reveal the battery life of camera brand X is a tad short or that the bedroom wall has a crack. The expert, Dziuda wrote, typically will present a limited amount of negative information, bulking up his or her credibility by appearing honest. But then the expert stops and again piles on positive information about the product to seal the deal, Dziuda says. The greater the number of favorable arguments that emerge, the more easily swayed the decision-maker.
Might the expert tell an outright lie? Probably not, Dziuda reasons, citing consumer protection laws and corporate ethics standards. But the buyer still might be manipulated by the withholding of negative information. As a decision-making consumer, or potential renter, “If you see only a few pieces of negative information, you should ignore it,” Dziuda advises. “That’s not necessarily a sign that the expert is honest, because he just tries to appear honest. But you should be smarter than that,” she adds. “A strategically smart listener will realize the subcontext of the expert and not be fooled.”
As buyers, we should realize that there are negative things we cannot see and certainly are not hearing about from the seller. “It’s just that we are not experts; we cannot even name what might be going wrong,” she says. Our minds tend to focus on the things we can see and the things we are told. To do differently would require a lot of thinking—and something we do not typically do as consumers.
The buyer does not take the information the seller reveals at face value; she tries to think why the seller revealed this particular information.
What Dziuda found surprising is that decision-makers are not influenced by unfavorable arguments. “At least in my model, which is a construction of reality, there’s this tension,” she says. “On the one hand, if the expert tells me just one negative thing, I know this negative thing exists. But on the other hand, given that he told me the negative factor, I might think he’s more likely to be honest. So there’s this tension, and in my model it washes out, and basically you say, ‘Well, I don’t know how to take this information; I’m not going to take this information into account when I’m making my decision’.”
Dziuda’s model is also on the hunt for equilibrium. In the world of numerical models about decision-making, “equilibrium” means that a balance of needs in an interaction between two or more individuals is achieved. In buying and selling, equilibrium occurs when the buyer guesses what arguments the seller will employ to convince her to buy; she then responds to these arguments by opening her wallet. On the seller’s side of things, equilibrium occurs when he guesses what arguments the buyer will find persuasive and uses them—because they will be to his advantage.
In short, the buyer does not take the information the seller reveals at face value; she tries to think why the seller revealed this particular information. What is he withholding? she asks herself. And if he offers a lot of negative arguments, does that fact mean he is to be trusted completely?
Dziuda, whose past work has included studies of collective bargaining and Europe’s changeover to the euro, says her recent research on information manipulation and conflicts of interest is based on her “real-world observations of interactions between people. Being interested in politics and the media.” She adds, “I was always interested in how people can be manipulated and how people who are informed manipulate people who are uninformed.”
The potential applications include not only consumer products but also scholarship, media, and politics. “If you think about how politicians try to influence people, they usually don’t lie, because if you are caught lying that has negative consequences,” Dziuda says. “So the way people try to manipulate the voters is by selective disclosure”—like the landlord who points out the crack in the bedroom wall while conveniently leaving out next door’s aspiring and, by the way, insomniac, drummer.
Dziuda ponders whether her model might shed light on when politicians can manipulate people and how people should react to what they hear. “They should realize there’s a lot of information that might be left out,” the economist says of voters, “because just because politicians cannot lie does not mean that they will not present information selectively.”
Related reading on Kellogg Insight
Firming Up the Foundations of Game Theory: Elucidating the role of information in strategic interactions
When to Come Out Swinging: Devise the right strategy for negotiations or online auctions
There’s No Place Like Home: Investment bias toward domestic funds
Dziuda, Wioletta. 2011. “Strategic Argumentation.” Journal of Economic Theory. 146(4): 1362-1397.