Sometimes, one company dominates an industry for an extended period of time, in spite of the best efforts of its competitors to catch up. For example, Intel has dominated the microchip market for decades despite the earnest efforts of Advanced Micro Devices. This may seem surprising since both common sense and empirical evidence suggest that, eventually, the follower should be able to catch up. In a variety of industrial settings, the cost of production decreases with cumulative experience, or “learning-by-doing,” but as experience accumulates, cost reductions from learning tend to diminish. If learning curves eventually flatten out, what would explain a market leader’s continuing dominance?
To explore this puzzle, David Besanko and Mark Satterthwaite, both professors of management and strategy at the Kellogg School of Management, teamed with Ulrich Doraszelski, an associate professor at the Wharton School, and Yaroslav Kryukov, an assistant professor at Carnegie Mellon University, to create an economic model that studies the dynamics of a market in which firms can benefit from learning-by-doing. A novel feature of the model is that it also accounts for companies’ loss of learning, or “organizational forgetting.” Just like people, companies can forget the know-how they gained, due to labor turnover, periods of inactivity, or failure to institutionalize knowledge that is being taken for granted.
“An analogy is tennis,” Satterthwaite says. “I used to play it very well, I practiced a lot, I developed muscle memory. But if I didn’t play for a few weeks, I’d start forgetting, and over the winter, I’d certainly forget. California kids would come and cream us kids from Massachusetts, because they didn’t have the chance to forget. They could play all year round.”
Moving Along the Learning Curve
The interaction between learning and forgetting is like “racing down an upward-moving escalator,” the researchers write in a recent paper in Econometrica. They explain that when a firm makes a large enough volume of sales to drive gains in know-how from learning faster than loss in know-how from forgetting, it moves down its learning curve and its marginal cost decreases. On the other hand, if sales slow down or stop, then the firm slides back up its learning curve and its marginal cost increases.
Specifically, the researchers found, organizational forgetting can make companies price more aggressively, which creates an equilibrium in which one company sustains a long-term market share advantage.
As a company sells more units of a product, it gains in know-how, and, conversely, the fewer units it sells, the more it is subject to forgetting. The researchers speculated that organizational forgetting might negate learning-by-doing, acting as a force that would undermine sustained market dominance by a single firm. Surprisingly, their results showed just the opposite: organizational forgetting leads to an even greater degree of market dominance by a firm that gains an edge in moving down its learning curve, and in some cases may lead to virtual monopolization of the market. Specifically, the researchers found, organizational forgetting can make companies price more aggressively, which creates an equilibrium in which one company sustains a long-term market share advantage.
Aggressive pricing by the leader is one factor that could cause a slowdown or stop the follower’s production, causing it to move back up its learning curve. “The leader’s interest is to be very aggressive in pricing,” Satterthwaite explains. “If the follower catches up, that firm is basically asking for a price war.” This is because the leader will cut its price in order to continue to produce more and keep its learning advantage. “The follower would have to work very hard to catch up; in fact, it would have to be lucky,” Satterthwaite says. “The leader would have to mess up in some way, for example, forgetting or badly managing. Then the follower could make a run and try to catch up, but that will more than likely be both very expensive and ultimately unsuccessful. If the follower is just careful and accepts its position, then the leader will not have to be very aggressive and the two firms can both make a very good living.”
The Leader’s Options
Besanko says the leader does not necessarily have to have lower costs. “We have examples of cases in which the competitors have equal costs, but the leader still has an advantage because the follower is reluctant to challenge the leader.” The willingness of a follower to accept its position in the industry is a source of market dominance that has been mentioned in academic papers but has not been shown in simple models, he says. “More complicated models, a computational model like the one we have, can show that theoretically, it’s a possibility. We see that as a core contribution of the paper.”
Summed up another way, “Bigness itself can be a source of sustainable advantage,” Satterthwaite says. “As opposed to ‘We’ve got better people’ or something like that. It’s more like, ‘We’re big and you know it, and you have to react.’”
Besanko notes that the team’s research raises questions about what kind of pricing strategies are appropriate. Under certain equilibrium conditions, he explains, firms may engage in predatory pricing, setting prices well below what they would have set if they were simply a monopoly in the market. He says that in this situation, companies may even set a price for their product that is lower than what it costs them to produce it. When that happens, in a practical sense, the company is paying people to take their products so that they can gain experience in producing them, he explains. “In fact, that’s what is happening to Intel in Europe—basically Intel arranged that if a buyer took certain minimum amounts, the buyer would get rebates,” Besanko remarks. “The EU antitrust authorities found Intel in violation; they considered that to be exclusionary behavior.”
“In our model, it is true that the pricing behavior is part of what enforces market dominance,” Besanko concludes. “Is that bad? That’s a question this paper doesn’t answer, but it does raise it as a public policy question.”
Related reading on Kellogg Insight