In this posting I’m going to discuss an academic article. I will greatly simplify it and pull out only the aspects that are interesting to us as pricing professionals. “The Effects of Consumers’ Price Expectations on Sellers’ Dynamic Pricing Strategies” by Hong Yuan and Song Han was published in Journal of Marketing Research, Feb 2011. This post discusses only a few interesting aspects of the article, but if you want to know more about the study you can either read the very brief description at the bottom of this article or the authors’ abstract.
One thing I like about this article is they hit us over the head with the concept that customers choose how much effort to put into searching for alternatives, and how our prices influence this decision.
Customers have a reference price, the price they expect to pay or the price they think is fair. If they see that your price is below their reference price, they will likely just make the purchase without searching much if at all for alternatives. However, if your price is higher than their reference price, then they will likely put more effort into shopping.
In a future post we can talk more about the many different ways reference prices are formed, but the one method used in this article is your previous price. For customers who purchase the same item over and over, they may remember the last price they paid and they likely made that their reference price. This means that every time you raise your price, you set your price above your customers’ reference price. This is the same as inviting your customers to search for alternatives. Don’t raise your prices (unless you can’t avoid it). This also means that if you are able to lower your prices, your customers are less likely to shop around.
Of course our costs don’t always support our ability to lower prices or even hold them steady, but as pricers we must do our best to manage our customers’ expectations, their reference prices.
This becomes one explanation for why prices rise so quickly when costs go up, but fall more slowly when costs go down. For example, when the cost of a barrel of oil goes up, gas stations tend to raise their prices the same day even though it takes time for that more expensive crude oil to get refined into gasoline and transported to the gas station. However, when the price of a barrel of oil goes down, we don’t see the price at the pump change so quickly.
The explanation, using the logic in this article, is that when we see an increase in costs we want to raise prices only once if possible. After all, when we raise prices we are inducing our customers to search for alternatives, which we’d rather not do. If we have to do it, let’s do it only once. As costs come down, we could lower our prices immediately and get the benefit of our customers not shopping around. However, if we lower our prices slowly we get the benefit of our customers not shopping around several times. This seems much more beneficial. Raise prices quickly when costs go up so we only induce search once. Lower prices slowly when costs go down to deter search multiple times.
The bottom line, when our price is below our customers’ expectations they are more likely to purchase. When our price is above their expectations they are more likely to search for alternatives. Raising price induces search, lowering price reduces search. As pricers and as marketers we must manage our customer’s reference price, the price they expect to pay.
=== About the article ===
The article used game theory where the players were companies that faced changes in their cost and consumers who faced a cost to search for alternatives. The authors proposed and supported 5 propositions, two of which were discussed above. P2 – “The higher the buyers’ observed price in the current period, the more they search; the higher the buyers’ expected price, the less they search.” P4 – “Sellers raise prices more quickly in response to marginal cost increases than they reduce prices in response to marginal cost decreases.”
The authors then described a couple of experiments in which students were given the role of seller or buyer. The participants were also given actual monetary rewards based on how well they did in their role. Their experimental results supported their propositions.