Costs, Competition, and Prices that Don't Clear the Market

We at Economic Forces are big fans of supply and demand. Critics, however, might ask how we explain prices that do not clear the market — in other words why, in some markets, there seems to be excess supply or excess demand at the current price that doesn’t result in a change in price.

For example, places like Chik-Fil-A charge the same price for a chicken sandwich regardless of the time of day and the length of the line. Last week, Brian tried to answer the question, “Why aren’t Pappy (van Winkle bourbon) and other rare whiskies sold at market prices?” that was first asked by Jeremy Horpedahl. And there is currently a big debate about why employers won’t simply raise their wages when they are having a hard time finding workers to fill their job openings. Each of these examples is about prices that are not clearing the market. Why doesn’t Chik-Fil-A raise its price at lunchtime? Why don’t stores increase the price of rare whiskey? Why won’t firms raise their wages?

In this post, I would like to argue that all of these questions can be answered by an appeal to the insights of the great Armen Alchian when it comes to thinking about costs.

Armen Alchian had two particular views that are valuable when thinking about market decisions. First, Alchian’s view of cost was primarily in terms of opportunity cost. What you are giving up when you make a choice is the next best alternative. This is the cost of an action. Second, Alchian thought it best to think about costs in terms of present value. Taken together, this view of costs can generate a lot of insight.

When thinking about investment, it is easy to understand why one might think about the costs and benefits of an investment in terms of present value. Some of the costs occur now. Some of the costs occur later. Almost all of the benefits accrue later. If one values the present more than the future, then thinking about this in terms of present value seems somewhat obvious.

However, Alchian’s insight about costs was not only about investment. Many of the models that we use in class are static models, but the effects of particular actions are dynamic. How can we think about dynamic issues in a static model? By thinking about things in terms of present value! To think about why this might be useful, let’s start with the Chik-Fil-A example.

In a previous post, I asked why prices are sticky. When I walk into a Chik-Fil-A at noon, the lines are long. When I walk in at 2 p.m., the lines are short. Yet, I pay the same price at both times of day. Doesn’t this violate supply and demand? Shouldn’t the price be higher when demand is high than when demand is low? What explains this?

Think about the costs to the consumer. The cost of buying Chik-Fil-A is the opportunity cost. It is what I could have purchased if I had not bought Chik-Fil-A plus what I could have done with my time if I had not waited in line. By this measure of costs, the cost of the Chik-Fil-A sandwhich is higher at noon than at 2 p.m.

Of course, that does not explain why Chik-Fil-A doesn’t raise its price. The restaurant does not gain any benefit from the cost associated with waiting. Why not turn that non-monetary cost into a monetary cost? Conceivably, Chik-Fil-A could monetize this willingness to pay by charging a higher price. This is where Alchian’s insights matter.

From the perspective of the customer, a constant price throughout the day creates a reliability in terms of monetary cost per unit of quality. Those with different time costs can therefore choose to eat at different times per day. Furthermore, consumers not only have reliable information about price per unit of quality, but also about the approximate wait times at various different times per day.

Now imagine that Chik-Fil-A installed menu boards in which the price was constantly adjusted based on the length of the line. This would eliminate costs associated with waiting because some people would get out of line. However, this strategy would create a cost associated with searching. If it is noon and I want Chik-Fil-A, this pricing strategy allows me to show up and avoid a line. However, I have a maximum willingness to pay for a chicken sandwich. If I show up and the price is above my willingness to pay, I have to eat somewhere else or I have to come back later to check the price.

One might think that solving this problem is easy. For example, why not just show up at 2 p.m. when you know that demand should be lower? You should be able to get your sandwich at that time with no line and even at a lower price. Well, everyone else knows that this is the case too. Thus, everyone might try to eat at 2 p.m. because that is when they expect the price to be lower. However, if everyone thinks this way and shows up at 2 p.m., the price will be high and if I don’t want to pay that price I likely won’t eat lunch.

The relevant question for thinking about which alternative that consumers would prefer appears to be whether the monetary costs and time costs with constant pricing are greater than the monetary costs and search costs with variable pricing. What Alchian tells us is that we should think about this in terms of the present value of these costs. I don’t just care about whether search costs are high today or the cost of waiting is high today. I care about the present value of the search costs, the present value of the waiting costs, and the present value of the monetary cost.

Now think about these two alternatives. Searching requires balancing a trade-off. All else equal, the longer I search, the lower the price that I will pay. I can balance this trade-off optimally so long as the marginal cost of my search is equal to the marginal saving that I get from searching. However, since the price is a random variable, even if I optimize my search, the time spent searching will also be random. Thus, even with optimization, I cannot completely minimize my costs because some fraction of my costs are randomly determined. With constant pricing, I do not have to search on price because prices are constant throughout the day, but I might have to wait, depending on when I show up. In this case, I can minimize my costs simply by showing up during downtimes. Furthermore, regardless of when I show up, the cost is predictable, unlike in the world of searching and constantly change prices. Evidently, the expected present value of the costs of searching and the resulting monetary price are greater than the present value of a constant monetary price and the cost of time.

The reason that firms choose this strategy is similarly based on the present value of the costs. If searching is more costly, then the demand will be lower for the product; not just today, but in the future. Thus, even if the higher price could generate more revenue today, the firm is stuck with lower demand in the future. The cost to the firm of these price adjustments is therefore related to the present value of the decline in demand that would result. The firm can prevent these reductions in demand by keeping the price constant. Although it cannot monetize the time cost of waiting, the firm is better off. What the firm is offering is reliability.

This seems to be the same point that Brian was making in his post about reliability. But while I think that the lesson about the price of whiskey is related to the above example, I don’t think that the answer has to do with reliability in this case. I think it has to do with firms thinking about costs in terms of the present value of opportunity cost; in this case, the present value of foregone demand due to a bad pricing strategy.

What the Chik-Fil-A example hopefully demonstrates is that what firms care about is not just the current demand for their product, but the present value of demand. A firm might be willing to trade-off a higher demand now for a lower demand later, or vice versa, if the present value is higher. However, what firms do not want to do is boost demand today if it means a lower present value of current and future demand.

As I alluded to above, rare whiskey tends to sell for a lower price at a store as it does in secondary markets. Why do firms forgo these higher prices?

These stores could simply auction it off to the highest bidder. Auctions aren’t that costly. As a result, an auction would clearly seem to generate more revenue to the firm. This seems like a no-brainer. Why don’t they do things this way?

I think the answer is related to thinking about demand in terms of the present value of demand. With an auction, only those with the highest willingness to pay will be given the opportunity to purchase the whiskey. Suppose that a competitor offers an alternative. A competing liquor store offers to allocate the rare whiskey through a lottery. By doing so, this liquor store will likely attract the business of customers who are interested in buying the whiskey, but who are unlikely to win the auction. Since entering the lottery will require visiting the store, this increases the present value of demand at the liquor store offering the lottery.

This strategy seems similar to the strategies used by companies who make video game systems. When first released, these systems regularly sell at prices on secondary markets that are much higher than the retail price. Why are these companies forgoing higher profits by selling these gaming systems below their market value? Again, the answer seems to be that some people are indifferent between one gaming system and another. However, the purchase of one gaming system locks-in the customer to that ecosystem of games. If one company charges the higher price, they could lose all of the customers indifferent between systems.

Of course, lotteries aren’t the only alternative to auctions. Another way to allocate the whiskey is to offer it to certain customers. When a shipment of rare whiskey comes in, the owner or the manager of the store calls a select number of people and offers to sell it to them. This, of course, seems somewhat corrupt. Why would the store do this?

Just like customers might prefer certain stores, stores might prefer certain customers. This type of allocation is like a “customer rewards” program for loyal customers. Those customers who already have a high present value of demand are rewarded with the opportunity to purchase rare whiskey. This offer isn’t available to everyone. It therefore encourages loyalty to the store. While these customers might not have the highest willingness to pay for the rare whiskey, they do have a higher present value of demand for the store’s products than possible auction winners. The cost to the store of losing these customers is the present value of these customers’ demand. This is almost certainly higher than the foregone revenue from selling the bottle for $200 rather than through an auction.

This same basic idea about costs can be applied to the current state of the labor market. Last week, the jobs report showed that the number of net new jobs created in the U.S. was considerably below the market forecast. What made this news a bit confusing to some is that the tepid job growth occurred at the same time as a bunch of anecdotal evidence that firms are having a hard time finding workers and that many firms are offering what are effectively signing bonuses to new employees.

The jobs number immediately sparked debate among economists, pundits, and policy makers about precisely what was going on. Some argued that the generous unemployment benefits created during the pandemic are holding back job growth since some workers can earn comparable income by staying home as they would returning to work. Others argued that with schools and day cares still closed in certain areas of the country, some people simply cannot return to work because they need to take care of their children. A number of economists noted that if firms are really having such a hard time finding workers, they should raise wages.

While it might not be immediately clear, this debate is actually related to the very concepts I have discussed thus far. The debate is about why wages are not being set to clear the market. Both those who blame unemployment compensation and educational/child care issues are arguing that the problem is with the labor supply. There is not enough labor willing and/or able to work at the going wage. There is excess demand. So why doesn’t the wage simply go up?

The answer is that these are temporary factors. As our discussion of Alchian’s idea of costs makes clear, when firms offer wages to workers, they care about the present value of those wages over the course of the worker’s employment. Unemployment compensation will go back to normal soon enough. Schools and day cares will continue to re-open as vaccination rates go up and case numbers go down. Given this knowledge, why would firms offer a higher wage to workers right now when they know that they can hire the same worker at a lower wage in the near future? In other words, if the labor shortage is driven by policies that will expire soon, the market-clearing wage today is higher than it will be in the future. In present value terms, the cost of paying higher wages now relative to paying lower wages later can be quite costly depending on how durable to the employer-employee relationship will be.

If firms know that the wage necessary to clear the market is higher now than it will be months from now, one thing that firms could do is offer higher wages now and then cut wages when the labor supply starts to increase. Of course, that is hard to do. Or is it? That actually seems to be precisely what firms are doing right now! They are offering bonuses for returning to work to workers who come to back to work now. These bonuses are one-time payments. However, they allow workers to earn more income now, when the market-clearing wage is higher, than they will later, when the market-clearing wage is lower. This is no different than hiring workers at higher wages now and then cutting their wages later when the labor supply increases.

So while many people are puzzled by what is going on in the labor market, price theory and some careful thinking about costs can provide us with insights that help us understand the world.

More generally, when we ask the question as to why firms seem to be pricing things in a non-market-clearing way, the answer offered by Economic Forces always seems to point back to Alchian. If we want to better understand the world, we have to restructure how we think about costs and supply and demand toward that Alchianian approach.