It's not price-only theory.

Extending price theory to better reflect pricing behavior

You are reading Economic Forces, a free weekly newsletter on economics, especially price theory, without the politics. Economic Forces arrives weekly in the inboxes of 13,000 subscribers. You can support our newsletter by sharing this free post or becoming a paid subscriber:

---

As the father of a soon-to-be-school-age child 👴, I recently found myself frantically pressing refresh on my school’s website when after-school care registration opened at 8:00 on a random Tuesday. I was frustrated with the process; I doubt I was alone. My immediate reaction was to holler, “Why not use prices to allocate the scarce slots to those who value them most?” (Paraphrasing, and no, my kids did not respond or even acknowledge my question.) The process seemed inefficient and even unfair in making parents compete on speed and flexibility rather than willingness to pay.

One reaction is to say, “The school district has no profit motive, no residual claimant.” That’s true. But that’s not the whole story.

Most markets don’t use only prices to allocate resources. Think about the last time you tried to get concert tickets or make a reservation at a popular restaurant, as Josh has written about. Chances are, you didn't get the opportunity purely based on your willingness to pay. There wasn’t an auction. Factors like speed/waiting, loyalty, and personal relationships often play a larger role. Prices don’t adjust that much.

So much for that price theory then, huh? What a waste of time!

In fact, when you look closely, most markets don't rely solely on prices to clear supply and demand. From airline tickets to iPhones to industrial equipment, firms use a variety of non-price levers to allocate goods and services. They develop long-term customer relationships, strategically manage inventories, and leverage strong brands to maintain stable prices despite market fluctuations.

When prices alone clear markets

Before diving into the role of time and dynamics, it's worth reviewing how each of the simple, static models handle price adjustments. Here I’m drawing on an old chapter from Dennis Carlton in the Handbook of Industrial Organization for my framing on the huge topic. These textbook models—perfect competition, monopoly, and oligopoly—form the baseline for understanding market behavior and pricing.

In the perfectly competitive model, price is the sole adjustment mechanism. If demand increases or supply decreases, price rises until quantity demanded equals quantity supplied. If demand decreases or supply increases, price falls until the market clears. The model assumes that these price adjustments happen instantaneously and without friction. There are no shortages or surpluses, no inventories, and no long-term relationships. Price does all the work of matching supply to demand.

The monopoly model introduces the concept of pricing power, but the basic price adjustment story is similar. The monopolist sets a price to maximize profits given their demand curve. If costs increase, the monopolist raises price until marginal revenue again equals marginal cost. If demand shifts, the monopolist adjusts price to re-optimize. The key difference is that the monopolist intentionally sets price above marginal cost. But price is still the primary mechanism for responding to changing market conditions.

Oligopoly models can get a bit more complicated and introduce strategic interaction between firms, creating a wider range of possible outcomes. Still, across most these oligopoly variations, however, prices still adjust to clear the market. The oligopolists may consider rival reactions and long-run incentives, but price remains the key variable firms use to balance supply and demand. If costs rise, oligopolists raise price, although the magnitude may differ from the monopoly or competitive case. As I like to point out, most of the predictions in price theory don’t depend on competition vs. monopoly vs. something more complicated.

In summary, the simple, static models all rely heavily on prices as the mechanism for market clearing. Prices adjust, either competitively or strategically, until the quantity demanded equals the quantity supplied. This adjustment is assumed to happen quickly and completely, without frictions or outside factors.

However, empirical evidence tells a more complicated story. Studies consistently find that prices change far less frequently than the simple models would predict. For many goods, prices remain unchanged for months or even years at a time. Even in industries with relatively frequent price changes, we often mean a daily basis, not continuously.

This price stickiness is difficult to reconcile with the instantaneous adjustments predicted by the basic models. If prices were truly flexible, we would expect to see them changing constantly in response to every small shift in supply or demand. The fact that we don't suggests that there are significant costs or frictions associated with price changes.

One option is to just say that “menu costs” make it costly to change every moment. That’s fine if you’re not actually interested in price changes but what to use those lack of price changes to explain something else. But if you’re interested in pricing, I think you can’t take menu costs as a given but should derive that from something more basic. It’s Wallace’s Dictum for monetary theory applied to price theory.

Time matters

To understand these departures, let's walk through how the simple models change when we add a crucial element: time. Not just passive time, where people are being hit by “shocks,” but where the people in the model think about time.

In the competitive model, each good at each point in time is its own market. How does the price today interact with the price tomorrow? As always we need to look at supply and demand. On the supply side, firms can choose when to produce and sell, maybe using inventories to shift supply between periods. Demand depends not just on today's price but on expectations about future prices. On both sides of the market, both buyers and sellers can smooth their choices, which will smooth prices.

Consider a temporary spike in demand. In the static model, this requires a sufficiently large price increase to clear the market. But in a dynamic setting, buyers may just wait longer for delivery while sellers ramp up production. The price increase can be much smaller, spread out over time.

Now, let's look at the monopoly model. Everything above about smoothing applies to monopolies as well. But there's another force. A monopolist doesn't just care about today's demand but also about how today’s price affects future demand. The competitive firms don't affect prices so do not take this into account. Carlton uses the example of steel scrap. If the monopoly scrap supplier hikes prices, downstream steel producers might scale back their plans for new furnaces. That's a hit to future scrap demand that the scrap seller must consider. Raising prices now might encourage substitution away from the product later, giving the monopolist an incentive to keep prices more stable than the static model suggests.

Inventories matter

Once we take time seriously, we need to take uncertainty into account. Time and uncertainty open a role for firms to hold inventories, which is again missing in the baseline models. Inventories were implicit in the discussion above but let’s make it explicit. Plus, this is a good time to cite Alchian (1969) on information costs. If you exactly knew tomorrow’s demand, you wouldn’t need inventories.

But firms don’t know who will show up at their store. By allowing firms to smooth production, buffer shocks, and intertemporally substitute, inventories enable more gradual price adjustments than the simple model predicts. In the classic production smoothing model, firms use inventories to maintain steady production despite demand fluctuations, avoiding costly over- or under-production. Inventories also allow firms to shift supply between periods in response to changing costs or demand expectations, without necessarily changing prices.

Perhaps most importantly, inventories serve as a shock absorber. An unexpected demand spike can be met by drawing down inventory, letting firms spread the necessary price adjustment over time as they replenish stocks. This helps explain why prices often respond gradually to shocks rather than jumping to a new equilibrium.

Inventories also interact strategically with market competition. Monopolists must balance current profits against future inventory levels, while oligopolists may use inventories to facilitate or undermine collusion. Inventory allocation can also serve as a non-price rationing mechanism, with loyal or large customers prioritized during shortages.

Finally, inventory levels provide a window into firms' expectations, rising with optimism and falling with pessimism. This makes inventory data a closely-watched economic indicator.

The ability to hold inventories varies widely across industries, contributing to heterogeneous responses to shocks. But overall, incorporating inventories is key to bridging the gap between simple models and actual market behavior. By allowing more gradual price adjustment and intertemporal substitution, inventories are a crucial lubricant for the complex dynamics of real-world markets.

Pricing isn’t costless

Up to this point, the discussion might make solutions like inventory seem like a second-best option. There’s some “problem” out there, and so we are stuck with inventories instead of prices. There’s some truth to that framing, but it’s not the only way to think about it.

Markets, in the sense of an explicit spot where trade happens and not an abstract idea, are costly. It's costly to create a market that clears solely based on price. You need physical and institutional infrastructure - think of all the resources the New York Stock Exchange consumes, or the time buyers and sellers spend haggling. Add in more product heterogeneity and that makes price-only clearing harder.

Most markets aren't always centralized and anonymous. Buyers and sellers form relationships, investing in understanding each other's needs. This lets sellers allocate goods based on factors beyond just price—loyalty, past sales, expectations of future business. Prices can differ across buyers based on these softer factors.

So where does this leave us? With a much more complex (and realistic) view of market dynamics than Econ 101 would suggest.

To be clear, none of this invalidates the use of simple models as a starting point. Supply and demand is still an incredibly powerful framework. The art is knowing when the simple model is enough, and when you need to layer on more complexity. As we continue to develop richer theories and dive into the institutional details of actual markets, I'm excited to see how our understanding evolves. The real world may not match the textbook models—but that just means there's more for us to discover.