The laws of supply and demand are a very useful first-order way to think about what prices mean: a large enough amount of money to motivate sellers, a small enough amount to motivate buyers. But those charts exist in two dimensions, when you really want three. A price is a snapshot of the economics of a product at some moment in time, but if you're trying to reason about the economy, you care about how those prices change things, and how fast those changes take effect.

A supply and demand model gets more dynamic when there's some kind of shock that changes one of these curves: an increase in demand raises prices, which leads to higher production, or perhaps an increase in production cost raises prices until an equilibrium is reached at a lower quantity of production.¹ All easy enough, until you start asking: how long does that happen?

Consider oil. Over very long periods, oil is quite elastic on the supply and demand side. We've consumed almost exactly as much as we've produced over the course of human history, so clearly when demand rises—when we discover that an ingredient in snake oil can be refined into an illuminant, or that waste products like gasoline are a useful fuel for vehicles—oil production rises right along with it. In the shorter term, oil is famously inelastic. You just don't have that much discretion over how much fuel you consume, and it's rarely the main input. (You're spending more in opportunity cost than in gas, whether that spending is for a daily commute or a multi-week road trip.) As oil people sometimes note, nobody celebrates a decline in gas prices by spending all day driving in circles.

Over longer periods, demand is plenty elastic. One reason that the world isn't crippled by high oil demand is that cars have gotten more fuel-efficient. And that process roughly tracks price regimes, on a lag: declining efficiency in the 60s, better mileage starting in the 70s and running through the 80s, another slow period when oil was cheap in the 90s, and more improvements in the 2000s both because of price pressure (before the financial crisis) and regulatory changes (after). But even that trend illustrates just how inelastic short-term demand changes are: for cars to get more efficient in response to the oil shocks of the 1970s, GM, Ford, and Chrysler had to design them, then build then, and then the fleet of cars on the road had to slowly turn over as people either traded older cars for newer ones or upgraded. One source of higher elasticity in this case was that Toyota and the rest of the Japanese industry had been worrying about fuel efficiency much longer, and were able to step in and provide now-affordable cars that the US manufacturers hadn't bothered to make.

On the supply side, elasticity also goes through different regimes. In the postwar period, the supply elasticity of oil was extremely high: the Texas Railroad Commission required companies to throttle their oil production, keeping prices both low and stable. The mechanism for that stability was that there was plenty of willingness on the part of producers to produce, constrained only by permission to do so. That made oil supply almost perfectly elastic, a setup that lasted until the moment all restrictions were lifted and there still wasn't enough oil to go around.

If oil illustrates one extreme, with highly inelastic supply and demand in the short run, digital goods are on the other side: the supply elasticity is almost perfect, and it barely makes sense to talk about the supply curve for something like a tweet or a Spotify play—the marginal cost is close to zero, so it almost always makes sense to deliver an incremental unit of the product. At least, that’s true as long as the cost of serving a page means looking up and displaying static content, rather than running inference. That's one reason these companies have economics so different from the Econ 101 model of a company, or from other businesses: they're playing by the same rules, but have extreme variance in capabilities, a bit like competing in a marathon while driving a Ferrari. (As with the Ferrari, they're paying a high fixed cost for this capability.)

Elasticity matters a lot when looking at products with physical constraints, especially cases where there's some part of the supply chain with less elasticity than others. The difficulty of manufacturing PPE at scale early in the pandemic often came down to specific components that were only produced by particular manufacturers. As with many aspects of life,  our experience of this phenomena tends to reflect something in the middle—your favorite cheap new restaurant evolves into a mid-priced one once they know customers like it, your job gets a bit less lucrative once other people start picking up the same skills. But the really interesting parts are at the tails.

Read More in The Diff

The concept of elasticity runs through many Diff pieces, especially on supply chains, price levels, and how quickly companies can and can’t pivot.

• We’ve covered the supply chains of bits and atoms.

• We looked at whether hybrid work makes oil demand more elastic ($).

• In the early days of the pandemic, we considered how brittle supply chains can be.

• One source of wonky elasticity: collusion, whether algorithmic or otherwise.

• A classic asset with inelastic supply is gold.

Join the discussion!

1  There are a handful of exceptions to this, some more interesting than others. The boring exception is that for some products, price is a signal of value—if you price a luxury product halfway between the usual list price and the cost of a knockoff, your prospective buyer assumes it's an unusually expensive knockoff and refuses to transact. The more interesting case is the Giffen good: if a product is a large share of the consumption basket, and is cheaper than a higher-quality alternative, price increases can actually increase its consumption because consumers can't afford the better alternative. In the real world, the best examples of this are typically staple food products in extremely poor countries; if you can only afford to eat meat once a week, an increase in the price of wheat, rice, or potatoes will increase your consumption of those foods, since you won't be able to afford meat at all. In the developed world, consumption baskets are diverse enough that this doesn't really hold.