Bryan Caplan  

The Age of Contestability

Dinner with James Manzi... Another Take on the Soap Opera...
When economists want to measure the competitiveness of an industry, they usually start by counting the number of competing firms.  If they see a lot of firms, they infer a lot of competition.  Few firms?  Little competition.  One firm?  No competition at all.

Three decades ago, a model came along to challenge this mindset: the contestable market model.  In the contestable market model, a single firm often behaves exactly like a perfectly competitive firm.  Why?  Because of potential competition.  Sure, your firm may be the sole supplier of a good right now.  But if you raise your price above the competitive level, all sorts of competitors might suddenly spring up.  To prevent this from happening, you need to keep your prices so low than no potential competitor bothers to become an actual competitor.

The contestable markets model enjoyed immediate attention.  But as you'd expect, believers in the traditional structure-conduct-performance model were not impressed.  It doesn't google, but I still remember Leonard Weiss's summary dismissal.  It went something like this: "The contestable markets model is empirically irrelevant and should play no role in policy."

There hasn't been much new research on the contestability model in recent years.  But I recently realized that the empirical and policy relevance of this much-maligned model is now very hard to deny. 

Consider: The internet has given us a long list of near-monopolies: Amazon, Apple, and Netflix are only the beginning.  Funny thing: consumers love these "monopolies."*  Despite their market dominance, these firms deliver amazing products at low prices.  And almost no one is "waiting for the other shoe to drop."  If someone claimed that Amazon, Apple, or Netflix were planning to gouge us as soon as they devour their last competitors, we'd laugh.

The contestable markets model has a simple explanation.  Sure, Amazon, Apple, and Netflix look a lot like monopolies.  But the people who run these firms can feel their potential competitors breathing down their necks.  If these "monopolies" start taking their customers for granted, they'll quickly cease to be monopolies.  Indeed, they may quickly cease to exist at all. 

In fact, the results we see are better than the simple contestability model predicts.  Amazon, Apple, and Netflix don't just keep offering customers the same attractive terms.  They're constantly trying to improve the terms they offer.  Why?  Because they realize that in the Age of Contestability, market leaders will fall behind unless they keep running full speed ahead.

* Yes, I know that Netflix recently burned up a lot of good will with some bad decisions.  But consumers still love the product.

Comments and Sharing

COMMENTS (22 to date)
Steve S writes:

I would argue that Netflix's quick turnabout of their "Qwikster debacle" is further proof of the contestable market theory.

As soon as they pulled the ultimate monopolist maneuver (reducing choice and convenience, increasing price), they were reminded that they weren't the only show in town, lost tons of customers, and changed their minds just as quickly to remain the market leader.

Josh writes:

What about Google? (Or Microsoft in the 1990s, for that matter). If I didn't know anything else about the legal context, I'd think they'd fall into the same camp as Apple or Amazon, yet there's a much higher likelihood of antitrust enforcement against them.

Jason Collins writes:

Although I agree with the general point, I am not sure that these examples provide strong evidence. Apple has around one-third of the smart phone market and a fraction of the personal computer market. When I buy a book online, site aggregators provide me with a mountain of quotes from competitors besides Amazon - and Amazon and Apple are competitors in the book market. I would not describe them as monopolies (or even near monopolies).

Left Outside writes:

It looks like there is a strong link between contestable markets and low entry costs.

So in low skilled services and manufactures, or internet retailers we should be less worried about monopolies, but in other circumstances contestable markets may not be a good model.

Julien Couvreur writes:

Are you becoming Austrian? ;-)

Some more examples: Microsoft (OS space and productivity software), Google (search and ads), AOL (used to be internet giant), MySpace (overtaken by Facebook), Intel, Sony, etc.

In all those cases, there are *high costs of entry* (developing an OS or a search engine is expensive, building critical mass for a social network is difficult, ...).
But even though each of those firms temporarily had (or have) a very dominant position they clearly are not secure like the monopoly model suggests (abuse your customers since they have nowhere to go).

Ed Bosanquet writes:

I have to agree with Jason Collins. I'm not convinced this concept actually occurs today, I would love for it to but it seems for each dimension of the companies listed business there are competitors. Many of the companies you listed compete with each other. Apple and Amazon compete on tablets, app stores and e-books. Amazon and Netflix compete on streaming video. There is healthy competition in the tablet and smart phone market. (Apple has claimed the high end for the moment).

Each one of these companies is a different bundle of services and thus doesn't have a full on direct competitor but in each dimension there is competition. Google and Microsoft are two other companies that offer some of the same services. Once again not bundled in the same way.

Thank you,

Zubon writes:
If someone claimed that Amazon, Apple, or Netflix were planning to gouge us as soon as they devour their last competitors, we'd laugh.
See, of course, the recent French lawsuit against Google maps for giving away its corporate API, explicitly on the theory that Google would drive out the competition and then gouge.
Steve Sailer writes:

People barely complain these days about annoying monopolies like Ticketmaster.

Robbie writes:

Another example may be sporting competitions.

Until the 1990s, the Australian Rugby League (ARL) competition was a monopoly. But it was in a contestable market, as demonstrated when the Super League was founded and attracted a large number of star players with high salaries. The two rival leagues eventually negotiated and formed a new monopoloy which continues today, the National Rugby League (NRL).

A similar story occurred in the 1970s when Kerry Packer founded World Series Cricket, which (again) attracted a number of star players from the traditional Test Match competition, which had previously been a monopoly - proving that it had actually been a monopoloy in a contestable market.

There is currently a dispute between two high profile team owners in Australia's soccer competition, the A-League. One of the owners is threatening to start a new rival competition. The A-League is a monopoly (at least, for now), but obviously participating in a contestable market for soccer competitions.

Dave Huntsman writes:

First off, your 'model' may be missing the larger lesson. Even more important than the number of current competitors - after all, whole industries often go uncompetitive - is the existence of new entrants; the ability for there even to be new entrants is critical, including to this 'model'. Amazon isn't that old; neither is Apple in its current configuration. And the 500 pound new entrant of course is google, given short-shrift here.

A Country Farmer writes:

A well timed post because Amazon just cut their prices on their cloud computing services which power a lot of websites (called EC2).

J_D_L writes:

People worry about Netflix-as-monopoly? I have a Netflix account, but during the past three weeks I've pulled three movies from local Redbox kiosks and a series of 8 discs from blockbuster. I see several services with different delivery strategies, and content which only partially overlaps. Since I'm only actually a billed member of Netflix, I can use the other two services to cover N's weaknesses, coming and going as I please. What Redbox doesn't license, I pick up at Blockbuster. What Blockbuster doesn't carry, Netflix mails to my house. Unprecedented selection and accessibility.

Navin Kumar writes:

Is this only 30 years old? Schumpeter said something very similar. His writings on monopolies is sadly forgotten.

RPLong writes:

I'm with Julien and Navin. The first time I read something like this was in Mises' Human Action.

Joe writes:

My goodness. In what world are Amazon, Netflix and Apple near monopolies? Amazon has hundreds of competitors. There are segments in which they have pricing power, but they are still focused on growth so they don't use it. Their strategy is not defensive at all.

Netflix continues establishing a new segment in the very competitive entertainment sector. They will accrue pricing power over time, but we'll have to wait to see what they do with it.

Of the three, Apple has the most pricing power in their core segment, and they use it. Their margins are absolutely incredible compared to those of the down-market offerings of their competitors. They in no way practice limit pricing.

Your model might be right, but those examples don't make the case at all.

Adam Thierer writes:

Bryan.. Interesting post. You might be interested in a piece I penned on this exact issue for Forbes a few months back. It was entitled, "Of 'Tech Titans' and Schumpeter's Vision."

Cheers -- Adam Thierer, Mercatus Center

Richard Manns writes:

I'm not exactly an economist,

But the most obvious factor of the Internet age has been the tearing down of the advantage of capital.

The net effect, if I'm right, is that companies are forced to be competitive if and only if that is the only way of maintaining themselves. Where it isn't, for whatever reason, they won't.

Further, the same company might behave competitively where it has to, and non-competitively where it doesn't. Apple is actually a good example of this; I suspect the iPhone 4S would have been a better product, had Apple not been able to rest upon its capital of millions of devotees.

Dan Carroll writes:

I would describe many tech markets as oligopolies, but agree that the definitions of those markets is subjective. Google clearly has a monopolistic position in search, with economics similar to the old airline ticket reservation systems, held in check only by the willingness of Microsoft to lose money. Apple is not a monopoly when the market is defined broadly, but when their phones are examined against their addressable markets, they win about half the time, with the other half going to a free product. Then there is music players. Amazon dominates online shopping distribution and fulfillment, but obviously not retail in general. So is it a question of substitute markets?

The problem with the traditional monopoly as taught in economics 101 is that it assumes that monopolies raise the price of the product above the rate at which a market with multiple competitors would hold. However, most monopolies exist at least in part because of economies of scale, and actually are able drive down the price below the multiple competitor equilibrium. Thus, they can undercut on prices and still earn "monopoly" profits. Call it barriers to entry, but it is really just a function that bigger is lower cost. So, in the extreme, a firm with 51% market share can undercut a firm with 49% share on price.

Even where networking effects predominate (Google, Microsoft), the value of the network to users far exceeds any price premium they might pay, though even in those examples it is not clear that they actually pay much of a premium - even with Apple, hardware quality and other factors mitigate the perceived price premium: tablets offered at similar price points to the iPad get creamed in the marketplace.

Jerry Ellig writes:

The contestable markets model is about entry of an identical competitor when there are no barriers to entry -- NOT entry by someone who figures out a way to do it better or at lower cost. When I read about the evolution of tech markets, it sounds a lot more like what Schumpeter described as "creative destruction" -- entry by a competitor who has a decisive cost or quality advantage over the incumbent firms. If the new competitor is innovative enough, it can be a threat even if there are "sunk costs" that would normally deter entry. Contestable markets theory does not predict this kind of competition because it assumes identical firms. And that's the difference between contestable market theory and more Austrian theories of competition.

Brian writes:

What are you talking about? When it comes to Apple they have a few to many competitors in almost all their product lines.

Apples does not have the cheapest product in any of its markets. The reason people buy Apple is good looks, great software, and religious devious to the brand, all at a high sell price (although not the most expensive). Long lasting hardware really is not a selling point other than its poor quality can add negatives to the overall image of a product.

However, people feel they get quality with Apple so they pay the higher price. The only real contestant quantitative high quality of Apple products when one takes the superficial away (design looks, and brand image especially as a status symbol) is superior software. However, products like iPods and itunes which were once the best in the market are no longer the best on the market. Smart phones are the same although they are still top tier which I would not say about a few other apple product lines. Apple is the most valuable company in the world because of software and genius marketing that multiples the value in the marketplace of the experience (which is almost all from software). However, they have always made their money off the hardware.

Dan Carroll writes:

Apple's meaningful competitors are:

iPhone - the OS is the operative market definition, as the hardware is more or less interchangeable: Android, Blackberry, Windows. RIM is in decline, Windows has tiny market share. Other smartphones are non-entities. That leaves one competitor, which probably only exists because it is a free OS.

iPad - Android, Kindle (Fire is based on Android), and (soon) Windows 8. Windows 8 is not out yet, Android tablets have small share, leaving Kindle as their primary competitor in a bifurcated high-low market.

iPod - Zune and iPhone. Nuff said.

Macs - Windows. Microsoft is the monopolist, but Macs skim off the top of the PC market.

Apple may make money off of hardware (in a tax-evading accounting sense), but their profits only exist because of the software.

As far as quality - Apple products have very tight integration between software and hardware (most apps are native), which yields performance advantages, but is a higher cost proposition due to the closed architecture. Durability in phones is not really an advantage; with tablets it remains to be seen.

However, one can point to substitutes: tablets compete against smartphones and PC's. Smartphones compete against dumbphones. However, usage patterns suggest that most people buy and use all three: PC's, tablets, and smartphones.

Bob Smith - Dallas writes:

I took a Anti-Trust class in grad school years ago, and the US vs Alcoa case came up. The professor said that when the US Supreme Court ruled for the break-up of Alcoa that the justification was that Alcoa was so efficient, that it could sell its aluminum so cheaply, that no competitor could sell for less and still be profitable. If that's true, then it strikes me that this would support the model's thesis.

Comments for this entry have been closed
Return to top