What's wrong with antitrust

Markets create their own equilibrium, and government does best when it leaves the shape of that equilibrium to the aggregate choices of market systems.

John L. Ries, a guy who I've seen around ZDNet since IP packets were delivered by horse-drawn carriage, asked in a response to a recent blog post whether Microsoft is a natural monopoly. The question was posed in order to determine what ought to be done about such a situation. To paraphrase Mr. Ries, do markets in such cases not need protection because they are self-protecting, or are they not worth protecting?

"Natural monopoly" is a loaded term, and it carries along with it the idea that companies in that singular position are unassailable. So, to rephrase the definition, I would say that Microsoft is "naturally dominant," but contrary to the implications of "natural monopoly, they are vulnerable, provided competitors truly understand what it is that Microsoft provides to its customers (for a hint, read my post on software ecosystems). Likewise, calling markets "self-protecting" assumes that there is a problem with naturally-occurring market dominance. To clarify, I would state that markets create their own equilibrium, and government does best when it leaves the shape of that equilibrium to the aggregate choices of market systems. In other words, antitrust is unnecessary.

That's a rather semantically complex answer, though, and as I need little prompting to go off on random economic tangents, I'll spend the rest of this post explaining, in long form, what that paragraph means.

Free markets are smarter than managed markets. That's because each of us has more information about our own economic situation (needs, wants, resources, connections, etc.) than anyone else. A system that puts as many economic decisions as possible in the hands of the people with information is necessarily more efficient than one that places those decisions in the hands of less-knowledgeable third parties. That dynamic explains the failure of socialist versus capitalist economies, and I would consider it as fundamental a law of nature as is possible in the social sciences.

That understanding is what guides my insistence that markets which consolidate around a single vendor (or a single products, as was discussed in my post on IE's market dominance) are doing what makes the most sense given all the factors important to customers. Those factors might get lost in the tangle of costs, benefits and incentives which drive a modern global economy, but it is the right decision because billions independently voted for that outcome. So, if consumers choose to use integrated product (and note that businesses standardized on IE as well), then integrated product provides an advantage that downloaded product doesn't.

The flip side of markets, however, is a framework within which choice takes place. For economic frameworks, I look to Hernando de Soto, a Peruvian economist known for his comparative analysis of economic legal structures in rich and poor economies. He found that successful economies have well-designed legal frameworks for property and contracts that encourage individual entrepreneurship and create a safe haven for buying and selling. Since successful legal frameworks are enablers of individual economic choices, that fits nicely alongside the "markets are smart" thesis previously described. What's added, though, is the notion that markets aren't accidents resulting from anarchic individual choice, but are outgrowths of considered and goal-oriented economic polices.

That's dichotomous. On the one hand, I'm saying government needs to stand out of the way so that individuals can make good economic choices. On the other hand, I'm also saying governments need to stay engaged in order to create sound legal frameworks (albeit ones oriented around enabling individual choice).

If governments create the economic framework within which markets operate, what's wrong with antitrust?

Just as third parties have trouble making good economic choices for other people, government bureaucrats responsible for antitrust management have trouble acquiring enough information to make a sensible market solution. They MIGHT make good policies. If I hand out hammers to a room full of teenagers, they might build me a nice house. They also might start whacking each other over the head with them, or build a house that accidentally traps half the class inside.

The history of American antitrust is a tale of antitrust regulators using their antitrust hammers badly. The AT&T antitrust "remedy" was an absolute catastrophe, in my opinion, for America's telecommunications market. Often, antitrust has served to actually reduce competition.  Staples was not allowed to merge with Office Depot on the theory that it would reduce the number of large office supply stores , even though the biggest competitor to either company were fast-growing retail chains like Wal-Mart and Target (who were magically defined as "not part of the target market"). Antitrust restrictions on pre-breakup AT&T that prevented the company from entering the computer market ensured IBM executives could sleep easier at night. Other instances just seem wastes of effort, such as the 15-year pursuit of IBM without ever attaining a conviction, or the breakup of Standard Oil just as the market they dominated (kerosene) was being overtaken by oil (a market they did not control).

Governments manage pretty well when their goal is to create policies which enable individuals to make choices, but have difficulties when they start moving into the realm of outcomes. Market dominance by one company is an outcome of billions of individual choices, and governments shouldn't be in the business of second guessing that outcome. For all the wonderful elucidation power of economic theory, primitive models of ideal competition are a poor substitute for real-world choices in a global marketplace.

That said, there are some things that governments can do which are "less bad," provided they can be convinced to limit themselves to just those things. For instance, I have fewer problems with bans on exclusive contracts, required licensing of interoperability protocols, etc...all of which are part of the remedies handed down in the Microsoft case by Judge Kollar-Kotelly. I always fear government bureaucrats will take their success at changing the oil as justification for pulling out the buzzsaw and redesigning the entire automobile, but if antitrust has any value, proponents should feature Judge Kollar-Kotelly's ruling prominently.

However, I don't agree with judges (or bureaucrats) who try to design software. That's the case with the European Union, which has decided that it knows better than Microsoft's customers whether they want robust media handling functionality out of the box. Likewise, I strongly disagree with structural solutions such as the one applied to AT&T by Judge Greene or proposed (and later overturned) by Judge Jackson with respect to Microsoft. Both cases are examples of government making decisions for the marketplace rather than limiting itself to policies designed to enable individuals to make choices.

So, to summarize, when dominance occurs naturally, it isn't a bad thing, and certainly doesn't require government intervention to "fix" it. I would even go so far as to argue that dominance that isn't natural (such as state-owned telecommunications monopolies) also does not require government intervention, but I've gone on long enough, haven't I?