A student forwarded to me a Washington Post article by economist Everett Ehrlich, which purports to use Nobel laureate economist Ronald Coase's 1937 article The Nature of the Firm to explain the rise of Howard Dean:
Back in 1937, an economist named Ronald Coase realized something that helped explain the rise of modern corporations -- and which just might explain the coming decline of the American two-party political system.
Coase's insight was this: The cost of gathering information determines the size of organizations.
Ehrlich then goes on to claim that the Internet has allowed mega-corporations to shrink, because the costs of gathering and transmitting information have gone down so dramatically:
Firms have become "flatter" and "faster," and the "networked" or "virtual" company has come into being -- groups of firms that use shared networks to behave as if they were part of the same company. ... So the end result of the Internet revolution on companies has been exactly what Coase's theory predicted: Cheap information has allowed firms to shrink.
Now comes the payoff, as Ehrlich claims the same thing will happen to political parties:
Coase's ideas are no less true for political organizations, as Dean's success shows. He is the first candidate to use the Internet effectively as a political organizing device....
The ability to have "virtual political parties" is the greatest challenge the two parties have ever faced. There are strategies available to them, of course -- deft positioning allows them to preempt competitors, as it does in every industry, and they can use the same technology, although Internet culture doesn't seem readily amenable to either Democrat.com or Republican.com. Being a Democrat or a Republican isn't enough of an advantage anymore -- there are simply too many other places where people can get political information and find political bedfellows in an age of low information costs.
My take? It's cute, but doesn't work. Ehrlich should leave the econo-editorializing to Paul Krugman. Problem # 1: Coase wasn't really concerned with the size of firms. He was concerned with the decision to conduct economic activity across a market or within a firm (the so-called "make or buy" decision). Coase explained that firms emerge when it is efficient to substitute entrepreneurial fiat for the price mechanisms of the market. In markets, resources are allocated by the price system, while in firms resources are allocated by authoritative direction. This was Coase’s fundamental insight: “If a workman moves from department Y to department X, he does not go because of change in relative prices, but because he is ordered to do so.” Accordingly, economic activity will be conducted within a firm when the costs of bargaining exceed those of command-and-control. From this perspective, Coase's argument doesn't look like a very good analogy for political parties does it? Problem #2: Information costs are only one of the transaction costs that explain why some economic activity is conducted within firms and some is conducted outside firms. The make or buy decision thus is impacted by such things as uncertainty, complexity, bounded rationality, opportunism and shirking, collective action problems, bilateral monopolies, and, especially, asset specificity. (I described these various costs and their relevance to the formation of business firms in my article The Board of Directors as Nexus of Contracts.) The internet reduces only one of these costs -- information gathering -- while introducing new costs of its own; namely, the necessity to filter and verify the information one gathers from the internet. Problem #2A Ehrlich is right that many mega-firms have downsized in recent years, but that downsizing cannot be attributed solely -- and perhaps not even principally -- to reductions in information-related costs. There are a ton of other relevant transaction costs that go into these decisions, such as the shift towards employment that emphasizes general rather than firm specific human capital. (If you're interested in my take on the transaction cost story about hierarchy and the flattening of organizations, see Privately Ordered Participatory Management: An Organizational Failures Analysis.) Macro factors like globalization of labor and capital markets also have been important. Again, when considered in this light, Ehrlich's analogy begins to break down doesn't it? Problem # 3: Businesses have a strong economic motive to reduce information and other transaction costs. People don't, especially when it comes to politics. A rational voter will expend effort to make informed decisions only if the expected benefits of doing so outweigh its costs. Even with Florida 2000 as a counter-example, rational voters know that their one vote will not be determinative of the outcome of any given election. So long as the effort and opportunity costs associated with making an informed voting decision are non-zero, voting is an economically irrational act. (I vote most elections, of course, which just goes to show that you can't use rational choice theory to explain the behavior of even lawyer-economists!) This is why I never complain about low voter turnout. Those who do not vote are being rationally apathetic. Political parties are an important way of reducing the transaction costs that produce rational apathy. When I vote, there are usually many issues on the ballot about which I know nothing. So I use decisionmaking shortcuts. Under conditions of uncertainty and complexity, boundedly rational decisionmakers are unable to devise either a complete solution to the problem at hand or fully assess the probable outcomes of their action. In effect, cognitive power is a scarce resource, which the inexorable laws of economics tell us decisionmakers will seek to allocate efficiently (to the best of their ability). Consistent with that prediction, there is evidence that actors attempt to minimize effort in the face of complexity and ambiguity. One way in which they do that is by using decisionmaking “heuristics.” For example, out here in California we face many ballot initiatives. My heuristic for voting on initiatives about which I know little is to vote no. I just assume they're a bad idea. I may be wrong, but the transaction costs of making an informed decision just aren't worth it to me. Now we come to the function of political parties. I usually know enough to make a decision about President or Senator or maybe even Congressman without having to resort to a heuristic. But then you get to stuff like State Assemblyman or, worse yet, Sewer Commissioner. I have no idea who these people are and there is no way I'm going to incur the opportunity costs associated with finding out! So my heuristic is to vote for the candidate of my favorite political party (guess which one that is). I assume that a person nominated by my party will have values roughly consistent with mine. Party affiliation thus is a signalling device, which reduces the transaction cost of voting by eliminating the need to know anything else about the candidate other than there mere fact of party affiliation. My heuristic may lead to bad decisions in some cases, but how much damage can a Sewer Commissioner do to my quality of life anyway? Besides which, how likely is it that my one vote will decide who gets to be Sewer Commissioner? The internet does nothing -- nada, zilch -- to solve the transaction costs that are solved by political parties. Why? because I'm not going to go on the internet to look up information about a sewer commission candidate and neither are you. In fact, let's be honest, we're not going to do it for State Assemblyman. How about Congressman? Maybe. But why bother given how gerrymandered Congressional districts are these days? Ehrlich may be right that the internet will make it easier for outsiders like Dean was a few months ago to capture a party's Presidential nomination. But as long as there are sewer commissioners to be elected there will be political parties. Or, as Arnold Kling put it: "I'll believe that the structure of politics has changed the day that someone can get elected to the local School Board or County Council without the endorsement of the teacher's union." UPDATE: Mickey Kaus writes:
Bainbridge [identifies] a specific way parties minimize costs--by eliminating the need to investigate the views of every Sewer Commissioner or State Assemblyman. (Instead, as noted, you can just rely on the party label.) But parties do a lot of other things that are made a lot easier by the Internet--like hooking like-minded people together in the first place and raising money from them and mobilizing them rapidly around the issues of the day. The Dean campaign does exist, after all. It can't be theorized away. And it could not have done what it has done without the Internet. Even Bainbridge says, at the end of his post:
Ehrlich may be right that the internet will make it easier for outsiders like Dean was a few months ago to capture a party's Presidential nomination.
'Only the presidency' seems a pretty big 'to-be-sure' concession. Ah, but remember the less glitzy offices, says Bainbridge "[A]s long as there are sewer commissioners to be elected there will be political parties."
Even at that level, will they be traditional parties or Dean-like Insta-parties? My guess is the Insta-parties will get down to the Sewer Commissioners soon enough. And if they don't ... how important are they if they can't elect a president?
Sorry, but I don't buy it. Only a hardcore political junkie would think that people will ever use the internet to study up on lower-level offices. The problem is that most of the people participating in this debate spend a lot of time on the internet writing or reading about politics and even when they're not on the internet they're thinking about politics. In contrast, I doubt whether most people - including most voters - are willing to give up watching Survivor, surfing porn sites, or playing Grand Theft Auto in order to use the internet as a source of political information. (Confirmation of that intuition may be found here.) If these people bother to vote, the party brand is critical. Indeed, one point I did not make explicitly in the original post was the function of parties as a brand. People develop brand loyalty as a heuristic -- always buying Tide saves you from having to make an informed decision every time you go to the store. Political parties perform the same branding function. Remember the old Texaco ad slogan: "You can trust your car to the man who wears the star"? That's what political parties do: You can trust your sewers to the man who wears the Democrat or GOP label. Who besides people like Kaus (or, admittedly, the people most likely to be reading these remarks) is going to bother looking past the label on down ballot slots? The other party functions Kaus identifies -- like raising money and so on -- are just incidentals to the branding function. In response to Kaus, the key question is whether the party's branding function has value at higher levels, like the presidency. Why is Howard Dean trying so hard to win the Democratic nomination? Because it has brand value. Even at the Presidential level, brands matter:
[Carroll Doherty, editor at the Pew Research Center] estimated that perhaps 15 percent of the electorate is up for grabs, though other experts say it's closer to 7 to 10 percent. The number of undecided voters is shrinking as people line up with the party that best represents their views, Doherty said.
In other words, people are choosing a brand and sticking to it. Why? Because it's easy. The brand label saves them the trouble of having to make an informed decision, which reduces the effort and opportunity costs of voting. Brand loyalty is a very powerful factor in decisionmaking; powerful enough that I doubt the viability of "Insta-parties." BTW, age matters a lot in whether or not people vote. Older folks tend to vote at much higher rates than younger ones. Why does that matter to my argument? Because brand loyalty increases with age:
Studies from sources including Advertising Age and the International Advertising Association indicate that consumer-buying habits of men under 35 are malleable and more receptive to advertising-inspired brand switching. (After 35, these studies say, brand loyalty is set in everything from deodorant to cars.)
Look at it this way: Howard Dean has used the internet to assemble a coalition of just under 540,000 contributors (as of today). Al Gore got 51 million votes in 2000 -- and still lost the Electoral College! Dean has proven that the internet provides a useful set of tools for capturing the party's nomination. He has not proven that you can use those tools to get from half-a-million supporters to 51 million voters. My guess is that you cannot get from half-a-million to 51 million without the benefit of a well-established party brand. When a third party candidate proves you can use the internet to win the presidency without the benefit of a major party brand label, then I'll be impressed. Bottom line? I would be willing to place a Long Bet that every President between 2004 and 2024 is either a Democrat or a Republican.
