Why Not Two CIAs?

Sometimes I get an idea that I think is very good, and a day later I no longer think its so good. Here is an example of post and a quick revision. Other times, I have an idea that I think is good at the time but later I think maybe that its better than that. An example might be my idea for a new kind of exercise bike. I love that idea; I would buy a bike like that. Another example might be my analogy that I made in a post two days ago about choosing your own budget. I tepidly liked the idea but I thought that it would not help make government more efficient because we don’t get to choose a better government technology:

However, the main issue I have with this proposal is that it would not address the fundamental problem with government spending: we don’t have a plan B. In the free market, if I don’t like the information the Washington Post provides, I can always buy the New York Times. However, if I don’t like the information the CIA provides (and who does?), then I cannot opt to spend my taxes on CIA2. We can hope the government reforms CIA1, but CIA1 is the only choice we have, except for not having a CIA altogether. Free markets duplicate organizations so that if one becomes dysfunctional, we can shift to a more efficient version. We never get that choice with the government. So while I kind of like the idea of having some choice about how my tax dollar should be spent, unless I actually get some choices, all I have is Hobson’s choice.

In the quote, I made the analogy between being able to choose between two different newspapers for our news and choosing between two different CIAs for our intelligence information. When you think about it, these are very analogous organizations. The ideal CIA would be like a great newspaper that just published its information with a secret or top-secret classification. The problem with the CIA is that a lot of the information it sends out is wrong. If the New York Times regularly prints garbage information, other newspapers will instantly show the world that the New York Times is wrong, and no one besides the New York Times suffers from the error. Unfortunately, we only find out that the CIA was wrong after we bomb the Chinese Embassy in the former Yugoslavia or after we invade Iraq and find no WMD’s.

So here is the obvious question: "Why not have two CIA's?" If we had a CIA1 and a CIA2 we would have two chances of getting those pieces of information right. News is too important a commodity to rely on only one newspaper and intelligence for the leading superpower is too important a commodity to rely on only one source as well.

However, there is a more important reason why two bureaucracies are better than one. With two bureaucracies, we have two chances of having at least one functional bureaucracy. Some bureaucracies degenerate into dysfunction. Reforming them is nearly impossible unless you are willing to fire almost everyone.

Bureaucracies tend to reinforce a particular groupthink. People who don’t exhibit that groupthink aren’t recruited into the bureaucracy or get discouraged within that bureaucracy and leave. It happens in the corporate world as well, but there the disgruntled employees can form their own firms and prove the old groupthink wrong. In government, there is only one groupthink per “industry” so the technology never improves.

We might be able to use the idea of “dueling bureaucracies” in other government agencies besides the CIA. However, an idea like this would need a proper trial. The CIA gives that opportunity for experiment. There is no better time to start the experiment than right now.

Back to homepage.

Also, check out my "Chocolate Shop," my favorite posts.

Chocolate Shop

• The Exercise Bike


• He Only Drinks American Beer


• Lottery Technology


• Free Food: Why Do Americans Love It?


• Too Much Ravioli


• Optimal Exorcisms


• Thanks, Amit