I have long held a theory about smart people that goes something like this: When someone comes up with a good idea, he/she eventually comes to a point where he/she attempts to use that idea to explain everything. There are a lot of good examples of this: Max Weber and his power-oriented theories of social interaction, Steven Landsburg's belief that the universe is made of numbers, Scott Sumner's obsession with NGDP, and so on.

There is at least one glaring problem with my theory: It's not inclusive enough. Clearly, if something like this is true of very smart people, it is probably also true of everyone else.

Two Theories On Thought-Bubbles
Arnold Kling sees the point, too. This morning, he begins by quoting an advocate of leftism, who believes that the crumbling of the political right in America is inevitable. To this, he contrasts the view of a rightist who feels exactly the same way about the political left. Finally, he says, "At least one of these folks is living in a bubble." Then he plays one of my favorite games: Assume that a theory is true, and imagine what that must mean...
What if we have reached a point where the scale and scope of government have become absurdly large? What you would observe is a growing gap between the theories used to justify government expansion and its practical impact. You would observe the cost of education and health care rising, without commensurate benefits. You would observe stimulus programs that increase employment according to computer models but not in reality. You would observe crony capitalism. You would observe budgets distorted by public-sector unions. You would observe fraudulent accounting that shifts costs for pensions onto future generations.
Kling also quotes a recent article by Lawrence Summers, which also argues for the inevitability of a big-government leftist utopia and serves as a contrast to Kling's own thoughts on the issue. Lubos Motl does an even better job of taking Summers down, including this among the many reasons why Summers is just plain wrong:
Summers' decision to include the demagogic reference to the interest rates is an example of what the Jews call "chutzpah". He is using the expected rise of the interest rates as a justification of a growing government. Interestingly enough, just a year ago or so, he would use the low interest rates as a justification of policies to increase the government, too. He would be saying that the interest rates were really low so no one needed to care about them and the government shouldn't be afraid to buy ever higher amount of money because things are OK.

Now, when it seems as though the era of cheap money is coming to its end anyway, he is using exactly the opposite observation to justify... the same policies to grow the fat spoiled brat known as the government, too. Why is he including all the irrelevant detailed numbers in his articles, pretending that they are arguments in favor of something, when it must be damn obvious to him that the true reason why he will advocate an ever bigger government is that he simply wants it, so he wants everyone else to believe that everything else must adapt to the "boundary conditions" that include the "fact" that the government has to be increasingly big? It's what I call demagogy.
Heads I Win, Tails You Lose
Of the two perspectives, I think Kling is a little closer to being correct. I don't think Lawrence Summers is knowingly engaging in demagoguery. I think Summers genuinely believes the economic evidence lines up in his favor. But that is the whole problem: When a data set happens to substantiate your theory no matter which direction it moves, then you know you have a bad theory.

It's the same with climate science: Warmer temperatures mean climate change; colder temperatures mean climate change; more extreme weather means climate change; less extreme weather means climate change. In short, everything means climate change.

It's also the same thing with Scott Sumner's NGDP targeting theories: No matter what happens in the economic data, he refuses to consider any data point other than NGDP, and no matter what happens to NGDP, he takes it as evidence buttressing his own theory.

And with leftism in general: Any major problem is evidence of insufficient regulations, including the failure of existing regulations!

And with rightism in general: Any random act of violence is terrorism, which requires the elimination of freedom; any absence of random violence is proof that eliminating freedom works to reduce violence!

These guys are all stuck in their own heads, unable to escape. It's not demagoguery, it's just a vast bubble from which they will never emerge.

Does Your Theory Specifically Serve Your In-Group?
I think a good indicator of when you're caught in a bubble is when your favorite theories all end up favoring your own in-group in particular.

That's why finance experts tend to favor monetary expansion, interest rate reduction, and "limited" government bank bail-outs (my bank, not the other guy's bank, because my bank just got caught up in the problem, whereas the other guys' banks were reckless!) and believe that the world would grind to a stand-still in the absence of investment mega-banks.

That's also why government pundits, lobbyists, and policy wonks all want to save federal government agencies and believe that the world could not possibly function without them.

That's also why teachers' unions all believe that if we don't keep pouring money into education, our children will become drooling morons. That's also why farmers believe that if you don't subsidize agriculture, the world will be unable to eat anything at all.

See, people often come up with theories that explain why their own interests are the most vitally important interests in the world. I think maybe there had been a time in human history during which people understood that these "theories" were really just political arguments. They started making these political arguments to carve out their own slice of government cheese and live their self-absorbed lives.

But somewhere along the lines, people within these in-groups started genuinely believing their own hype. It wasn't a political argument for more government cheese anymore, it was an "evidence-based" conclusion from incontrovertible evidence.

The problem is that it's not incontrovertible. These theories are still just political arguments being made for political gain. People get caught up in their own little individual universes and forget that the financial sector is not the most important thing in the world to a non-banker, that the bureaucracy is not the most important thing to those of us who live outside of the bureaucracy.

In the end, people would be really, unpleasantly surprised if they understood how little impact they are making on the world. The truth of the matter is that unless you are a Bill Gates or a Norman Borlaug, your life does not have much impact on the world.

We are mortal. Our time here is fleeting. We should resist the urge to explain the whole universe in self-serving terms, even when that explanation seems to involve data.