Bashing on Brooks

I don't understand Ryan Avent's venom towards David Brooks at all.  Brooks wrote a thoroughly reasonable column about rationalizing our infrastructure investments:

Major highway projects take about 13 years from initiation to completion -- too long to counteract any recession. But at least they create a legacy that can improve the economic environment for decades to come...

Moreover, an infrastructure resurgence is desperately needed. Americans now spend 3.5 billion hours a year stuck in traffic, a figure expected to double by 2020. The U.S. population is projected to increase by 50 percent over the next 42 years. American residential patterns have radically changed. Workplaces have decentralized. Commuting patterns are no longer radial, from suburban residences to central cities. Now they are complex weaves across broad megaregions. Yet the infrastructure system hasn't adapted.

The smart thing to do is announce a short-term infrastructure initiative to accelerate all those repair projects that can be done within a few years. Then, begin a long-term National Mobility Project.

Create a base-closings-like commission to organize federal priorities (Congress has forfeited its right to micromanage). Streamline the regulations that can now delay project approval by five years. Explore all the new ideas that are burgeoning in the transportation world -- congestion pricing, smart highways, rescue plans for shrinking Midwestern cities, new rail and airplane technologies. When you look into this sector, you see we are on the cusp of another transportation revolution.

A mobility project would dovetail with the energy initiatives both presidential candidates have offered. It would benefit from broad political support from liberals and business groups alike. It would rebalance this economy, so there is more productive weight to go along with Wall Street wizardry.

For some reason, this makes Ryan, whose blogging is usually very well measured (as well as quite brilliant), go postal:

So, please, let's not pretend that a 14% increase in government spending has been all about stimulus packages, okay? And really, why in god's name does Brooks think he knows the first thing about whether stimulus does or does not work? All stimulus plans work, to some extent-$100 billion in spending or tax cuts doesn't just vanish.

The question is how well such plans work (and there's data available!), and how efficiently they can be crafted. These subtleties are seemingly lost on Op-Ed Doofus. But having accepted that the world is going to pass another round of stimulus, despite the rhetorical power of his inane, uninformed babbling, he suggests that infrastructure is the way to go. Which we can all get behind. Unfortunately, he immediately starts writing about more stuff he doesn't understand, namely, transportation economics . . . Now, perhaps I ought to read this charitably, as the harmless urgings of a writer who really wants us to build a better America. I'll leave it to you all to do that on your own time. To me, it sounds like Brooks really wants to make it easier to build a bunch of highways. Commuting patterns have changed, you see (in response to what, I wonder?), and transportation hasn't kept up. Plus we have to reduce congestion (which has grown in response to what, I wonder?), and somehow this would all "dovetail with the energy initiatives both presidential candidates have offered." O RLY?

Look, a broad project to improve the nation's infrastructure is very desirable. We should be spending a half a trillion dollars on this. But if we approach this effort from the position of willful ignorance that has created so many transportation problems, then we're not helping anyone. So let's review a few things, shall we?

Induced demand is no joke. If you build new infrastructure people will come, so it's important to think about where you'd like the people to go when you build new infrastructure. A massive wave of highway construction will push people outward. It will increase commuting times, and in the long-run it will increase congestion. It will increase energy demand, because no matter how you power your car, longer trips mean more energy consumed. In the short-run, it will mean an increase in dependence on fossil fuels. It will massively increase maintenance costs. If you think caring for today's highway network is expensive, then just try doubling it in size. If we try to accommodate a 50% increase in the country's population with entirely auto-dependent, horizontal growth, then we're all going to end up miserable and poor.

This is, quite frankly, the stupid way to think about transportation. Brooks did mention congestion tolling. The use of congestion pricing would immediately solve the congestion problem while raising billions of dollars that could be used for reinvestment in the nation's transportation systems. Instead of building more roads, we could focus on maintaining the ones we've got, and on upgrading the busiest corridors within and between cities to faster, greener, higher capacity transit and rail systems. That's how you tackle multiple problems at once. Make that shift, and you reduce maintenance costs, reduce congestion, reduce energy use, and increase economic productivity.

The downside is that there's a higher up-front capital cost. But that's a good thing, given that we're also talking about stimulus, correct? Right now, when we need to spend the money, and when government borrowing is fairly cheap, we should be investing a great deal in these projects. Then later, when the economy is rolling along and we want to reduce spending, we'll have an infrastructure that will allow us to do it.

For starters, Brooks's statement about stimulus doesn't strike me as particularly controversial.  The point of stimulus is to use the government to artificially expand aggregate demand by borrowing at fairly low interest rates and spending the money.  The problem is, if you send out checks, most of the checks seem to get saved, which creates a closed, useless cycle.  And spending, other than through channels like unemployment and welfare benefits which already have well defined and rapid cash-transfer mechanisms, take so long to get off the ground that they rarely come until it is too late.  Monetary stimulus is faster and usually more effective, unless you are, as we are now, in some kind of liquidity crisis where banks won't lend.  It is dispreferred mainly because it too needs to be applied early, and politicians have little control over it. This being about what Brooks says, in somewhat less technical language.

The rest of the column does not actually seem to be at odds with Avent's overheated rhetoric.  Brooks did not, to be sure, stand on his chair screaming "Spend it all on trains, you fools!  You'll kill us all!"  But spending it all on trains is first of all, not politically very likely, second of all, inappropriate to the many sparsely populated areas of the United States, and third, not desireable according to most of the experts I've talked to, not least because things have to get from the trains to other places.  If we want to, for example, expand our freight rail, this may imply more highways to rail hubs. 

Perhaps I'm just prejudiced because my father, a rock-ribbed Democrat and HUGE rail buff who sat on the Senate's recent commission on surface transporation finance, emailed me a loving note about that column this morning, saying that it read as if he'd been eavesdropping on the commission.  But even beyond that, the very worst you can say about Brooks' column is that it's not as gung ho about stimulus and rail as Ryan is.  I respect both of Ryan's positions, while disagreeing with them, but Brooks' positions are well within even the center-left mainstream.  For the life of me, I cannot imagine what there is in there to get angry about--unless you're a hard libertarian who hates any and all government spending.