« John Taylor on the Fed's "industrial policy" | Main | Why being President-elect is a difficult job »
Ed Glaeser's words of wisdom
While the mechanics of a payroll tax cut are simple, spending hundreds of billions wisely on infrastructure is hard. Currently, the federal government spends about $40 billion a year in transportation, and another $20 billion on other forms of infrastructure. There is a case for significantly increasing this amount. Our roads do need repairing, and it makes sense to invest more in a downturn when unemployment is high. But even doubling the current federal infrastructure expenditure, a vast increase, would represent only 8 percent of a $750 billion package.
Here is much more, on the mark throughout. Here is related material by Mark Thoma and Paul Krugman. Here is a related post by Alex.
Addendum: Obama now is calling for $300 billion in tax cuts as part of the stimulus plan, roughly forty percent of the total.
Posted by Tyler Cowen on January 5, 2009 at 07:35 AM in Economics | Permalink
Comments
As usual, Glaeser is right on the mark.
His point about giving money to local governments makes sense. It remarkably counterproductive for so many states to halt projects that are already underway, some on the verge of completion. That's very inefficient.
The first place federal money should go is to all the states in sufficient quantity that they can a) complete all projects already approved and b) start a few new projects on the margin, presumably those that would have been funded next had the money been available. States that don't have many projects could use their pro-rata share for tax rebates to their citizens.
This won't add up to all the needed stimulus, but at least it would eliminate the current foolishness we see where most states are halting previously-approved projects.
Posted by: a student of economics at Jan 5, 2009 8:48:47 AM
I'm glad to see Obama has completely decoupled spending and taxing. Now he can reduce taxes and increase spending in any amount he wishes. Soon all ponies will be free! Won't that be grand?
Posted by: K T Cat at Jan 5, 2009 9:02:43 AM
This is good news, isn't it?
But even doubling the current federal infrastructure expenditure, a vast increase, would represent only 8 percent of a $750 billion package.
If we can double infrastructure spending at low cost, why not? I hope no one is letting the cart lead the horse, and thinking that we've "got" to spend $750B, rather than ... just looking for good and appropriate stimulus.
(I think stimulus should "time shift" needed expenditures forward, but not "make work" with wild abandon.)
Posted by: odograph at Jan 5, 2009 9:14:49 AM
IMO this is an opportune time to do a tax cut that could save money in the long run. Eliminate the SS and Medicare taxes now! This would end the charade that SS is a retirement program and allow the benefits to the rich and middleclass to be cut in future. It would also undermine the arguments against raising the retirement age, which argue the fact that the poor pay into SS and then die younger that the retirement age.
Posted by: floccina at Jan 5, 2009 9:27:05 AM
I think it is interesting to note Keynes's view of start-stop public works-infrastructure spending. This is from a 1942 BBC radio talk:
"The difficulty of predicting accurately the appropriate pace of the execution of the building programme is extremely tiresome to those concerned. You cannot improvise a building industry suddenly or put part of it in cold storage when it is excessive. Tell those concerned that we shall need a building industry of a million operatives directly employed – well and good, it can be arranged. Tell them that we shall need a million-and-a-half or two million – again well and good. But we must let them have in good time some reasonably accurate idea of the target. For it the building industry is to expand in an orderly fashion, it must have some assurance of continuing employment for the larger labour force. (Keynes, Collected Works, vol. XXVII, p.268)."
Posted by: Mario Rizzo at Jan 5, 2009 9:31:53 AM
Yes, let's end Social Security and instead force private citizens to send their retirement funds to trustworthy fund managers (who, according to "conservatives," ought not be regulated.)
Posted by: meter at Jan 5, 2009 9:44:09 AM
Our roads do need repairing, and it makes sense to invest more in a downturn when unemployment is high.
I realize this sounds eminently reasonable, but I think it is wrong. During a recession, everyone needs to cut back on discretionary spending. Americans overconsumed during the housing bubble, so the way to fix that is to underconsume now. By throwing hundreds of billions in borrowed money at anything that moves, the federal government simply makes it that much harder for Americans to replenish their net wealth. (I am assuming my household balance sheet goes up more when I am allowed to invest $1,000 than when the government borrows that much on my behalf and spends it on "infrastructure.")
Posted by: Bob Murphy at Jan 5, 2009 10:29:18 AM
Giving additional money to big spending states, without any efforts to get them to review spending programs, reward states that are big spenders while punishing states with prudent management. Plus how do you reconcile the need to make users pay for benefits and giving checks to state governments
Posted by: DanC at Jan 5, 2009 10:37:09 AM
Bob makes an interesting argument. The problem is that political incentives make "do nothing" an impossible answer, especially since there is widespread belief (it certainly may be wrong as I'm not a scholar of the period) that Hoover did nothing and that prolonged the Depression.
Posted by: mk at Jan 5, 2009 10:53:52 AM
Over the next few decades of the 21st century, there will be a shift in emphasis from physical infrastructure to online infrastructure. Most of the economy will eventually shift from the bricks and mortar to online, paralleling the previous historical shifts from rural to urban and (in the US) from urban to suburban.
Much of the physical infrastructure that we take for granted today will be left to decay, much like the inner city downtowns and the grand old buildings in Detroit. A lot of the urgently needed repairs and maintenance that generate so much discussion nowadays will be put off, more or less forever.
A world with continent-wide 3G or Wifi and where you can carry everything you need for work or entertainment in the form of an iPhone/Kindle/netbook combination will be a world where there will be much less need to travel anywhere quickly, because you will be able to do whatever you need to do wherever you are. Perhaps a world of slow-motion nomads, stopping to sleep overnight in Japanese-style "coffin motels", which will be as ubiquitous as Starbucks is today.
Better to concentrate on the truly crucial infrastructure for the future: energy, sewage, water supply, very-high-speed wireless and wired networks, ubiquitous sensors and M2M communication, augmented reality, etc. And put much less emphasis on repairing old bridges that might see an order of magnitude less commuter traffic thirty years from now, or repairing old grade school buildings that presume students will always need to be physically present in some classroom.
Posted by: at Jan 5, 2009 11:22:29 AM
DanC is concerned that sending money to the states may "reward states that are big spenders while punishing states with prudent management."
The solution is to send back to states an amount proportional to what their citizens currently send to the national gov't. If the state wants to use the funding to maintain or expand infrastructure investment, great. If they don't want or need such investment, they can simply rebate the money to their citizens directly.
Bob M. assumes that "my household balance sheet goes up more when I am allowed to invest $1,000 than when the government borrows that much on my behalf and spends it on "infrastructure.""
That assumption is not grounded in standard economics. Because of the existence of (partially) public goods like bridges, roads, public health, basic research, etc, it is entirely possible that the marginal dollar will create more wealth via investment in public goods, which is systematically under-provisioned by private spending. The oft-repeated claim that government spending is inherently less efficient than private spending reflects an ideological view, not basic economics.
Posted by: a student of economics at Jan 5, 2009 11:23:41 AM
@meter incase your post was targeted at me I did not say that we should end Social Security, I would just like to end the SS and Medicare taxes enabling needed reform latter on. If your post was not targeted at I apologize.
Posted by: floccina at Jan 5, 2009 11:37:30 AM
Tax cuts are a great idea. He's smarter than I thought.
Yahoo headline says 300 MILLION, which is a dollar per person!
Posted by: Al Brown at Jan 5, 2009 11:58:01 AM
Ed Glaeser, from the link provided: "fairness and economic efficiency dictate that infrastructure should generally be paid for by users, not general tax revenue. It is appropriate that gas taxes pay for federal highway aid. Using general revenues to build highways means more subsidies for carbon-emitting cars."
That's funny. Ed didn't mention that 15% of gasoline taxes paid by vehicle drivers actually subsidize the tiny portion of commuters nationwide who use mass transit. I can only hope that he will use his status as an eminent urban economist to promote such fairness in funding of mass transit. It is only fair that train riders pay for their trains.
Posted by: John Dewey at Jan 5, 2009 12:20:01 PM
It seems like they shouldn't spend money on construction, but destruction. Tare down the unwanted homes, developments, and infrastructure and convert them to farms or wilderness. It would help with the housing glut and decrease future maintenance expenditures. There are places all over the country that could use this, from the rust-belt to the sun-belt.
Posted by: Jason at Jan 5, 2009 1:08:49 PM
John Dewey is unhappy that a portion of gas taxes support mass transit. But automobile commuters do benefit when other people take mass transit, since it reduces congestion and other negative externalities.
In fact, that's probably why a famous study found that 98% of U.S. Commuters favor Public Transportation for Others.
Posted by: a student of economics at Jan 5, 2009 2:15:18 PM
//I realize this sounds eminently reasonable, but I think it is wrong.//
I disagree, with the first part only: To me it is completely counter-intuitive to suggest that we waste money (for, as we all well know if we are honest with ourselves, government expenditures result in much waste and inefficiency) during a recession. Also, it makes more sense, to me, to invest when we actually have the money to do so. As Mr. Murphy suggests, we have spent too much as it is; my basic common sense gut-feeling is that now we - both gov't and consumers - should save. Investment will pick up again when it should, but now's not the time to push that OR more spending. But that's only the common sense side, which clearly holds little sway in the economic realm.
And to "student of economics": //it is entirely possible that the marginal dollar will create more wealth via investment in public goods, which is systematically under-provisioned by private spending.//
Is it not possible that investment in public goods (roads, bridges, etc.) is (would be-how could we ever know since it's hardly been given a chance in private hands?) entirely OVER-provisioned by gov't, and poorly/inefficiently priced to boot? Witness road congestion and *wait for it* over investment in suburban homes due to easy and convenient access to distant workplaces via public roads. Oh, and I think an Austrian economist like Murphy would know that sometimes his assumptions are not based in standard economic theory. They're, um, based in Austrian economic theory.
Posted by: Tim at Jan 5, 2009 2:47:46 PM
a student of economics: "But automobile commuters do benefit when other people take mass transit, since it reduces congestion and other negative externalities."
The reduction of congestion on highways provided by light rail has been very slight. In cities such as Dallas, the introduction of light rail has not led to increases in mass transit usage. Those who formerly rode on busses now ride on expensive trains. Furthermore, the increase in congestion around light rail parking lots has significantly burdened those drivers who use nearby roads.
If a city wants to reduce congestion, the most cost-effective method by far is to build more highway lanes.
Please tell me, student, what are the negative externalities you believe light rail trains have reduced?
For those who didn't follow the link provided by Student of Economics, he/she is citing a "news" story from The Onion.
Posted by: John Dewey at Jan 5, 2009 2:57:19 PM
student of economics: //since it reduces congestion and other negative externalities//
Let's be clear: An externality, properly defined, is a market failure. Congestion on public roads is decidedly not in this category: The roads are provisioned by gov't for public use, so the failure becomes one not of the market but of the gov't to properly price the usage thereof. Private roads solve the problem easily with tolls, which are efficient, and so there is no "market failure." That's where I disagree with John Dewey: The most cost effective way to reduce congestion is not highway expansion -which will soon enough be clogged up again anyway- but by tolls. Congestion is not an externality unless one abuses the term to encompass anything and everything that imposes some "cost" that is not adequately provided for. If you're going to call congestion on gov't roads an externality, we may as well designate as the same smelly people, ugly people, beautiful people, and harmful gusts of wind.
Posted by: Tim at Jan 5, 2009 3:35:00 PM
Tim: "The most cost effective way to reduce congestion is not highway expansion -which will soon enough be clogged up again anyway- but by tolls."
You are correct in an economic sense, of course. But I do not foresee major cities suddenly switching their free highways to toll roads. The "political will" is just not there. So the choices are usually:
- add mass transit, which is then rarely used at levels that accomplishes anything;
- convert free lanes to HOV lanes, which usually just makes everything worse;
- expand free highways with free lanes;
- expand free highways with combined high occupancy and toll (HOT) lanes;
- add new toll roads.
Few corridors exist for adding new toll roads.
The easiest way to personal relief from congestion is often simply to live close to one's workplace.
By the way, I do not agree that expanded highways will "soon enough be clogged up again". I have observed highway expansion in Texas that reduced congestion for a decade or more.
Posted by: John Dewey at Jan 5, 2009 4:05:07 PM
Sorry, John, I did not realize you were talking about politically realistic options. I agree toll roads aren't coming soon: nobody wants to pay a toll, or vote for someone who makes them do so. And if that's off the table, then you are probably right about reducing congestion with more highway space.
I will say that they've done it in London, and although I've read positive and negative reviews of this policy, it seems to have worked decently as far as reducing congestion. We'll see if the "political will" changes to incorporate this other, more efficient, option.
Posted by: Tim at Jan 5, 2009 4:40:43 PM
What does the currently most referred to economist say? Keynes is dead, but he talks alot.
Concerning the Great Depression (his "slump"), Keynes said investment should go to new projects. Keynes had economic reasons for choosing "new", although I am not sure how "old" "new" could be -- repair roads?
Certainly, one wants to avoid a glut of money into one area, which might have diminishing returns on larger expenditures (or are there economies of scale?).
Certainly new would include projects like windmill factories -- 1 million large windmills would match all current production going into electric lines.
It would include a remodeled city with population on a 10 mile diameter circular train track and 100,000 people at each train stop as in Hong Kong.
It would include funding software programmers to solve public problems; eg, making free quality anti-virus software and stoking the database of viruses needed by that software (like free flew shots).
It would include creating a Battery Labs that hired an international cadre of experts to research better batteries -- comparable to Singapore's concentrated government funding.
How can we extract ourselves when we hardly listen to others; unlike Singapore, the new U.S. administration will not hire a single appointed government official who is a quintessential expert with a foreign nationality.
We would never hire a Keynes as Treasury Secretary.
When the decision about spending $1 trillion gets made, hopefully, a quant will have a perspective of probabilities and gains/losses, rather than a lawyer gassing out his head.
Posted by: Jameson Burt at Jan 5, 2009 4:53:57 PM
JD asks: "Please tell me, student, what are the negative externalities you believe light rail trains have reduced?"
A: Well I didn't limit transit to light rail. Busses, subways, etc. count, too. But since you asked, congestion is reduced (as you note in your own post), though the benefit is not as great as you would like. Also, air pollution, accidents, global climate change, funding of terrorists also are externalities that are reduced.
Posted by: a student of economics at Jan 5, 2009 5:13:34 PM
Tim claims: "Private roads solve the problem easily with tolls, which are efficient, and so there is no "market failure."
In theory, if you set the toll price exactly correctly, it can be efficient. That's why I favor congestion pricing.
However, unregulated private ownership of roads will almost surely set the price too high, creating a monopoly deadweight loss. Consider the losses created by the extortionate tolls every few miles in medieval Germany.
Furthermore, there are substantial complementarities and interdependencies among roads -- consider pricing at or near intersections or simply long journeys requiring multiple roads -- so multiple competing road owners would not take into account these externalities unless they formed some sort of cartel, but then we're back to the monopoly problem.
Perhaps a mutual company which was owned by all the potential users and subject to the discipline of voting would work, albeit imperfectly, as the owner of the roads.
We could call it a "democracy".
Posted by: a student of economics at Jan 5, 2009 5:24:33 PM
@John Dewey and a student of economics
That's funny. Ed didn't mention that 15% of gasoline taxes paid by vehicle drivers actually subsidize the tiny portion of commuters nationwide who use mass transit. I can only hope that he will use his status as an eminent urban economist to promote such fairness in funding of mass transit. It is only fair that train riders pay for their trains.
It is even worse than that...In the USA mass transit uses more energy per passenger mile than cars do!
I like the idea of safer motorcycles/scooters like the BMW c1 200. Also safer cars through electronics (sensors etc) rather than through heavier vehicles. These two things might decrease congestion, save time money, fuel and lives.
Posted by: floccina at Jan 5, 2009 5:34:36 PM
