How to pay for infrastructure

How to pay for infrastructure

By Matthew Yglesias

SlowBoring.com

March 14, 2021. 

Construction along a highway

(Getty Images)

Back in January, when the Biden economic team originally outlined its American Rescue Plan, they also said that they anticipated rolling out a second plan — which at the time, they dubbed “recovery” — which would be more focused on infrastructure.

What they said was that they expected that second plan to be ready in time for Biden to address a joint session of Congress, which new presidents normally do in late February, but they did not commit to a specific date. Obviously it is now early March, Biden has not addressed a joint session of Congress, and there is no plan available.

But members of Congress are starting to talk about it, with moderates expressing a few reservations.

I myself have reservations about the idea of a multi-trillion infrastructure bill, but the specific reservation I am hearing — which is focused on the budget deficit — strikes me as precisely the wrong thing to worry about. If you have a good infrastructure plan (and I really hasten to emphasize that this is a big “if,” not a trivial thing) then it is incredibly reasonable to pay for it by borrowing money. The part of the Democratic agenda that needs paying for is the stuff in the American Rescue Plan that’s scheduled to expire next year, but which most Democrats want to make permanent.

Democrats want to “pay for” things

Axios did a piece on Joe Manchin’s thinking: “Manchin said the infrastructure bill can be big — as much as $4 trillion — as long as it's paid for with tax increases. He said he'll start his bargaining by requiring the package be 100% paid for.”

And reporting from Sarah Ferris and Burgess Everett in Politico makes it clear that it’s not just Manchin; lots of Democrats are excited about spending but leery of even more debt.

“At some point we’ve got to start paying for things,” said Sen. Angus King (I-Maine), who caucuses with the Democrats and is worried higher interest rates could become an albatross on the economy. “It’s got to be paid for. It’s just a question of who pays. Are we going to pay or our kids going to pay?” [….]

“Some of it needs to be paid for,” said Sen. Jon Tester (D-Mont.), who suggested an “all of the above” strategy for paying for an infrastructure package that includes spending cuts and raising new revenues. […]

“I think it's time for everybody in this body, in the country, to start injecting that term back into the vernacular — ‘paid for,’” said Rep. Dean Phillips (D-Minn.), who added that he and other centrists are already in talks with top Democrats about the importance of paying for the next package.

Since this exact phrase keeps recurring, I want to make a fussy point about it, namely that every debt-financed initiative from the Iraq War to the Trump tax cuts to the COVID-19 relief bill was, in fact, paid for.

During the Revolutionary War, the Continental Army would sometimes make ad hoc requisitions of supplies and then leave those whose goods had been confiscated with IOU notes of questionable real value. And while that was an unusual episode in American history, governments attempting to mobilize real resources through coercion rather than financial payment is an actual thing that happens. We have a volunteer army today, but conscription is a way to get a lot of soldiers without fully paying for it.

I acknowledge that this is a somewhat pedantic point, but I do want to insist on it. If you want to renovate your house but you don’t have the cash, there’s a difference between getting a renovation loan and hiring contractors who you then stiff. Paying for things with borrowed money is a way of paying for things, not an alternative to paying for them. King put it more squarely, which is that the difference between debt financing and tax financing is whether you try to constrain current consumption to pay for investment or future consumption. And I think the right answer is future consumption.

Here’s a household analogy

It’s supposed to be bad to analogize public finance to household finance. But it also seems inevitable. So why not go for it.

Let’s say you own a big old house that’s pretty nice, the mortgage is paid off, and you’re going to leave it to your kids when you die.

Then one day, your gas furnace breaks and the house is very cold, so you call someone to look at it. He quotes you a price for a replacement, but you’re environmentally-minded, so you also ask about the option of an electric heat pump. The heat pump is more expensive, but also more ecological, and it saves you money in the long run. But speaking of which, if you’re interested in the environment he can also recommend a bunch of upgrades to your insulation, and that you replace your windows with modern, energy-efficient ones. Plus, your roof is well-situated for rooftop solar. All this stuff costs money, but it also has long-term benefits. And since your house is all paid off, you have plenty of home equity that you could tap for a loan.

Now, I’m not saying that you definitely should mortgage your house to do a bunch of energy efficiency upgrades.

But what I am saying is that “think of the children!” should not push you in one direction or another. What’s needed here is for you to run the numbers on the benefits of the efficiency retrofit versus the interest rate on a loan. If the retrofit is a good deal, then you are hurting your children by not taking the loan! If it’s a bad deal, then you are hurting your children by accepting it.

Today, even after the COVID-19 relief bill’s passage, interest rates remain extremely low so it is very easy for projects with meaningful public benefits to clear the test of being worth financing with debt.

Chart showing daily treasury real yield curve rates by days in March

If you insist on offsetting the cost of useful projects with tax increases, you naturally generate political opposition, because people like to have money. It’s true that taxing the rich is popular, but you’re still bound to get more pushback than if you weren’t doing it.

King should take his “think of the children” mantra seriously. If we can leave our children with abundant clean energy, they will be extremely grateful. I will be the first to admit that I am not super-knowledgeable about the policy specifics in the energy world. But if there are promising solar projects, promising wind projects, promising nuclear projects, promising geothermal projects, promising battery projects, or anything else that has high upfront costs but generates an ongoing flow of cheap clean energy, then economically speaking, we should do them and not let financing issues stand in the way.

Do we actually have lots of good projects?

To me, though, the really big question in this space is “is it actually true that Congress could spend trillions of dollars on useful infrastructure projects?”

I am more familiar with the transportation space than the energy space, and I have some serious doubts there. The great thing about old-fashioned financing of roads through the gas tax is that it functioned as a form of user fee. You could do that in a more modern way with a vehicle-miles-traveled tax. But neither of those seems to be under consideration today. If you pump money into the existing highway formula, then whether that money comes from cheap debt or from taxing the rich, you are just doing more to subsidize sprawl and driving.

In theory, if you put it into transit instead, that could be great. But in practice, TIGER and New Starts have funded a lot of projects that are, at best, bloated and overpriced.

I’ve written before about station bloat in things like the Second Avenue Subway in New York and Boston’s Green Line extension. But both of those are at least legitimately worth building — just better to do more cheaply. 

Something like the Silver Line in D.C. has been much more of a mixed bag. Not only did it cost a lot of money, but it’s just a poorly conceived project that’s actually reduced capacity on the Blue Line and therefore made transit service in the region worse in some ways. The Obama era brought us a lot of mixed-traffic streetcars that do absolutely nothing to improve transportation, and I suspect in many cases diverted local funds from what could have been bus lines.

There are various useful things one could do in intercity passenger rail. But is Amtrak actually prepared to do any of them? 

My broad view is that on the transportation side, despite endless rhetoric about “crumbling bridges,” we mostly have structures in place where either significant expansion is undesirable (highways) or the existing institutions are mismanaged or incompetent. 

Of course, I don’t want to prejudge legislation that doesn’t even exist. But my initial instinct is to be skeptical. More to the point — there’s no good way to pay for a bad project. A good project is worth paying for with debt. But a bad project just shouldn’t be done.  

Making the rescue permanent

If I were to go through the trouble of scrounging up a congressional majority to raise taxes, I wouldn’t use the money on infrastructure at all.

What I’d do instead is pay attention to the fact that the just-passed American Rescue Plan has a bunch of temporary elements that are very good ideas:

It makes the Child Tax Credit significantly more generous and transforms it into a monthly allowance.

It increases Affordable Care Act subsidies so people can get decent plans.

It makes the Earned Income Tax Credit more generous.

It pumps a bunch of money into federal Title I aid to K-12 schools.

It provides a bunch of rental assistance money.

This is mostly stuff that Biden campaigned on, and it’s stuff that Republicans have been complaining isn’t “really” about addressing the pandemic emergency.

And there is some basis for this complaint. The money that Biden characterizes as being “for school reopening” is only loosely tied to that goal. But it’s fairly similar to an education policy goal he outlined way back in May of 2019. Either way, though, it’s a good idea. A fresh meta-analysis by C. Kirabo Jackson and Clare Mackevicius confirms that pumping money into schools, especially those serving low-income kids, has significant benefits. The CTC provisions, as has been widely observed during this debate, could cut child poverty by as much as 50%.

All this stuff should be made permanent. And to make it permanent under reconciliation rules, you need to offset it with tax increases. What’s more, this kind of redistribution is exactly the kind of spending that should be offset in the long run — the goal is to transfer resources from the wealthy to the poor, to parents, and to students. And unlike with infrastructure, there’s no big question of program quality here. We know the government can cut checks. That’s where the tax focus belongs.