The Economic Implications of More Infrastructure Spending

The idea that something is free, that we can do something with no consequences, is always appealing to the people who want to spend the money or the beneficiaries of it. It seems very unlikely to me, that we’ve discovered a magical way to say, double the size of our economy and that’s simply by spending money.
— Russ Roberts

Russ Roberts is a fellow at Stanford University's Hoover Institution and the host of the podcast, EconTalk. He's also starred in two rap videos about John Maynard Keynes and F.A. Hayek, and is the author of several books, including most recently, How Adam Smith Can Change Your Life: An Unexpected Guide to Human Nature and Happiness. 

In this interview, Chuck Marohn and Russ Roberts discuss the political appeal of infrastructure spending vs. the economics perspective. They also talk about how to ensure a good return on investment and how to focus on smaller-scale projects.

This interview is part of our ongoing conversation on federal infrastructure spending.


+ Podcast Transcript

Chuck Marohn: This campaign season, our leading political candidates have indicated that they want to make some game-changing investments in infrastructure. In a nation that seems deeply divided on so many fundamental issues, the need for large infrastructure investments is seemingly the one place that we have a broad consensus. At 'Strong Towns,' we understand that America's approach to growth and development is bankrupting our cities. This begs the question: if we're committed to spending more money on infrastructure at the federal level, how do we do it in a way that actually makes us better off? To explore that, today we're honored to be joined by a returning guest Russ Roberts. He's a fellow at Stanford University's Hoover Institution and most of you know him as well, as the host of the podcast 'Econ Talk.'

Russ, welcome back to 'Strong Towns.'

Russ Roberts: Great to be with you, Chuck.

Chuck Marohn: The presidential candidates are recommending a large surge in infrastructure spending. I'd like to get your immediate reaction to that idea.

Russ Roberts: I think you want to distinguish between the political appeal of infrastructure spending and the, what I would call the economics perspective. Certainly, the general public views infrastructure as something akin to health apple pie. Who's against it? Infrastructure? That's like ... Being in favor of infrastructure is like saying, "I want to build up my core when I work out." It's like a foundation of the whole economy and our whole country. Of course we need roads, bridges, and tunnels. Of course we need physical health.

Infrastructure, I think, taps in to that primal appeal when politicians talk about it. I think that's the reason that the candidates of both major parties this year, happen to be pretty gung-ho on it. Economists don't tend to be gung-ho on any one thing. We tend to ask the question: how much do we have and what's the value of a little bit more, relative to other things?

There are a lot of things we like. Infrastructure is very important. There are many other things we like. The question is: is the bang worth the buck? That's where the economics's perspective can be very different from the political one. Even within that economics's perspective, there's some differenceS of opinion.

Chuck Marohn: I've read enough Paul Krugman and actually saw an article from Larry Summers a couple weeks ago, talking about the infrastructure spending today as a no-brainer. Economically, this is a absolute no-brainer. I have got questions, myself. Either I'm not grasping something they're saying or they're saying something different than what I'm understanding. Can you give the best spin on why an economist of the Krugman or Larry Summers mindset would say this kind of spending economically is a no-brainer?

Russ Roberts: I think they make a couple arguments and I want to say up front, that I'm skeptical of these arguments. I want to try to give them the best case I think they would make, which is the following: Interest rates are at historically low levels, which means that borrowing money is very cheap. Long term projects, seem on the surface to be very attractive. Of course infrastructure tends to be long-term projects. Bridges, tunnels, roads, high-speed trains, airport construction, et cetera. If we're going to to those kinds of projects, if we need those kind of projects, this would be a good time to do it because the interest rates are low and the cost of borrowing is low. That's that first point they would make.

The second point they would make, is that when unemployment is ... Even though unemployment is relatively low right now, around 5%, employment, the number of people working, is not so healthy. It'd be great to get more people back to work. They would phrase it in the following way, they'd say, "Our potential ... We haven't reached our potential economic growth. We're off the long-term trend. Certainly, there's an important role that infrastructure plays in long-term growth," they would argue and they would also argue that there's a multiplier effect, that the money that would go into the hands of these newer workers would go forward and stimulate other parts of the economy and that would be a good thing.

The third point I think they would make is that we have some serious deficiencies in infrastructure and therefore, this is ... Given the first two arguments. That's why it's a no-brainer. For them it's sort of a .. They see it as a sort-of a free lunch. I say sort-of because it's hard for me to say anything's a free lunch, even when it's on behalf of other people but when they say it's a no-brainer, they're saying, "We need to spend the money. Interest rates are low. We got this extra kick for the stimulus effects of a multiplier, so it's a no-brainer."

Chuck Marohn: I struggle because one of the things that we obsess about here at 'Strong Towns' is the return on these investments. Even in the no-brainer article that Larry Summers put out, the numbers that he used in terms of the actual monetary return and the interest rate that we would pay on this debt, I did the math. It doesn't actually pay back under his rosy scenario for over 120 years. That's a long payback period.

Russ Roberts: That's a long time. That's a long time and of course, there are a lot of things to say on the other side. One of those is the value. It's one thing to say infrastructure's a good thing or we need more infrastructure, but we'd have to specify what particular kinds of investments and specific projects we're talking about because we all know that there are a lot of ways to spend money badly and it doesn't matter how cheap the money is.

There's some skepticism in my mind about whether it has much of a multiplier effect; therefore, I would want to make really sure that the actual projects were good investments. The idea that, 'Oh, it's just all infrastructure," just isn't ... That's not relevant. That's not a sufficient argument. You've got to say which particular kinds of infrastructure are you talking about and make the case for me that they're worth while. That would be the first question and the meta question around that would be, what's the mechanism by which we decide the answer to that question through the political process? For me. The biggest problem I have with federal spending on infrastructure is, the accountability isn't there.

If the federal government allocates some money to general infrastructure again, whatever that means, and it ends up building a road that is relatively untraveled or a bridge that isn't particularly helpful in terms of reducing commute times or a train that doesn't cover it's costs. Even when you take into account say, third party effects, like potential pollution and so-on. Who's going to pay the price politically for that decision?

In the federal government, I think the answer is no one because it's very vague and opaque about ho's fault it is. Who bears the burden? It's spread out across all the tax payers. They don't pay close attention to any one infrastructure project. The advantage of having infrastructure being sourced at the state or the local level is that the people who make the case for it have to raise the money for it and that seems to me to be a really good check on bad investments and bad kinds of infrastructure.

The other point to make I think, is that interest rates are low right now. They're going to be higher later. I hope, because I think they're artificially low right now or at least I worry that they are. Some of these calculations about this being really cheap may not turn out to be true but even-so when it's cheap to borrow it doesn't mean you should borrow. For example, it's cheap to borrow, that doesn't mean I should buy a $5,000,000 house just because, "Oh, well the interest costs will be really low." The interest costs are really not the real cost. They're part of it but the real cost is the principal plus the interest costs, so I can't afford or it's not valuable enough to me to build the $5,000,000 house, which it is not. The fact that the interest rate is low is irrelevant. I've always been puzzled by that argument.

The final point I would make on the stimulus multiplier effect is that, in my mind it's most of the benefits go to the people who pour the concrete and they also are politically well-connected so I understand why they're eager to make the case that infrastructure is really going to be important for everybody else but I know it's important for them. It's not surprising to me that their always advocating for it. I'm not so confident that it's going to draw people into the labor force that the optimists say it will. It will certainly raise the prices of concrete and the wages of workers in say, the construction industry, but whether it will necessarily help the people who are not finding work right now isn't necessarily the case and I'm a skeptic about the general multiplier effects of that.

We have a small historical example recently of Japan, which over the last two decade, or a single decade, has spend trillions of dollars on all kinds of concrete-based infrastructure. They're economy remains stagnant so I'm a skeptic really on all three points.

Chuck Marohn: Does the notion, and I think it comes from the depression, I don't know maybe it comes from somewhere else, that in times where we need to get the economy moving, if we just pay people to dig a ditch and fill it back it, we'll be creating this multiplier effect. It's fascinating to me because when we get to infrastructure, I here that argument made sometimes, that it really doesn't matter if these are good investments or not, as long as we're putting people to work.

I step back and look at cities and I say, "Well, okay. We come in and build this project and then we walk away but now it's not a ditch that's filled in, which is essentially a net-zero, it's actually a long-term liability. Now you've got to go out and fix that bridge, repair that frontage road, repair that pipe. Those are bills that come do. Is that something that economists talk about and fret about and worry about, or is that a lesser concern in today's economy, like a back seat to just growth and job creation?

Russ Roberts: I think you make an excellent point that I think economists would certainly recognize, that sometimes, well I should say almost all investments have future costs associated with them, not just the building of the infrastructure at the time that we're discussing. The point I want to emphasize, again, is that, if you pay people to dig ditches and feel them back in, it's really good for people who make shovels and really good for people who are likely to get the ditch digging and filling in jobs. Ditch digging is an example, I think, that is somewhat deceptive because almost everyone can dig a ditch and fill it back in but most of the infrastructure projects we're talking about have a lot of specialized talent, so it's not always going to be the case that unemployed or underemployed workers are going to be able to fill the jobs that are going to expand because of the increased infrastructure spending.

I really like your point about the extra expense. Let's say I decide to build an extra extension to my house and cover it over with a nice roof and it's very pretty and of course, that extra square footage of my house is going to have extra costs associated with it. If I just merely built it and tore it down and hauled it off, that is a different case and that is the case that the so-called Kanesians often invoke, and my point is, while that will be good for people who build houses and tear them down. Whether it will be good for the economy generally, doesn't seem to me to obviously be the case and it means I don't have money to spend on other things.

I think that you have to add that in also. Those are real resources. They're not free. The idea is that the workers are sitting around looking for something to do but the concrete's not sitting around waiting for something to do and the steel is not sitting around waiting for something to do and a lot of the workers aren't either. They're already employed and they're just going to get paid a little bit more, which is lovely for them but not so good for the economy as a whole if it's not a good investment.

Chuck Marohn: I heard and argument from Richard Duncan, it was last August, Arguing essentially that, the deficit doesn't matter argument. The idea that globalization has pushed down the labor costs, that we have this deflationary cycle, and that we can essentially spend without consequence, at least in the short-term.

There's a Minnesota, mid-western part of me, that just reflectively resists that yet, I these are not stupid people making these suggestions. Am I hearing that wrong or is there actually a theory of Economics that say, "Look, In an economy that is structured the way ours is right now, we can print money and we can spend money and there really is not going to be a consequence, at least not in the short-term that we're going to have to deal with.

Russ Roberts: There are serious Economist's who worries about the debt are at least over-rated. Whether those worries are irrelevant, I think, is a particularly strong example but there are Economists that say that government borrowing money is not a big deal. We owe it to ourselves. Yes, we lend some of it to the Chinese say, or other nations but a lot of it is just our own internal accounting. That does not persuade me. I've never been persuaded by that argument. Whether it's true or not, it does appeal to a certain kind of magical thinking.

The idea that something is free, that we can do something with no consequences, is always appealing to the people who want to spend the money or the beneficiaries of it. It seems very unlikely to me, that we've discovered a magical way to say, double the size of our economy and that's simply by spending money. One way to think about it is to think about the ditch-digging example. Suppose we pay people to dig ditches and we pay them $50,000 a year while they dig the ditches and fill them back in.

Now let's suppose, for the second year, we're disappointed at the impact on the economy as a whole. We've got a lot of people doing this by the way, a few million around the country digging ditches and filling them back in. Instead of paying them $50,000 a year, let's pay them $500,000 a year or $5,000,000 a year and we'll just print the money, or we could borrow it, either one. I don't think either of those is going to lead to prosperity. It's unlikely that that strategy of adding zeros to the end of the paychecks of the people digging the ditches, even though it's going to add to what is called aggregate demand in some Economist's models, it seems unlikely to me that that simple strategy is going to allow us as a nation to have more goods and services at our disposal. We've got no more productive.

The claim would have to be that we would have to have more resources that have been unemployed now employed. I don't think that way of doing it is actually going to have an effective impact so the idea that borrowing money or printing money is a free lunch is deeply appealing but I don't think it's correct. The problem with the costs of those strategies is that they're often pushed off into the future. The United States right now, has no trouble meeting it's debt obligations. We continue to borrow money very easily.

Treasuries, which government bonds, are very attractive in the world market and that encourages to think that we can just keep doing this. We can keep rolling our debt over. It's akin to having a large number of credit card applications I can continually fill out and use my credit cards, an increasing supply of credit cards to cover the previous credit card balances I've accumulated. That can work for a very long time and maybe almost forever for a certain kind of nation. A nation like ours with a very strong economy.

The problem is, it works until it doesn't. When it doesn't, you end up like Greece. Now, we're not Greece. We're nothing like Greece, except for one little thing, which is we spend more than we take in as a nation in the public sector and there is the possibility, we don't know when, we don't know how big it has to be, that we'll eventually reach a point where people say, "You know, I'm not so sure this is a good investment," and we won't be able to roll our debt over, which means we won't be able to finance the new debt without making a lot of heavy sacrifices, which means there would be a temptation to default. The consequences of that are probably not so pleasant. It certainly means we won't be able to borrow again down the road if we need to but that might be the least problematic part of it. I think there is a inevitable tendency to view debt and printing money as "free", when they're probably not and almost certainly not.

Chuck Marohn: When we think about the federal government spending money on infrastructure, today if we look back at the stimulus bill at the beginning of the Obama administration, there was a focus on shovel-ready projects and essentially putting money into the current system. In terms of reform for people who want to change that system, is the counter-veiling theory, the notion that we should starve these systems and they'll change on their own, that seems maybe to me, to be a little shallow, or is there a way to change them by lubricating the system in a sense?

Let's say Hilary Clinton or Donald Trump gets into office. We're going to spend a trillion dollars on infrastructure. The money's allocated at Congress. What would be the least destructive way for that money to find it's way from Washington D.C. to an infrastructure project?

Russ Roberts: When you lay it out like that it's ... even though you said it in a very neutral way. I think as neutral as you could say it. When we start to think about ... There was some phrase there like, "And then Congress will spend the money." What we know about Congress, and this is not a, I don't, a cynical view, I think it's a realistic view, it's that individual members of Congress tend to favor projects that go to their home districts. It's really nothing complicated there and that's going to be the case whether those districts desperately need the projects that are in the discussion, or whether it would be somewhat pleasant, or whether it's not need at all, but there's an important benefactor of that member of Congress who wants to build a particular project right there.

You've really put your finger on the problem. I don't know an easy way to solve that, which is why I'm a skeptic about the value of federal infrastructure spending. I want make it clear, there are projects that span the nation, interstate highway system being the obvious one, space travel being another one, if we thought that was an important part of our future, that aren't really best sourced at the state or local level and are best sourced and funded and designed at the federal level.

Most projects aren't like that. They're local. They have some overlap, maybe across a couple of states, so a piece of highway or a bridge or a tunnel, maybe might be jointly beneficial to a couple of states or to nearby states or residents of nearby states, but in general, the beneficiaries tend to be the people near the projects themselves and as a result, it's not reasonable to think that Congress is going to do that in a way that serves the nation as a whole. They're going to do it in a way that serves their own district.

There's no easy way to improve that that I know of that is constitutional. You could talk about changing the Constitution in how infrastructure spending is allocated. You could imagine different ways you could do that rather than putting it through the sausage grinder known as Congress but under the current system certain kind of reforms are unlikely to lead to radically different results. One example would be a cost benefit analysis. You could say, "Well, for a project to be funded, it has to pass some kind of cost benefit analysis," my guess is that's already there and my guess is that's hard to enforce in a meaningful way. That's my somewhat depressing view.

Chuck Marohn: One of the things that we have put forth at 'Strong Towns' is that we see the highest returning investments being small today. When we look around and compare the Big Bertha Tunnel going on in Seattle and any possible financial return from that, it is dwarfed by the idea that if we just made little low-risk improvements in our neighborhoods, things like planting trees and putting in crosswalks, that those are far higher returning investments but there's no pot of money for that. I'm not asking you to buy into that theory but I am wondering if you have a notion of how we would do smaller projects in a system that is so top-down?

Russ Roberts: Yeah. Of course there's that political challenge that smaller projects aren't as attractive to politicians because they're not as noticeable and therefore they're not as likely to get credit for it. I certainly sympathize with the idea that marginal improvements, that's a word that in Economics means a small step in a particular direction, unfortunately in every day English it means almost insignificant.

When I say a marginal improvement, it certainly denigrates it more than I mean to but I think small projects are the way to think about things and in particular, small projects of different kind that would allow you to get information about the value of a certain kind of infrastructure. Rather than creating a massive light rail system that turns out to be not very well traveled, you might want to try different smaller steps in terms of expanding bus options or different kinds of bike lanes and other things if you want to encourage less car travel.

That example, by the way, shows you another one of the challenges of making marginal steps, it's hard to have a marginal improvement in your light rail system if you don't have one. You're either going to have to build one that goes lots of places or not have one at all, so the tendency is to build one that goes lots of places and then you find out that, "Boy that was really expensive," and not many people ride on it.

I think there is a huge challenge, I'm getting the pronunciation of his last name wrong, but Ben Friberg, has done a lot of research on what he calls mega projects and he finds that almost inevitably, they do not pay for themselves. Now, that's not to say that they're all mistakes or that they are disaster, right?

A project that, let's say, barely breaks even or loses a little money, at least has some benefits and you could argue that those benefits may extend longer into the future than what you're counting and there could be other benefits that you're not measuring, et cetera but he finds often that it's not just that they don't quite pay off, it's that they're awful. I think that's the tragedy of these big specially federal subsidized projects. Remember they're local projects, like the project you were talking about in Seattle, but they're often funded heavily by subsidies form outside the area.

Chuck Marohn: One of the things that I see Economists do a lot is look back at past infrastructure spending and then use that data to project out into the future. As an Engineer, my gut reaction to that is ... When we build the 35W bridge the first time, it creates a lot of growth around it. Now this new connection. Everything changes and people start to invest and build things. When the 35W bridge fell down and we had to replace it, when it went back up it was shiny and new and it looked good but there was no growth that came after that. How do Economists deal with, or not deal with, the fact that past results don't necessarily indicate the future trajectory?

Russ Roberts: In that case, there may not be new investment growing up around it, but to enjoy the fruits of that past investment, you'd want to rebuild the bridge. I think the real challenge there is that when Economists do statistical analysis of this kind, they usually are using some kind of measure that's denominated in dollars so they're going to lump together bridges and roads and tunnels and airports.

All of these things are incredibly important as you say, the first ones, they're very valuable. I'm certainly not suggesting that the government shouldn't do anything in these areas. I am suggesting it's better done at the local and the state level but certainly, it shouldn't be zero. There are many things that the public sector can do much better than the private sector and these are the things we're talking about. The real question is, when Economists go try to measure the impact, do they measure it in a way that makes sense? If you're lumping together all these kinds of projects and trying to figure out the bang for the buck of an extra dollar of so-called infrastructure spending, I think you've aggregated past the point of being sensible.

If you're looking at any particular project though, I think you have to look very carefully at it's setting and what the value is of maintaining it, expanding it. If you build a bridge and then three other bridges get built and so that value of that original bridge is smaller, you might not want to replace it if there are now new ways and new ways that people get around. I think that's the challenge that's hard. You have to go by a case by case basis.

Chuck Marohn: I was in a meeting once and we had this project we were working on and it was a ridiculous project. For me, I looked and thought, "This is just a silly waste." One of the meetings we had, the Engineer for the project was coming in with the economic analysis and I thought, "Well, this is going to kill the project," and when he showed up it had an 8 to 1 benefit to cast ratio. I was stunned. I'm like, "How is this possible?" Because we had done some of the math locally and it didn't make any sense. It was a huge bypass project.

What they had done is they had converted time saved in congestion, wear and tear on you vehicle, an estimate of what they thought safety improvements would be, they had converted all these things into cash for the sake of the economic analysis and then they compared that to a cost, which actually was cash. I know it's an honest way to do things but it seems like a lot of our financial decisions are based on a comparison of esoteric values to actual cash values. Can we run an economy that way or are we misapplying that approach?

Russ Roberts: That's a tough one, right? You have certain benefits often the costs are almost always measurable in terms of dollars and so what you want to do is decide, "Well, are the benefits worth the costs?" And I would suggest that that's not the only way to make a decision but it's important. It's relevant. If you fund lots of projects that have negative net benefits, you're going to be poorer than you otherwise would be overall.

You still might decide it's worth doing because of some specific benefit you think is undervalued but the point you're making is a different one really, which is when you have a cost benefit analysis like that and you have the non-cash components, it's a nice idea to put them into monetary values. The problem is there's always a tendency to over-inflate them if you're the consultant paid to do the estimates for the non-monetary value or even some of the monetary value because there tends to be double counting and sloppiness in those situations.

My favorite example is this, is stadium construction, which you could consider a form of infrastructure for a city. Stadium construction for say, a sport team, take national football NFL team, that's a building that's going to be used eight times a year by the home team and it might be used some other times for concerts and other events but not too often. Most of the studies done by objective Economists find them to be terrible investments for the city and of course, wonderful investment for the owner of the team.

The owner of the team goes to the city and says, "We need a new stadium or I'm leaving," The city says, "Well, we want to be considered a major league city, we have a professional football team, we wouldn't want to lose it, and while it's true that the benefits go overwhelming to the owner and some to the fans, which there are many in the city, the owner holds the city somewhat hostage and threatens to leave otherwise.

Then they hire a consultant, an Economist typically, who estimates the net value of that stadium. They always over estimate it strangely enough. They often double count the benefits of people eating in local restaurants after the game, neglecting the fact that many of those people would be doing something else and still be eating in those restaurants after the game even if their weren't a major league team there or football team or a stadium. They would do other things. Other visitors would come for other reasons.

There is a, what I would call a scientism, meaning a fake, faux science to these kinds of studies. They have lots of decimal points. They have lots of detailed measures down to the last dollar and there's a certain fraudulent aspect to it, a certain sham and fraud to it that's a temptation because a lot of people have a stake in having that stadium being built or whatever that bypass is or whatever the case may be.

Chuck Marohn: I want to ask you one last thing. It's a bigger picture question. When we study economics, we tend to look back at events, the 1870's the 1920's and 30's. Are we living in some very fascinating times to be an Economist and to do a thing like host a podcast where weekly you get to talk about economic issues, it seems like now is a fascinating time to be doing what you're doing.

Russ Roberts: Well, I love my job. I love thinking about these things. The problem I have is that my field has gotten increasingly more mathematical and I don't think that's necessarily ... And also more sophisticated in terms of it's application of statistics and I have not been convinced that that's been a move towards greater understanding or clarity.

I worry that we don't make much progress on some of these problems that are persistent back to the 1870's and the 1920's, questions like, "What should the government's role be I the money supply? What should the federal reserve do?" What do we do when there's a down-turn in the economy?" I see a lot of people with strong ideology arguing back and forth and as ideologues, and that would include me, and not so much as Scientists and yet we have this reputation somewhat as scientific, partly because of the notation we use in our articles and the techniques that we use in our statistics and I think that's a little bit dangerous.

The fun part is that most people ... A lot of people are really interested In how the economy works so they're interested in trying to understand it and that part is I think, what makes it fun to be an Economist today alongside the incredible explosion of technology that allows us to communicate. I like to point out that when I was growing up and becoming an Economist, there were two Economists, or maybe three, who talked to the general public.

One was Milton Friedman. The other was Paul Samuelson. They had alternating weekly columns in 'Newsweek' magazine. You can add John Kenneth Galbraith to that list if you want. He wrote popular books trying to explain economics That was about it. If you wanted to try to teach to the public at large, that's about all that was available. There weren't many slots and because of podcasting, because of blogging, because of the internet, because of technology, the opportunity to share your insights, whether they're true or not, with the world, has grown tremendously and that's really what makes it fun to be alive right now for somebody like me.

Chuck Marohn: Russ Roberts, thank you again for taking the time. It's great to chat with you and I still listen to the 'Econ Talk' every week and I absolutely love it.

Russ Roberts: Thank you Chuck.

Chuck Marohn: Hey, you take care.

Russ Roberts: You too.

Chuck Marohn: Thanks everybody for listening. Keep doing what you can to build strong towns.

(Top image from YouTube)