These numbers are ridiculous on their face, but dig deeper and the ridiculousness shifts dangerously to outright offensiveness. That’s because, while infrastructure spending is done with real taxpayer dollars, the costs Americans would supposedly be saving themselves from are anything but hard currency.
Real Investment, Paper Returns
When it is reported that an infrastructure investment will save $1 trillion over the next decade, most people outside city-building industries, and a scary number of them within, assume that this means 1 trillion dollars. As in, this infrastructure investment will result in 1 trillion dollars that can be spent on something else: education, health care, perhaps defense.
I suspect this might have been true at some point in the past, before infrastructure spending became an embedded cultural belief. It is not close to being true today. Take the example of $430 billion over a decade that the ASCE suggests will be lost by businesses due to inadequate infrastructure spending. This is not dollars lost; it is a dollar equivalent for their estimate of the value of lost time.
Here is how this works: A project is proposed to add a lane to a congested highway. The highway is carrying 100,000 cars per day. Each trip is projected to be 30 seconds quicker once the new lane is completed. Project advocates would then take those 100,000 trips, multiply that by 30 seconds of saved time per trip, and conclude that the project will save 3 million seconds – roughly 830 hours – of otherwise lost time each and every day.
That’s a lot of time, especially when you realize that the median worker in the market makes $25 per hour in wage and benefits. At that rate, saving 830 hours of time is equivalent to saving $21,000 per day.
And when you factor in that there are 356 days in a year and that the added lane can be expected to last at least 50 years, all of a sudden saving each person a mere 30 seconds of time on their commute results in a total of $380 million saved. Do that calculation over thousands of different projects and it adds up to hundreds of billions of dollars. Now we’re talking real money!
Or are we? When confused for actual money, this saved-time calculation becomes farcical. First and most obvious, we can’t pay for the asphalt, concrete, steel, engineers, and construction workers with 30-second increments of saved time. Bondholders likewise do not transact in saved time. They all require real currency. So, if the project doesn’t result in the kind of growth that generates real dollars, all that saved time isn’t going to matter much.
It is also important to step back and understand what is causing the lost time the project is alleviating. It’s not long-haul truckers traveling from Houston to Kansas City, from Los Angeles to Tulsa. It’s commuters, the kind of people induced by transportation investments to live in the developments that, as shown in Chapter 3, are bankrupting our cities through the illusion of wealth.
And despite what engineers and economists model in their spreadsheets, commuters are humans and thus react to change in complex ways. Quicken their commute by 30 seconds, and they might sleep in half a minute more, move a mile further away from where they work, or decide to drive to the store during rush hour instead of waiting until mid-morning. Either way, it’s not credible to suggest that saved time is going to result in increased worker productivity, let alone create real dollars that can be recaptured to pay for the project.
The ASCE Failure to Act report refers to saved time, but it also refers to reduced wear and tear. This is a method to calculate how much humans will save not having to repair their vehicles as often due to improved road conditions, again, as if humans all respond in the same way. It’s also quite a one-dimensional calculation; there is no consideration given, for example, to the economic benefit from employing more mechanics, auto dealers, and car manufacturers. That’s not because they are any less real but because, as is obvious by now, this math the Infrastructure Cult uses is not trying to suggest anything real. It’s merely assembled as propaganda.
Building a highway? Calculate the time commuters save in transit but ignore the delays they have during construction and maintenance.
Putting in a traffic signal? Calculate the value of potential new business growth but ignore the cost of time delays for people having to sit at red lights.
Putting in a frontage road? Installing a new sewer line? Building a water tower? Calculate the value of the new Walmart, McDonalds, and Jiffy Lube but ignore the negative economic effects from closing the old Walmart, McDonalds, and Jiffy Lube up the road.
A belief in the power of infrastructure investments to generate growth and prosperity is so deeply entrenched that it has impaired rational analysis. Otherwise thoughtful people will not question ridiculous notions, such as whether spending $2.2 trillion over a decade is worth it to avoid $1 trillion in losses during that same period, because it is – in their minds – unquestionable that infrastructure investment is a positive thing.
Psychologists call the human tendency to seek out affirming data, and to ignore conflicting data, confirmation bias, but this goes much deeper. The leadership class in America holds infrastructure investments in such high regard that the overwhelming benefit from new growth is simply assumed. It’s a foundational belief not open to serious examination.
Accounting for Infrastructure
The cult-like belief in the value of infrastructure is evident in the way municipalities track their own wealth. In accounting terminology, the balance sheet is a ledger that lists a city’s assets and liabilities, the wealth it possesses, and the claims against that wealth.
It is relatively easy to understand why a pension promise would be considered a liability. The city agrees to pay a pension benefit in the future. The value of that promise is a dollar amount that can be calculated, based on actuarial tables of life expectancy, historic rates of return for different investment approaches, and other discernable trends. Money is collected from the employee, and some contribution made by the municipality, for the purpose of meeting this future obligation.
The amount of money the city has saved to pay pensions is an asset. The amount the city is obligated to pay out for pensions – calculated in present value – is a liability. The difference between these two is either a surplus or, more likely for pensions, a deficit. It is recorded this way on the municipality’s balance sheet. To overcome a deficit, more money must be set aside and/or a reduction in benefits is necessary. This is straightforward.
It is logical to assume that infrastructure is tracked in a similar way, especially since doing so is seemingly easier than tracking pension liabilities. A new development is built. The cash flow derived from the wealth of the tax base – the taxes from all those new homes and businesses added together – is the community’s asset. The easily estimated future maintenance cost is the liability. Generating a surplus year-to-year across all these developments is how the city stays in business. Again, pretty simple.
Only, that’s not how infrastructure works. The generally accepted accounting practices for municipalities counts infrastructure as an asset, not a liability. There is no accounting of the tax base or the revenue from the community’s wealth; it’s simply ignored. With this approach, the more roads a city has, the more pipes in the ground, the more public buildings and pumps in its inventory, the richer that city is. It’s backward.
Take two cities. The first has a billion dollars of tax base on one block of street. The second has only a million dollars of tax base but has five miles of street. While the first has a plenty of wealth and not much in terms of future maintenance promises to keep, the second city would have the stronger balance sheet. That’s because the second city has five miles of street as an asset instead of only one block. No matter that the billion-dollar city has 1,000 times the tax base, by the method cities use for their accounting, the million-dollar city is wealthier.
Any rationally minded person understands that the street in front of your home is not an asset for the community. It can’t be picked up and sold to the neighboring town. It can’t be pledged as col
lateral against a debt. The street is a liability, plain and simple. In the infinite game of running a city, it represents an eternal commitment to ongoing maintenance and repair.
The community’s asset isn’t the street in front of your home; the asset is your home. Or at least the future tax revenue related to that property. The value of your home represents wealth within the community, wealth that can be tapped to meet the promises the community makes, like maintaining roads and pipes. That cash flow can be – and frequently is – pledged as collateral by the local government when they issue bonds.
It almost feels silly to have to explain something so obvious. Yet, obvious or not, every state and local government in the United States tracks its infrastructure liabilities as if they were assets, while few bother to account for, let alone track, their real wealth. If our municipalities used accurate accounting, most of them would be insolvent.
Assuming Secondary Effects
The few times when I’ve heard members of the Infrastructure Cult pushed to account for infrastructure spending’s return on investment, they always fall back on the assumed secondary benefits. They’ll argue that infrastructure spending makes the economy more efficient, for example, asserting this as de facto truth. The claim isn’t tempered by the fact that nearly every American can point to a local list of wasteful infrastructure projects, investments that cost a lot but didn’t seem to amount to all that much.
Somehow building a new frontage road so the Walmart on the old frontage road can be relocated is assumed to be efficient. The same with building a new highway interchange so a few hundred homes can be built an hour away from the regional employment center. When looked at individually, the public-sector math behind most of these projects makes no sense. Yet, the true believer is comfortable assuming that, if you add up all the negative-returning infrastructure investments, the network effects from them produces a platform of economic efficiency. Not only is there no evidence to support that conclusion; it defies logic.
Another favorite secondary benefit to cite is job creation. The case here is stronger at the federal level and in states and the handful of cities with an income tax, but the financial case is still more wish than rigorous reality. If half a project cost is labor, and a fourth of those wages end up paid in taxes, and 5% of those tax revenues go toward infrastructure, then the tax revenue from temporary construction jobs are covering 0.6% of any given project. I think that’s optimistically high, but even if it wasn’t, there are far more cost-effective ways to create jobs.
It is true that construction workers need services themselves, so putting money in their hands has secondary benefits for the overall economy. Again, we’re not capturing that revenue anywhere measurable, and so correlating spending on projects we desire with positive outcomes we observe – and ignoring any contrary data – is merely a theoretical exercise in bias confirmation. Sure, that construction worker has money in his pocket to buy a coffee and that will put someone to work, but the person or the business paying the taxes on that infrastructure – or the debt on the interest – makes economic choices, too, decisions that are ignored in these one-sided calculations.
Economists like to argue that, in times of economic hardship, if we pay someone to dig a ditch and then pay them again to fill it back in, we’ll stimulate the economy in helpful ways. The case for infrastructure spending is an extension of that logic: If instead of digging and filling a ditch, we spend the money on concrete, steel, and asphalt, we’re actually better off because we’ve built something useful.
What is obvious but not acknowledged in that narrative is that, with the ditch, we ultimately end up back where we started. No long-term liability. In contrast, when we build a road or a bridge or a mile of pipe, we’re left with an eternal maintenance obligation. If the project costs more than the wealth it creates – which most of the projects we are undertaking today do – then we’re just getting poorer, regardless of job creation. We’d be better off digging and filling the ditch. Or just giving people money without laundering it through an infrastructure spending program.
The time savings and wear-and-tear calculations from the ASCE report are also frequently applied to transit projects, particularly large rail projects. Transit projects are generally subjected to more financial scrutiny than highway projects, but that’s still a very low bar. The most commonly cited financial benefit of transit comes from freeing up congested highways, an assumed effect never witnessed anywhere in human history.
Beyond that, transit is often cited as an investment in social equity because it is seen as a benefit for the poor. As with highway investments, there are all kinds of calculations made to show the financial benefits for the disadvantaged of having improved bus or rail service. In other words, the Infrastructure Cult argues that it’s a positive economic benefit to have a development pattern that spreads everything out as far as possible; then they argue that there’s even more economic benefit to provide a marginal transportation option to the growing number of people who can’t afford to live in that very expensive, spread-out development pattern. I find this incoherent.
The latest fad is to tout a project’s carbon-reduction benefits as a contribution to fighting climate change. For example, with all seriousness, project supporters will make a series of intellectual contortions to calculate the amount of carbon saved on a congested freeway, under the assumption that their capacity-building project reduces the amount of stop-and-go driving. They conveniently ignore the more obvious and intuitive fact that making it easier to drive means more people will drive, and more driving means more carbon, not less.
Rather than producing a multitude of unmeasurable benefits, it is far more likely that our passion for infrastructure spending is satiating a more human instinct. Psychologists call this confirmation bias. We include in our analysis the things that make the case for what we want to do, and we ignore the things that would undermine our cause. In other words, we exhibit cult-like behavior.
Calculating secondary effects and comparing them against costs is sometimes a worthy exercise, but only for projects that have a positive, real rate of return. If the project is going to lose money, and the costs continue to compound indefinitely as maintenance cycles repeat over and over, then secondary benefits are meaningless as an accounting exercise.
A Real Return on Investment
What would it mean to break out of the Infrastructure Cult and make capital investments that had a real return on investment? First and foremost, it would require us to spend public money on infrastructure projects that covered their own costs, not only today but indefinitely into the future.
This is nearly impossible with a development pattern that forces us to build in large blocks to a finished state. Such an approach forces us to make predictions about what will happen after an infrastructure investment is made, something humans have a long track record of being poor at doing. We need to know what will be built, what the market will pay for it, that it will be maintained, that the value of the developed properties will increase at rates greater than construction inflation, and so on.
For all practical purposes, this is impossible, which is one of the reasons our ancestors built their cities incrementally. Projections are not necessary and propaganda math is irrelevant when things are built incrementally with ongoing feedback driving adaptation.
If we’re going to break out of the Infrastructure Cult mentality, we would need to design our systems to respond to feedback. There is no clearer feedback on value than someone’s willingness to pay for something, yet our infrastructure funding mechanisms have a large degree of separation from the user’s willingness to pay for what they want.
We all pay our gas tax and expect the roads to be maintained, improved, and expanded when needed. Yet, as the ASCE has suggested – I think correctly – the gas tax comes nowhere near funding transportation to even a minimum tolerable condition. The natural response then is to raise the gas tax, but doing so reduces driving, which reduces reve
nue, which would require an increase in the gas tax. And so on.
And the gas tax tells us nothing about the viability of a specific project. At best, it’s more of a referendum on the public’s perception of the value of the overall transportation system. A more direct and actionable form of feedback would be a toll, yet, in the handful of instances where infrastructure projects have included tolls as their funding mechanism, the data suggests that humans value their time differently than Infrastructure Cult models suggest they should.
The best example is the I-65 bridge in Louisville connecting Kentucky to Indiana. The states spent $1.3 billion on a congestion-reduction project including the construction of the I-65 bridge, which is now tolled up to $4 per crossing. Not only is post-construction traffic not meeting projections, it’s been cut nearly in half from pre-construction levels.11 In addition, drivers willing to avoid the toll are now detouring through a longer, slower route to use a nearby, untolled bridge. Traffic is up 75% on the free-of-charge-but-slower bridge, despite the pre-construction claims that saved time is the same as saved money.12
Beyond tolls and usage charges, we could extend our infrastructure funding approach to include value capture. This is the approach public and private developers for centuries have used to pay for their projects, from Augustus in Rome and Napoleon in Paris to the trading companies given land charters in the New World.
It’s largely how Japan has funded its acclaimed high-speed rail system. Buy the land around the station at pre-development prices. Build the high-speed rail, which makes the land far more valuable. Sell the land at the new price and use the profits from the sale to pay for the rail line. This is great public policy, although the lobbyist for the land speculator is not going to be very happy when their anticipated windfall profits end up going instead to the project’s investor and risk-taker: the taxpayer.
Strong Towns Page 8
