The Myth of Infrastructure Spending
Trump's $1-trillion infrastructure spending plan continues to be one of his less controversial proposed policies. In Washington, and even among many in the general public, there is a consensus that government spending on more roads and bridges is always necessarily a slam dunk.
This is stated matter-of-factly in a recent CNN article about the president's plan:
Infrastructure spending is a proven winner for the economy: It creates jobs, fuels growth and allows Americans to get from point A to B faster, making them more productive.
This statement is not a quotation from any supporter of the the plan. The author of the article is simply stating what he believes to be an indisputable fact.
The fact that statements like this are made in such a blasé way should not surprise us.
This attitude goes back at least to the early nineteenth century when it was assumed that government infrastructure programs were all necessarily "proven winners" for the metaphorical construct known as "the economy."
200 Years of Boondoggles
Perhaps the most famous example of this is the transcontinental railroad which is still held up as the great "proof" that massive infrastructure projects are indisputably net gains for the nation.
In reality, the transcontinentals were never "proven winners" and were actually huge engines of debt and corruption in nineteenth century America. They siphoned off vast amounts of wealth from other uses where money would have been better spent.
Even worse, the railroads — like all infrastructure — required constant upkeep. Thus, the creation of new infrastructure was really a creation of future government commitments for more government spending.
In his huge study of the intercontinental railroads, historian Richard White noted:
[The transcontinental railroads] were less an asset that one generation passed on to the next than a debt that the past imposed on present and future. I mean this quite literally. The seeming durability of the iron, steel, and stone that made up a railroad was, if not an illusion, then largely relative. The wind and rain, the ice and the snow, the flood and the landslides, the rust, the rot, and the fire that ate away at railroads meant that they had to be constantly repaired and rebuilt, or they ceased to function. When nineteenth-century editorialists proclaimed that the fraud, deception, and loss of capital that the railroads entailed ultimately did not matter, because Americans now had the railroad, they did not know what they were talking about. Without constant new investment and labor, the railroads would have been useless...
All of the watered stock, the money siphoned off into private pockets, waste, and fraud that characterized the building of the railroads created a corporate debt that had to be paid through higher rates and scrimping on service. A shipper in 1885 was still paying for the frauds of the 1860s.
Thus, infrastructure spending is not some one-time thing that blesses the nation with new wonderful things that everyone can use indefinitely. It is often a commitment to simply spend much more money in the future. So, you'd better be sure you're buying exactly what you need.
White also observed that shipping through the railroads did not become competitive with sea-based shipping for decades. Western states did not become wealthier in the wake of railroad construction across the frontier. In fact, incomes in Western states went down during the same period. So much for the economic slam dunk.
In spite of it all, we continue to ignore the true costs of this sort of spending. Charles Hugh Smith observes:
Building bridges to nowhere isn't just a waste of money in the present; it saddles the economy with productivity-draining costs for decades to come.
If there is anything the political left, right and center can agree upon, it's the lasting benefits of spending more (borrowed) money on infrastructure: roadways, rail lines, airports, seaports, pipelines, dams, electrical lines and so on: the physical networks of advanced civilization.
That Roman roadways constructed 2,000 years ago are still visible illustrates the longstanding value of reliable infrastructure: Roman political control and trade depended on roadways and sea transport to tie the sprawling empire together.
This is the basic assumption behind the notion that virtually any and all infrastructure spending will create value far into the future.
But is this really true? Does rebuilding and/or adding infrastructure create economic value?
To answer, we need to look at two issues: productivity and cost-benefit.
Infrastructure creates new value when it boosts productivity, generally by lowering costs of moving goods, energy, etc.
The value created by increased productivity must far outweigh the cost.
Consider the classic "bridge to nowhere" infrastructure project: a bridge is constructed between a sparsely populated island and the mainland. The payoff is a handful of residents are spared the time and inconvenience required to ship their vehicles between the island and mainland on a ferry.
Does this time savings translate into increased productivity, or merely extra leisure? And what was the cost to gain this very modest increase in leisure/productivity? Spending tens of millions of dollars on the bridge actually reduces the productivity of the entire economy due to the opportunity cost: the millions of dollars could have been more productively invested elsewhere, and spending the money on a low-value-creating bridge deprived the economy of the capital, labor etc. that could have been better invested in productivity-generating projects.
This is especially true of new projects, of course. Building a new bridge or a new road brings totally unproven or hypothetical benefits, especially — as the railroads proved — in less industrially dense areas. What's worse, the promoters of such spending do not take into account the opportunity cost of the spending.
After all, this sort of spending requires that the money must first be removed from the pockets of taxpayers. It is assumed that this is justified because it creates "stimulus." It creates new jobs in building highways, and makes roads everyone can use. But what would the taxpayer and the worker have done in the absence of that spending?
Patrick Trombly wonders:
The government, via taxation, produces something that the consumers have not already chosen to buy. If they were already producing and buying such things, no government intervention would be “necessary.” Thus, whenever the government spending program — on infrastructure or anything else — ends, we end up with workers who took the “stimulus” jobs instead of the jobs that would have been created by the private economy’s spending or investing the same money. Those workers will have invested their time and energy in the development of skills not actually in demand by the consumers. This is a form of malinvestment, and it impacts employees of these firms in a manner similar to the workers who were misled by Fed-created malinvestment booms into the home construction fields in the 2000s or the oil drilling fields in the 2010s. Of course, workers need not worry about other employment if interest groups can convince politicians to keep pouring billions into these industries indefinitely, even though the taxpayers couldn’t be bothered with voluntarily investing in those industries to the same degree.
"But we're only repairing the infrastructure we already have!" will be the refrain of some supporters of the new spending.
First of all, given the experience of the past, we can say that this claim is almost certainly untrue. Federal infrastructure plans designed for "repairs" have always tended to also result in bridges to nowhere, and shiny new expensive projects where far less extravagant spending would have sufficed quite well but for the easy access to federal largesse.
But even if spending were made merely on repairing existing facilities, the debt and cost that appears on the back end is usually ignored.
Replacing existing infrastructure is also problematic. It may well be necessary, but since it won't boost regional productivity (since it's merely replacing existing structures), it acts as a tax on the regional economy: if the replacement costs $1 billion and generates no real gains in productivity, it is in essence a tax that bleeds capital from the economy that could have been productively invested elsewhere.
Rebuilding a bridge generates higher spending on materials and wages, but if it doesn't generate additional productive capacity equal to its cost, this additional spending (in our world, always paid for with borrowed money that accrues interest for decades to come) runs out once the project is complete, but the costs of paying for the replacement continue on for decades.
As a rule of thumb, if a replacement bridge costs $1 billion, it will cost users and taxpayers $3 billion over the life of the loan/bond that funded the project.
Borrowing immense sums to spend on infrastructure that doesn't boost productivity actually cripples an economy by channeling scarce capital and tax revenues into projects that only boost spending for a few years at best, while the costs of borrowing the money pile up for decades to come.
In other words, building bridges to nowhere isn't just a waste of money in the present; it saddles the economy with productivity-draining costs for decades to come.
This high future cost for no-productivity gain infrastructure effectively bleeds the economy of income and capital for decades, for the temporary sugar-high of infrastructure spending today.
A rigorous cost-benefit analysis might conclude that some aging, marginal infrastructure should be torn down rather than replaced. If self-driving vehicles will reduce vehicles on the roads significantly — and some estimates range as high as an 80% reduction in traffic — perhaps we should wait for this technology to mature before spending trillions of dollars on infrastructure that is about to be under-utilized.
We should instead ask: where are the big gains in productivity going to come from going forward? The answers to that question should guide our public and private investment decisions.
In the meantime, we should question whether proposed infrastructure spending is actually an "investment in our future" or just another bureaucratic boondoggle designed to enrich crony-cartels and justify rising bureaucratic budgets.
Indeed, the tactics of the infrastructure-spending enthusiasts are very similar to that of the global-warming zealots. They put forward massive government spending plans, while ignoring the opportunity cost of the spending or the economic realities involved. But who needs to consider costs when the benefits of the spending — in the minds of supporters — are 100% completely and obviously good?
Anyone who raises questions of cost is simply dismissed as a hopelessly small-time thinker, or a "denier" of reality.
As Always: Decentralize
If a new highway is needed in Peoria, let the Peorians pay for it, either with a toll road or local taxes. If the benefits of infrastructure spending is so abundantly obvious, then it will pay off for every metro area that builds a new road. The great benefit of federal spending on infrastructure, however, is that the new spending can be forced on the taxpayers with much less political effort than would be required to obtain a new local tax. Members of Congress — who are far more insulated from voters than state and local policymakers — need not sorry about how it will be paid for. They can simply rely on the central bank to help the Feds go more deeply into debt to spend another few hundred billion dollars.