Just after 6 PM on August 1, 2007, at the peak of rush hour, 111 vehicles were driving across the I-35W bridge over the Mississippi River in downtown Minneapolis when a thin metal plate in the bridge’s central span ripped. The bridge collapsed, plunging vehicles and passengers into the river more than 60 feet below. Thirteen people died, and 145 were injured.

This was not an isolated incident. In May 2013, a bridge on the I-5 north of Seattle collapsed, injuring three people, when a truck carrying an oversize load crashed into it. And in February 2015, a chunk of concrete fell from the bottom of the I-495 overpass in Maryland, crush­ing a car.

These incidents should not have come as a surprise: according to the American Society of Civil Engineers, a quarter of the United States’ bridges are structurally deficient or obsolete. Bridges are carrying more traffic, with heavier vehicles, than they were originally designed to handle, and in 2013, the average bridge was 42 years old.

The problem with the United States’ infrastructure is much broader than fail­ing bridges. The nation’s roads are congested and full of potholes. In 2014, the typical urban commuter spent 42 hours stuck in traffic, up from 20 hours in 1984. Americans consumed over three billion gallons of gas as they sat in grid­lock for almost seven billion hours, at a cost of $160 billion in wasted fuel and time.

In March, meanwhile, the entire subway system of Washington, D.C.—the second-largest in the country—was completely shut down for a day for emer­gency safety inspections. The inves­tigation exposed so many problems that officials have warned that they may have to close large parts of the system for as long as six months for repairs.

The root of the crisis is clear: the United States has underinvested in its infrastructure. The federal gas tax is the main source of federal funding for roads, bridges, and subways. But Washington has not increased that tax, of 18.4 cents per gallon, since 1993; in real terms, its value has thus fallen by over 40 percent. Expert groups such as the American Society of Civil Engineers, business associations such as the U.S. Chamber of Commerce, and unions such as the AFL-CIO have all called for trillions of dollars of new investment. But Wash­ington has failed to act.

The Road Taken, a timely and insightful book by Henry Petroski, a professor of civil engineering and history at Duke University, helps explain why Washington has been unable to solve this problem. In part a history of infrastructure, in part an appeal for greater investment, Petroski’s book offers a rare engineer’s perspective on a debate too often dominated by economists and politicians. Yet engineers can do only so much on their own. A healthy national infrastructure requires not just competent engineers but also a government—and a public—willing to pay for it.


In the years after independence, the Founding Fathers debated whether the Constitution allowed the federal government to play a role in funding what were then referred to as “internal improvements.” In 1791, Treasury Secretary Alexander Hamilton delivered the famous Report on the Subject of Manu­factures to Congress, in which he argued that “many internal improvements of primary magnitude might be promoted by an authority operating throughout the Union, which cannot be effected as well, if at all, by an authority confirmed within the limits of a single State.”

The real heyday of federal action came with the presidency of Dwight Eisenhower.

Hamilton’s view was rejected by President James Madison and his successor, James Monroe, who both vetoed major infrastructure legislation passed by Congress. They believed that the Constitution did not grant the federal government the authority to fund infrastructure and that it would take a constitutional amendment to establish that authority. Yet as he vetoed the legislation in 1817, Madison acknowl­edged the need for infrastructure invest­ment, writing, “I am not unaware of the great importance of roads and canals and the improved navigation of water courses, and that a power in the National Legislature to provide for them might be exercised with signal advantage to the general prosperity.” Still, five years later, Monroe vetoed a bill that sought to repair a road running from Cumberland, Maryland, to Wheeling, in what was then western Virginia, arguing that the states through which the road passed should pay for it because Congress lacked the authority to do so. His veto helped end the Era of Good Feelings, when the country was united behind one party, the Democratic-Republicans, and contributed to the revival of partisan politics.

From that point all the way up to the early twentieth century, Washington provided little funding for roads and bridges. Yet with the growth in automobiles, the building of the Panama Canal, and, in 1907, a favorable ruling from the Supreme Court in Wilson v. Shaw on the constitutionality of federal spending for infrastructure, public support grew for federal action. In 1916, U.S. President Woodrow Wilson signed the landmark Federal Aid Road Act, the nation’s first federal highway funding legislation. Five years later, the Federal Highway Act of 1921 mandated a national highway system.

But the real heyday of federal action came with the presidency of Dwight Eisenhower. In 1919, as Petroski writes, Eisenhower had traveled in a military convoy from Washington, D.C., to San Francisco and was dismayed by what he saw. Later, during World War II, he encountered firsthand Germany’s superior Autobahn. After the war, as president, he made infrastructure spending a priority. In 1956, he signed a law that established the interstate highways that still traverse the United States. Eisenhower saw infrastructure as both an economic and a national security priority. On the campaign trail, he had argued that “a network of modern roads is as necessary to defense as it is to our national economy and personal safety.” The law itself made a similar point, reading, “Because of its primary importance to the national defense, the name of such system is hereby changed to the National System of Interstate and Defense Highways.”

Emergency personnel stand on the remains of the collapsed I-35W bridge in Minneapolis, Minnesota, August 2007.
Emergency personnel stand on the remains of the collapsed I-35W bridge in Minneapolis, Minnesota, August 2007.
Scott Cohen / Reuters

Eisenhower’s highways were part of a series of great infrastructure projects that helped usher in unprecedented prosperity. Government investment and private entrepreneurship laid railroads across the continent; built huge power plants, such as the Hoover Dam; and provided universal phone coverage. Those projects generated economic growth and united the nation.

But they required public investment, and in recent years, that has been lacking. Since Eisenhower launched the inter­state highway system, and President Ronald Reagan expanded it, the national gas tax has served as the main source of federal funding for highways, bridges, and public transit. The gas tax used to generate substantial revenue, and it could still raise significant funds now to upgrade the country’s roads, fix its bridges, and expand its transit systems. Gasoline sales are at an all-time high. Although electric cars may one day dominate the roads, today they make up less than one out of every 1,000 cars, and SUV sales are at record levels. Yet Congress has not raised the tax in more than 20 years. Since the tax is not indexed to inflation, the rate actually declines in real terms every year. In 1993, it made up almost 20 percent of the price paid at the pump; today, it makes up around seven percent.

The United States is still waiting for a genuine infrastruc­ture stimulus.

In 2009, U.S. President Barack Obama signed an $800 billion stimulus bill. It provided some additional infrastructure funding, but, as Petroski points out, only four percent of the bill’s total went to transportation projects, roughly the same amount of federal investment in a non-stimulus year. Instead, Washington used most of the stimulus for temporary tax cuts for individuals and businesses, amorphous “aid to the states,” and a host of other initiatives unrelated to infra­structure investment. The United States is still waiting for a genuine infrastruc­ture stimulus.


Even when Washington has invested in public infrastructure, the results have sometimes been disappointing. Historically, the federal government has funded infrastructure projects by distributing money to the states, thus empowering state legislatures and bureaucrats in the executive branch at the state level to decide which projects to fund.

Congress did occasionally pick projects itself, through a process known as “earmarking.” Although earmarking always made up a relatively small portion of overall government spending, well under five percent, it grew dramatically: there were six earmarks in a 1980s high­way bill and over 6,000 in a similar bill in 2005. The most famous earmark was the “bridge to nowhere,” a proposed $400 million project in Alaska, which prompted so much outrage that Congress ultimately banned the practice of earmark­ing. This has shifted even more control over infrastructure spending to the federal executive branch and to elected officials at the state and local levels.

At those levels, Petroski demonstrates, vocal minorities wield great influence over infrastructure choices, often in undesirable ways. Take speed bumps on local roads. Small groups of committed citizens who want to discourage speed­ing in their neighborhoods often push for municipalities to install such obstacles. But as Petroski points out, the money spent on them could be better used to fill potholes. Speed bumps, he writes, create “a classic infrastructural dilemma of choice: to spend money deliberately raising for a complaining individual or small group an otherwise undesirable bump in the road—or to spend it for the good of all by filling unwanted holes in the pavement.”

A pothole in Los Angeles, California, February 2016.
A pothole in Los Angeles, California, February 2016.
Mario Anzuoni / Reuters

Given these problems with the political processes that shape infra­structure, it’s worth considering whether engineers ought to have more influence on decisions about what to build and how to pay for it. After all, a 2013 Pew Research Center poll found that 63 percent of Americans believe that engineers con­tribute “a lot” to society, a figure that far exceeds that for public trust in government, which has hovered at around 20 percent for several years.

Yet left to their own devices, engineers often overshoot their budgets, creating “signature projects” that priv­ilege aesthetic choices over the cost of construction and maintenance. Petroski never resolves how to balance these competing incentives. Policymakers should give engineers free rein to innovate within a constrained budget. Both should ensure that decisions made at the beginning of a project, which have an outsize impact on its success or failure, are not just cost effective in the short run but likely to yield low operating and maintenance costs in the long run.


The problems that plague American infrastructure are deep-seated and com­plex. Yet there is a way out. Wash­ington and the public must recognize that world-class infrastruc­ture does not come cheap. High-quality infrastructure is vital to global economic competi­tiveness, and the United States is falling behind. The United States invests less than two percent of its GDP in infras­tructure; Europe, by contrast, invests five per­cent. Furthermore, the bureau­cratic system that oversees public infra­structure spending has become hope­lessly “siloed,” with separate agencies at each level of government dedicated to different modes of transport. Each has its own stakeholders, champions, and opponents. It’s a wasteful, inefficient system; reforming it will require several steps.

Wash­ington and the public must recognize that world-class infrastruc­ture does not come cheap.

First, governments should eliminate silos. A unified department should merge the federal highway, transit, aviation, maritime, and railroad administrations; the Army Corps of Engineers, which controls investment in ports; and the Environmental Protection Agency’s water programs, which provide federal funding for sewer systems and drinking water. This unified depart­ment of infra­structure should incentivize state and local authorities to make smarter choices with federal funding. For instance, it could coordinate the timing of different projects, such as a sewer-line expansion and a road repair, so that the government has to dig only once. It should also consider instituting a so-called corridor-based approach, similar to the one the United Kingdom uses, in which the government evaluates a set of projects designed to solve a particular problem and chooses the most cost effective among them. For example, to improve the flow of people between Washington, D.C., and New York City, policymakers should com­pare the costs and benefits of investing in highways, high-speed rail, increased airport capacity, and more efficient freight rail and shipping to decide how best to solve the problem, rather than doling out funds to each of the various trans­portation agencies without coordination.

Perhaps most crucial, however, is the need for the United States to invest more. For some time now, it has been depreciating its capital stock, free-riding on the efforts of past generations. When previous generations built the interstate highway system, they faced a choice: pay for it themselves with a large gas tax and bequeath it to their children, or issue bonds that they and their children would later have to repay. Eisenhower, thinking it necessary to make his pro­posal politically palatable, proposed relying on bonds. Yet Congress instead chose the former, asking Americans to pay more than their fair share to lay the groundwork for the prosperity of future generations. Presidents from both parties—Eisenhower, Reagan, George H. W. Bush, and Bill Clinton—have raised the federal gas tax and increased infrastructure investment. The public is much more willing to support user fees that are dedicated to infrastructure projects than amorphous tax-and-spend proposals. So today’s Republicans need to exempt user fees, such as the gas tax, from their broader “no taxes” mantra. And Democrats need to accept that the gas tax is meant solely for infrastructure and not for other pur­poses (in 1993, Clinton raised the gas tax for deficit reduction).

Finally, policymakers should not fear the private sector. Good ideas frequently come from sources outside the govern­ment. Cities and states should allow both the public and the private sectors to submit unsolicited proposals for innovative infrastructure projects. Of course, these projects would have to be rigorously and publicly evaluated. But the fact that private companies are moti­vated by profit is no reason for the government to ignore them.

The United States has reached a fork in the road. It can let its infrastructure crumble, its bridges collapse, and its roads grow ever more congested. Or it can raise the gas tax, allow people to travel faster and more safely, and grow its economy. Which path Washington chooses will make all the difference.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • Aaron Klein is a Fellow in Economic Studies and Policy Director of the Initiative on Business and Public Policy at the Brookings Institution.
  • More By Aaron Klein