From President Trump’s Camp David retreat with cabinet officials and congressional leaders at the beginning of this year, word emerged that the president and his advisers are divided on the best policy for infrastructure. Gary Cohn, director of the National Economic Council, presented a detailed plan to make $200 billion in federal investments in order to unleash $1 trillion of total infrastructure investment through public–private partnerships, a plan that has now been leaked to the media. The president himself, meanwhile, reportedly prefers a more straightforward national building program.

Any discussion of infrastructure spending needs to recognize the stark reality of the American cost disease. As explained in a December New York Times report on the New York subway, when the United States builds infrastructure, it often costs more than any similar industrialized country would consider spending. New York City brings the cost disease to its highest fever, but even cities that excel at cost containment by American standards would have their numbers thrown out on their ear in many other countries. Liberals sometimes wave away cost concerns by reemphasizing the need for any particular project, and conservatives sometimes blithely presume that any project is wasted money. But all parties involved must recognize and address the cost disease, which drastically reduces the amount of infrastructure Americans can get out of any particular budget figure. Building a tunnel six times more expensive than one in France means that you get one-sixth the tunnel that you should. As transit researcher Alon Levy has shown, the American cost disease is real, and the situation is dire.

While the entire basis of these cost overruns is still not known, it is clear that American labor costs significantly contribute to project-cost inflation. Prevailing-wage standards, set under the Davis-Bacon Act, are a frequent target of the ire of conservatives, who charge that the requirements empower unions to run up prices and drain the public purse. These prevailing wages certainly inflate costs, and repeal or reform of Davis-Bacon would help the taxpayer receive a fair value for his investment. However, competitor nations such as France and Spain cannot be said to possess weak unions or ungenerous labor laws, and those nations still manage to build infrastructure for a fraction of American per-mile costs. Davis-Bacon repeal is no silver bullet, and further reforms will be needed.

As Jarrett Walker explains in his book Human Transit, operation costs in industrialized countries are dominated by labor. Simply put, people are expensive in rich countries, and hiring workers requires paying significant wages and benefits. Thus, one of the most effective ways to exercise fiscal prudence is to ensure that human personnel are not wasted in their transit work. Unfortunately, wasting person-hours seems to be American transit’s most consistent accomplishment. Subway trains that should be able to be run by computer often must be managed by one or two drivers, and tunnel-digging machines that the French operate with fewer than ten people are managed by more than two dozen well-compensated Americans.

Capital costs, meanwhile, are also inflated by “buy American” procurement rules attached to federal infrastructure financing. When local governments take advantage of federal grants or loans to expand their infrastructure, they are required to buy at least 60 percent of rolling-stock components, such as rail cars and buses, from American manufacturers. (Current law requires that level to rise to 70 percent by 2020.) Manufactured goods, on the other hand, must be 100 percent American in materials and manufacture. While well-intentioned in their concern for American manufacturing, such policies can further inflate the cost of infrastructure. For example, according to the American Action Forum, Americans pay 34 percent more for their metro cars than the global average. Even with such policies in place, contracts often go to the most competitive global firms, which then set up separate manufacturing facilities in the United States. The profits are passed back to the foreign headquarters, while taxpayers pay the price for not being able to access the normal industry supply chains.

For the American taxpayer to receive assurance that his money is being spent wisely, any major infrastructure investments should be accompanied by actions to treat the cost disease. Already, the governors of New York and New Jersey are expressing indignation that the Trump administration has renounced an Obama-administration plan to fund half the ballooning cost of their new tunnel-building program. They would do well to turn that indignation toward their own transit authorities for wasting historic amounts of money. The 50 percent of the projected cost that the governors were already willing for their states to pay should be beyond sufficient to complete the entire project, and then the concerned states would not have to go through federal procurement channels at all.

Furthermore, when the national government picks up significant portions of the tab, it often incentivizes projects that should never have been undertaken at all. In 2010, self-described “recovering engineer” Charles Marohn pointed to a project in Staples, Minn., that cost $9.8 million to build an overpass above a railroad in order to connect two state roads and ease the congestion that came from waiting for train cars to pass. Staples has a population of 3,000. The federal government offered it $8.8 million for the project, and the state of Minnesota chipped in for the other $1 million. While the good people of Staples might enjoy their uncongested cross-town connection, Marohn wryly predicted that if they “were asked to simply pay 10 percent of the cost, . . . this project would not be happening.” Most federally supported projects are not so heavily subsidized, but a more customary 80 percent federal match was enough for the Louisiana city of Shreveport to attempt the decidedly retro project of bulldozing a working-class, mostly black neighborhood to build an urban highway connector through the city in the name of economic development.

The good news is that even as Washington continues to argue over the best way to make infrastructure investments, private actors are already emerging to offer innovative means of transportation. Whether with cars, trains, or the humble bicycle, new companies are stepping up to unleash American mobility, and each innovation holds the potential to reshape demand for other infrastructure components as people adjust their living and travel patterns.

Virginia’s McAlester’s Field Guide to American Houses conveys this recurring effect in a few pages as it details the development of American neighborhoods. Towns and cities were first built to be accessed most regularly on foot, meaning that homes, workplaces, and shops necessarily intermingled, all built on relatively narrow plots of land. The advent of horse-pulled streetcars stretched out development along a commuting pattern that opened up land for neighborhoods of residential rowhouses. The electric streetcar created spokes of development, populated by detached houses, emanating out from city centers. Because the neighborhoods still had to be navigated on foot after residents disembarked from the streetcar, though, homes in these early suburbs were built on relatively narrow lots. The automobile filled in the land between the streetcar spokes and eventually pushed out to fields opened up by freshly paved highways, allowing direct access to ranch houses and split-levels built on much wider lots.

We may now be approaching a similar point of transformative change through the explosion of private transportation services. The most well-known newcomers to the transportation scene are ridesharing companies such as Uber and Lyft. By enabling people to turn their personal cars into de facto taxis, the services upended the long-standing taxicab-medallion cartel system and tapped an explosive reserve of unmet consumer demand for point-to-point mobility. Uber and Lyft are also among the most active investors in what is widely projected to be the next phase of the automobile’s development: the autonomous vehicle.

Cars are far from the only mode of transportation undergoing significant innovations, however. In Florida, the “Brightline,” the first private passenger-rail project to be constructed in the United States in a century, is taking paying customers. The privately funded, financed, built, and operated line connects West Palm Beach and Fort Lauderdale, with stations in Miami and Orlando set to follow over the next few years. And Texas Central recently passed its first major federal environmental review on its way to constructing the first true high-speed-rail system on the American continent. Texas Central will connect Houston and Dallas, the fourth- and fifth-largest metro areas in the country, and it will run without state subsidies.

Creativity is also bubbling up in the bicycle world, as many American urban centers have seen bikeshare programs emerge. The market appears to have decided that the time is ripe for such systems to make money. Companies such as Ofo, Mobike, and Limebike are surging into city centers and finding huge numbers of customers. Seattle, for instance, had just wound down its failed city-run bikeshare program when three dockless bikeshare companies filled the void, building the second-largest city bikeshare fleet in the country without spending a single public dime. In China, such dockless bicycle companies, which offer cheap and easy last-mile connections, have dried up the ridesharing services’ market in short-range trips and driven demand back into transit.

To commit enormous federal funds right now while the forms of American mobility are so rapidly shifting, then, would be to bet one’s stack of chips while one’s hand is still being dealt. Instead of rushing to build new roads and highways based on past habits, we should turn our focus to rescuing and reinforcing the investments we have already made. The “crumbling” bridges and roads that President Trump decries will not crumble any less because a new bypass is being built on the other side of town, and maintenance liabilities are already outstripping many communities’ capacity.

Instead of starting another highway-building program, the United States would do well to focus on maintenance, to devolve planning and funding decisions to localities, and to ensure that the playing field is level enough to accommodate whichever road the future of transportation goes down.

– Mr. Coppage is a visiting senior fellow at the R Street Institute, where he studies conservative urbanism and the built environment.