There is widespread agreement that our infrastructure — roads, bridges, water systems, railroads and mass transit, flood control and power grids — is in great need of repair, rebuilding or replacement. The American Society of Civil Engineers (ASCE) gives it an overall grade of D+, pretty poor for the world’s leading economy.
There is much less agreement on how to pay for it, especially when we consider what needs to be done. The proposals of $50 billion in infrastructure spending by both major presidential candidates fall far short of what’s needed. The highly successfully Interstate Highway Program proposed by President Eisenhower amounted to one-half to one percent of GDP (Gross Domestic Product) during its initial years.
U.S. GDP in 2016 will be over $18 trillion. If we were to match the Interstate Highway Program, that would be between $90 billion and $180 billion annually for 10 years or $900 billion to $1.8 trillion in total. Indeed the ASCE estimates that about $950 billion is needed right away. No wonder politicians shy away from this subject.
Fortunately, there is a straightforward way to raise the necessary funding. Let us create a federal infrastructure bank, modeled on some long-standing federal agencies. The infrastructure bank would sell bonds to the public and then relend the money to states, municipalities and agencies like the Port Authority at low interest rates and for long terms.
Here is the twist. U.S. corporations are parking huge amounts of cash overseas at low, even negative, interest rates. Encourage these companies to bring their money home by buying federal infrastructure bank bonds. As long as they hold them, the money will not be considered repatriated. They would undoubtedly be willing to accept one percent on a ten-year bond under these conditions.
Over the longer-term, these infrastructure investments will improve the productivity of the U.S. economy. In the near-term, it will provide substantial employment for people with many levels of skills and education.
Perhaps such concrete spending will result in the old-fashioned 3X multiplier we counted on in the 1950s and 1960s. Debt service on $1 trillion would amount at most to $10 to $20 billion per year. Surely this can be afforded in our $2 trillion federal budget.
To do such a major program right will require one or two years of careful planning and will need to involve may different constituencies. The rebuilding program itself will take at least 10 years.
The time to start is now.
Thomas Synnott
Cornwall
Editor’s note: Synnott is an adjunct professor of industrial engineering at The Cooper Union for the Advancement of Science and Art in New York.