Should America borrow trillions of dollars to build new infrastructure?
A number of commentators have recently advocated devoting significant federal dollars to rebuilding the nation’s infrastructure. The Obama Administration has followed these suggestions by implementing or proposing major spending programs in several areas including high speed rail, smart electric meters, and expanded broadband service.
Proponents make several arguments in favor of these initiatives. First, additional infrastructure would increase the nation’s productive capacity, leading to faster economic growth. Second, infrastructure projects create jobs, which would help reduce unemployment from its current high levels. Third, the nation’s current infrastructure could definitely use improvement. The American Society of Civil Engineers has rated various categories of the nation’s infrastructure. Bridges get the best grade with a C. The Society recommends spending $2.2 trillion over the next five years. Finally, interest rates are at historic lows with the government issuing 30-year bonds for under 5 percent.
There is another, more subtle, rationale as well. The recurring financial crises of the last two decades have been exacerbated, if not caused, by the enormous amount of “hot money” in the world’s financial system. This money crosses borders quickly to seek the highest rate of return, often through short-term speculation. It leaves just as quickly, often precipitating financial collapse. Much of this money has recently been invested in developing countries and commodities, sparking significant problems in both. If future crises are to be avoided, it is enormously important that this capital be devoted either to consumption in the rapidly developing countries or to productive investment anywhere rather than further speculation.
The idea certainly has merit, but several points caution against quick action. The first is that the United States is already on a path of unsustainable deficits. Adding trillions of dollars of more debt, no matter how good the cause, would only hasten the date of economic collapse, when America’s creditors are no longer willing to extend additional credit at rates that the government can afford. Even if the possibility of a Grecian financial crisis is discounted, the burden of debt already being placed on our future strongly argues for funding new programs by cutting existing spending rather than issuing new debt. It is not clear that the government is ready to do this.
One problem is that many people confuse investment with consumption. Given that a lot of the current economic problem is due to the fact that America borrowed a lot of money to finance consumption, it is very important that we not do more of the same. Any borrowing has to be productively invested, which means that it has to earn a rate of return. If an investment earns a rate of return higher than the cost of borrowing we are better off, provided we can safely manage the financing and liquidity problems. If it does not generate a high enough rate of return, we are worse off. Yet there has been very little analysis of the exact rate of return that different infrastructure projects might be expected to produce. A great deal of past spending has represented wasteful consumption.
The rate of return is strongly influenced by the cost of any investment. Federal requirements such as Davis-Bacon requirements for wages, environmental restrictions, and permitting processes often unnecessarily raise the cost of investments, thereby pushing many from the investment to the consumption category, lowering the rate of return on all, and reducing the amount of investment that Americans can get for for any given expenditure. Labor restrictions also reduce the number of jobs that are created on any project. One way to measure how important proponents of infrastructure are is to see whether they are willing to relax some of these restrictions on even a temporary basis so that we can get as much investment for the dollar as possible.
Some of infrastructure proposals recognize the current budget constraints by proposing loan guarantees rather than direct spending. There is a great danger, however, that the true cost of federal guarantees will be underestimated in order to make it seem like they are cheap. This is what was effectively done for Fannie Mae and Freddie Mac even though in those cases the guarantees were not explicit. The result is that taxpayers are likely to end up with a huge bill several years later when projects turn out to be riskier than was promised. Nor is it clear what market failure a loan guarantee addresses. Most builders, whether they are private companies or other governmental units are able to borrow on their own credit. Most likely the real purpose of a guarantee is to let the project borrow at the same low rate as the federal government even though it is a worse credit risk.
It is possible that investments in highways, rail, broadband, the grid, and other areas could be very productive. These could be funded through a combination of direct expenditures, explicit loan guarantees, and privatization. These are less likely to lead to hidden liabilities. But in order to make sure that the projects do generate a high rate of return, it is very important to approach these them as large, interconnected projects rather than as discrete pieces. For example any spending on the smart electric grid should follow the development of a national plan that shows how all the pieces of the grid fit together and that includes other reforms such as faster approval for transmission construction and charging higher rates for peak use. In the absence of broader planning, much of the money recently spent on smart meters will not generate a strong rate of return because in the absence of rate changes individuals still lack any incentive to shift their consumption to nonpeak times or because the meters being installed turn out not be be compatible with later technology. Similarly, highway expenditures would have a higher return if they are accompanied by congestion pricing and if they are wired to convey traffic conditions on a real-time basis and accommodate the greater use of IT in cars.
Finally, there is the continued risk that technology will outpace spending on infrastructure. Just imagine a country that spent billions on new land line phone systems just as the wireless revolution was happening. For instance, a modern highway system needs to incorporate the capability to carry not just vehicles, but also information at high speeds. It is not clear yet what the best technology for broadband expansion will be, but it is likely to be faster and less expensive than today’s choices. And the smart grid still needs standards and a comprehensive vision in order to adapt to whatever the future brings. Absent foresight we could find that much infrastructure spending is outmoded soon after it is installed.
Despite the budget problems, the current impediment toward more infrastructure investment is neither a lack of feasibility nor a lack of capital. The federal government can still borrow significant sums if bondholders think that the funds are going to investments that make the country richer in the long run. The problem is that voters just do not believe that Congress will spend the money wisely. Rather than propose vague but grandiose plans, the Administration needs to develop a well-thought out vision for each major part of the nation’s infrastructure, accompanied by policy changes that increase the value of the components while minimizing their cost. Then it will be in a position to make it case to Congress and the American people.