Monday, February 10, 2014

Fred Thompson: Is A Separate Capital Budget The Solution To Our Infrastructure Problems?

Once again, Fred Thompson makes a timely contribution to the blog:

Noah Smith, one of my favorite economic bloggers, recently posted about the need to fix our infrastructure and proposed that one way to do this is by separating the capital budget from the regular budget:
We need to rebuild our infrastructure, and now is the perfect time to do it. Interest rates are at historic lows, but they are unlikely to stay there forever... 
But infrastructure budgets have been cut, not expanded. Why? One reason is that in the race to cut the deficit, infrastructure spending has been lumped in with other types of spending. That is a tragic mistake. Unlike government “transfers,” which simply take money from person A and give it to person B, infrastructure leaves us with something that helps the private sector do business, and thus boosts our GDP growth. Infrastructure is a small percentage of overall federal spending, but tends to be a politically easy target. 
One idea to boost infrastructure spending, therefore, is to treat government investments differently from other kinds of government spending by having a separate capital budget.
This grabbed my attention. We all have our manias. Mine is infrastructure investment, especially our inability to make large-scale infrastructure investments, which reduces economic growth and, based on studies by Duflo and Pande and Lipscomb et al., may increase poverty. For me the issue isn’t limited to public investment. We strangle all sorts of large-scale infrastructure investments. Here, the poster child is Keystone XL. Recently, the U.S. State Department’s final environmental review of the proposed Keystone XL pipeline was released for comment. It concludes that Keystone XL would have little or no effect on the rate of extraction of oil sands or on the consumption of oil, that its main effect would be to increase the safety with which that oil is moved about (a little) and its efficiency (a lot), and implies that the opposition to this good thing is largely symbolic. The report’s basic economic argument is that, even if the oil from the tar sands were shipped by rail, its cost would still be less than the current or expected future prices. Consequently, building the pipeline won’t affect the marginal price/cost of oil and, therefore, final supply or demand.
As I see it. The main thing that is strangling large-scale infrastructure investment is that we have politicized these decisions in a system of governance that is highly biased in favor of the status quo. So far as public investment is concerned, the big problem seems to be an inability to manage IT and ICT investments. Even so, granting those claims, wouldn’t it make sense “to treat government investments differently from other kinds of government spending by having a separate capital budget”? Of course it would and, for the most part, we do. More than 80 percent of the US’ nonmilitary public infrastructure is owned by state and local governments. This means that, if we have insufficient public infrastructure, it’s primarily a state and local matter. Most state and local governments already have separate capital budgets.
(Parenthetically, it might be mentioned that the Feds offset the presumed budgetary bias against investment projects by annualizing capital costs in authorizing ‘bricks and mortar’ projects and in appropriating funds for them.)
Therefore, let’s come down to the level where it matters, Oregon, and to an infrastructure issue I know something about, highways. Analysts at the Department of Transportation (DOT – and more disinterested bodies, like the Oregon Transportation Research and Education Consortium – OTREC) are quite concerned about the sustainability of our highway transportation network. They believe that the social returns on certain kinds of highway investments are quite high: maintenance, > 35 percent; projects oriented to reducing urban congestion, > 15 percent, etc. In contrast, ignoring effects on aggregate demand, most other highway projects produce relatively low or negative returns. Because the lion’s share of state money goes to maintenance, this evidence is entirely consistent with the conclusion that we ought to be investing more and that now is a very good time to make such investments (in addition to low borrowing costs, the state doesn’t have to bid fully employed resources away from the private sector). Moreover, the relatively high payoff to the maintenance portion of the DOT budget tends to reinforce Noah Smith’s argument: maintenance seems to be where we are under-spending the most and maintenance spending is usually charged to the current account rather than to the capital account.

Nevertheless, there are a couple of policy issues that are probably more salient than is budget format to fixing infrastructure problems. In the first place, maintenance investment is far less risky and seems to have higher expected rates of return than other highway investments, but gets little or no federal matching, while new construction projects get treated pretty generously by the Feds: about 80 percent of the cost of a project (up to some predetermined limit, then none of the costs, so that the state bears most of the risk of cost overruns, which tend to be high for investments that break new ground, whether literally or figuratively). The solution seems obvious, shift federal transportation subsidies to a formula-based grant, which is how most states allocate motor fuel taxes to local jurisdictions. At the very least, federal matching shares should probably be lowered for new construction and raised for highway maintenance.
Second, highway engineers tell us that road wear is a cubic function of axle weight and a quadratic function of road speed. Oregon’s weight-use-mile tax, imposed on commercial vehicles, approximates the damage they do to the highways (and also provides a very real incentive to reduce axle weight and to comply with the 55 mph speed limit on trucks, thereby, also reducing highway maintenance costs). The gas tax once did pretty much the same thing for private vehicles, insofar as speed and vehicle weight were the principal determinants of fuel consumption. However, the legislature’s consistent failure to raise motor fuel taxes in tandem with construction costs, together with reduced private driving and the shift to more fuel-efficient cars, has depleted the state’s highway fund (it has also increased uncertainty about the benefits of new highway construction and, therefore, the riskiness of these projects, which was already high on the cost side). The solution to this problem lies in increasing revenue.
Currently, Oregon’s DOT is experimenting with GPS systems that would monitor how and when vehicles use the highways and allow the state to bill their owners directly for the road damage their vehicles cause. This would permit revenue from highway x to be devoted to maintaining highway x, and give DOT a better guide to allocating funds for new construction. Furthermore, these charges could be varied by time of day or week, thereby lessening congestion, further reducing the need for new construction. It might make sense to delay most new, large-scale highway construction projects in the state, especially those justified by high congestion costs, until we see how these experiments play out – technically and politically. A moratorium on new construction would also free up a little more money for maintenance projects.
Finally, there is one other point that should be borne in mind: most of the benefits to infrastructure investments accrue directly to users or indirectly to their customers (or their customer’s customers). Insofar as highways are concerned, these benefits accrue about equally to private automobile operators and commercial freight handlers (and/or their customers). If public investment really does produce returns of 20-30 percent, one would expect producers of primary products, manufacturers and freight handlers to be clamoring for the state to impose taxes or float bonds to upgrade roads and highways. The effect on business profits would be high and the cost to business low. They are not or, at least, I don’t hear the clamor. What I hear the business lobby saying is that tax rates (and regulatory burdens) are too high, not that public investment is too low. If that’s what business owners and their employees truly believe, perhaps, it might be wise to treat claims about high payoffs to infrastructure investments (especially risky, big new projects) with a modicum of skepticism.


Doug Gabbard said...

Based on my experience working with local governments in Oregon, I have to say that your point about prioritizing maintenance investment is spot on. I like to imagine a world in which "project cost" is not limited to initial construction, but rather includes the present cost of construction, maintenance, and perpetual replacement. Whenever you construct an asset, you have to make a deposit in a fund that can be used only for the maintenance and replacement of that asset. If such a requirement existed, we would never be tempted by the illusion that sprawl is affordable.

Fred Thompson said...

Doug, Thanks for your comment. Do you (or anyone out there) know of any hard evidence of skimping on maintenance? What I have are a lifetime of anecdotes, which as I remind my students, isn't a plural form of data.

Patrick Emerson said...

That's funny, I use the plural of anecdote is data all the time (thinks to Krugman) - but do so ironically most of the time.

Doug Gabbard said...

Data on deferred maintenance is difficult to get unless you stumble across some internal condition assessment. It's much easier to quantify deferred replacement. With a budget document or capital improvement plan, you can see planned expenditures by type of infrastructure. Then, with the help of a CAFR, you can compare the planned expenditures with the actual depreciation booked for that category of asset. If spending is less then depreciation, you are losing ground. The City of Salem, for example, will spend $9.5 million per year for water and sewer infrastructure over the next five years. Meanwhile, annual depreciation for those assets is $16.2 million. Of course, if your population is growing, you might be losing ground even if your spending is greater than depreciation.

Fred Thompson said...

E.A. Littrell and I actually wrote a piece on how to improve the CAFR to generate this sort of info: "Fixed Assets Reporting" and "Reply", Public Budgeting & Finance. 18/1 (Spring 1998) 94-98; 124-127.

Fred Thompson said...