Both the Economist and the Washington Post have recently denounced the poor state of US infrastructure; it is obviously becoming fashionable to do so. However, much of their criticism is ill-founded. Infrastructure failures are inevitable because infrastructure decisions have become politicized and suffer from the results of that politicization. Infrastructure investment has thus become a particularly economically inefficient area, the reform of which would add greatly to national and global wealth.
When the Duke of Bridgewater built the Bridgewater canal in 1758 to transport coal from his mines at Worsley to newly industrial Manchester he didn’t need political permission to do so. Neither did he raise outside capital; the canal was financed out of his own resources. Hence his own entrepreneurial judgment that money could be made from the canal was the sole factor in his decision. His investment was spectacularly successful; the price of coal at Manchester dropped by three quarters within a year after the canal was completed in 1763. The canal operation remained a profitable independent company until being sold to the Manchester Ship Canal (also private) in 1885 – the canal itself remains active to this day.
Since around 1900, most infrastructure investment has been financed by the public sector. This has introduced a number of additional complications that the Duke of Bridgewater did not need to consider. Politicians are elected on a short term electoral cycle, so long-term projects have little appeal – even the modest-scale Bridgewater canal took 5 years from conception to completion, longer than most electoral cycles. Much of their power comes locally, so projects that benefit several jurisdictions are difficult to get off the ground and small projects benefiting only the immediate community are preferred.
Politicians are also almost entirely ruled by popular fashion. Thus in periods when infrastructure is fashionable, say 1920-1960, it is built in great profusion (think for example of the complex spaghetti of poorly planned, poorly integrated and environmentally destructive highways built by New York’s Robert Moses during that period). Conversely in periods such as that since 1970 in which environmentalist and “not in my backyard” objections have been given priority, infrastructure spending is neglected.
As an example of how this works, consider the state of Connecticut. Connecticut benefited enormously from the first wave of suburbanization; it already had the New Haven Railroad in place and no state income tax, so until 1980 it filled up rapidly with the more affluent toilers in the New York metroplex. The Hudson Valley had equally good infrastructure, in the form of the local rail line, but its towns of Ossining, Peekskill, Beacon and Poughkeepsie were blighted by the remnants of earlier industrial development; in any case it did not offer wealthy New Yorkers the same tax benefits as Connecticut.
The downside of having no income tax was that the state was always short of money, and so built infrastructure less aggressively than New York. It also made the mistake of attracting financial services company headquarters to Stamford, and then building no road infrastructure to support them – in the 1980s, when the major influx occurred, it still had no state income tax, and was attempting to maintain a Massachusetts level of social services on a New Hampshire tax base.
Connecticut voters foolishly elected Lowell Weicker, a leftist independent as Governor in 1990, and as a result they were saddled with a state income tax but no improvement in infrastructure (the Democrats had been terrified of introducing a state income tax themselves for fear of facing the wrath of the voters.) Nevertheless, Connecticut’s state income tax remained lower than New York’s, and its suburbs of Greenwich and Westport in particular were highly fashionable, so in the late 1990s and onward it attracted large numbers of hedge fund managers, who found they could carry out their nefarious but profitable activities from the comfort of Connecticut, without the bother of commuting into Manhattan.
The result of economic growth without infrastructure provision has been gridlock, more or less permanent except at 5am Sundays, on the Connecticut Turnpike between Greenwich and New Haven, a distance of 45 miles. Even outside the rush hour, it can take about 2½ hours to navigate the Turnpike, which now runs at over 200% of “capacity.” There are few alternatives available, since Connecticut has built a maze of small highways that may serve local voters but don’t go anywhere and relieve little if any of the congestion in East-West traffic. The cost of this gridlock is immense, not only in the time of drivers but in the additional environmental damage caused by hundreds of thousands of vehicles proceeding in low gear for several hours.
The solution is simple if you look at a map. Indeed it was contemplated by Moses half a century ago, even before the Connecticut Turnpike was completed, when he built the Sunken Meadow State Parkway across Long Island with a spur facing Connecticut. However even Moses, a Yale graduate born in New Haven, could not get the Connecticut state government to cooperate, so he doubtless ground his teeth in frustration and proceeded to demolishing more of the South Bronx. Connecticut needs to be connected to Long Island by a bridge, and connecting highways and another bridge need to be built linking Long Island to New Jersey.
The principal purpose of such construction would be to allow traffic to move from New England to the Mid-Atlantic states without having to drive 270 degrees round New York, a city that is particularly awkwardly situated for modern road traffic. A bridge could be built west of New London to the tip of Long Island, crossing 12 miles of water, from East Haven, crossing 20 miles of water (but connecting conveniently with Connecticut’s meager north-south road network) or from Fairfield to Moses’ Sunken Meadow State Parkway, covering 14 miles of water. At the New Jersey end of Long Island, a bridge from Rockaway Point to Sandy Hook would cross only 8 miles of water. Short links would connect the Connecticut bridge terminus to the Connecticut Turnpike, the New Jersey terminus to the Garden State Parkway (or a longer link to the New Jersey Turnpike) and the Long Island termini to the island’s extensive road network.
Constructing such bridges would not be particularly difficult or expensive – the waters to be crossed are much shorter than that crossed by the Chesapeake Bay Bridge-Tunnel, completed in 1965, let alone the Channel or Seikan tunnels. As well as removing the 110 mile New York circuit for long distance north-south travelers, the bridges would relieve the Connecticut Turnpike of most long-distance traffic, allowing the remaining locals to enjoy their gridlock in peace. They would also free Long Islanders from long-term imprisonment, allowing them to visit the rest of the United States without driving through New York.
Such a project would be dear to the heart of the Duke of Bridgewater. However it is unlikely to proceed because it cuts across the jurisdictions of three different states, would take a decade to build and would run into enormous opposition from various local interests, as well as from environmentalists who would have six different court systems and five appeals courts in which to harass it. Senator Barack Obama has proposed a $60 billion infrastructure fund at the federal level. This project would be an obvious candidate for such a fund, but is unlikely to fare well there because of the diffuse nature of its beneficiaries – even at the federal level it is easier to fund “pork” in a single district, so the local Congressman gets the benefit.
The other and better way to fund infrastructure is through the private sector. This only began to be squeezed out by public sector financing with the success of the Erie Canal, financed by New York State guaranteed bonds in 1817-25. From roughly 1930 to 1980, public sector financing was assumed for all major infrastructure projects, since the state was able to borrow more cheaply than the private sector. In addition, private sector finance, which had focused largely on debt through the nineteenth century, from the 1920s onwards focused increasingly on equity, a very expensive means of financing infrastructure needs. However after 1980 the increasing popularity of the private sector and budgetary constraints in the public sector swung the pendulum back towards private finance.
The ideal infrastructure finance involves both debt and equity, with the equity used as a cushion to provide assurance that the debt will be repaid. That should be readily available in a universe that includes junk bonds. The problem is that infrastructure brings only long term returns, which are generally fairly low but very secure. In addition, political harassment can hugely increase the cost of even the simplest infrastructure project, primarily by delaying it.
The Eurotunnel project, which ran so far over its cost estimates it was eventually forced to declare bankruptcy, is a prime example of an infrastructure financing gone wrong. It required agreement on even small details between two governments and two public sector rail operators. Its form, a rail-only tunnel, was chosen for political reasons when a road bridge was clearly economically preferable. Finally, it was selected through the kind of public sector bidding process that almost always results in costs running astronomically over the heavily massaged estimates that win the bid.
Even with the relevant governments lined up, private finance for infrastructure today is questionable, because of the excessively short-term orientation of most financiers. The infrastructure specialist Macquarie Bank’s approach, in which the project arranger extracts a high return quickly through financial engineering, works only when the investor takes over existing infrastructure in return for an up-front payment to the seller. The Macquarie structure can be very attractive to local governments with aging infrastructure and urgent budgetary needs; it is no solution when more than minimal new construction is involved.
There is however a natural match between the long term modest but fairly certain returns of infrastructure projects and the long term investment requirements of life assurance companies and pension funds, which have a need to diversify their assets beyond stock and bond markets. Rather than investing fiduciary money in fly-by-night short term operators in hedge funds and private equity funds, such institutions should form consortia to hire engineers and undertake infrastructure projects directly. As representatives of thousands or even millions of life policyholders or pensioners, they would have the political clout to deal with recalcitrant local governments. For them, infrastructure projects would provide that ultimate diversification: long term returns that were inflation-protected and independent of stock and bond markets, though not of the economy overall.
To build better infrastructure, we need superb engineers and sober, long term oriented, moderately paid fiduciaries. Not expensive fast-buck financiers and not governments. The problem is one of US economic structure, and it urgently needs to be solved.
-0-
(The Bear’s Lair is a weekly column that is intended to appear each Monday, an appropriately gloomy day of the week. Its rationale is that the proportion of “sell” recommendations put out by Wall Street houses remains far below that of “buy” recommendations. Accordingly, investors have an excess of positive information and very little negative information. The column thus takes the ursine view of life and the market, in the hope that it may be usefully different from what investors see elsewhere.)