The current conflict in Washington, dealing with the expenditure of hundreds of billions of dollars, is the much anticipated, frequently promised, but poorly defined ‘infrastructure bill’. As it stands, two versions are possible – one preferred by centrist Democrats and a few moderate Republicans that limits spending to a few categories, especially roads, ports, and the like, and a much larger one encompassing topics like ‘human infrastructure’, including expanded Medicare and other services not typically considered under the title of infrastructure. It remains to be seen if either bill will muster enough votes in both the closely-divided Senate and the Democratic majority House – doing so will require getting a broad consensus within the Democratic party, a unenviable task in a party that ranges from Joe Manchin to Alexandria Ocasio-Cortez.
The pressure is on, though – partly because such a bill has been promised for years, and because it is clear that in many ways America’s basic economic infrastructure is falling behind other countries – not just in rail, where the US has lagged Europe and Asia, but also in broadband internet and even with roads and bridges. Moderates frequently cite the high federal deficit as a reason to forestall further spending, while proponents of more spending note that interest rates are historically low and so the real price of taking on new debt is small. However, what neither side explains is precisely why, given the immense structural hurdles to getting anything done in Congress, the federal government is the only entity that can make this happen.
Part of the answer lies in the nature of infrastructure. Improving a road from Elko to Reno, Nevada may or may not be worth the trouble – few people live along the route and the cities it is connecting are relatively small and probably do little business with one another. But, if Elko is already connected to Salt Lake City, and Reno is already connected to San Francisco, then the value of that additional link is immense – instead of two big small towns, it’s helping to connect to large and growing metropolitan areas. However, the benefits of this connection are not going to be felt entirely or even primarily by Nevada – rather, both Utah and California will be benefitted from their great connectivity.
These benefits, which increase the value of a given product the more its linked to others – are called network effects. It is for this reason that the federal government had to subsidize first the railroad connection between Utah and California, and subsequently the interstate – it would be irrational for Nevada to pay for that all on its own, but beneficially to the whole country if the link is built. For this reason, large scale physical infrastructure has long been a project of the federal government, and this pattern only accelerated in the 1950s with the construction of the interstate highway system. These kinds of projects cannot be efficiently undertaken by the states alone.
However, much of what has been proposed in the larger infrastructure bill being considered by the House is not subject to the efficiency problems of network effects – expanding medical and home based care, childhood education, local commuter rail and bus lines, and many other infrastructure investments can be effectively undertaken by state and local governments. A train from San Francisco to Los Angeles, for example, would benefit primarily Californians, and thus if it is a good investment for the federal government, it would also be a good investment for California, should the federal government fall through. The same is true of more local projects, and especially true of services provided to individuals.
So, why instead to proponents of these programs instead prefer federal funding? Well, of course it’s natural to seek money wherever you can get it, and if you can get it without taxing your own voters so much the better. But while that might be a reason to ask for funds, it doesn’t explain making projects dependent on federal funding.
The reason is that states attempting to fund such infrastructure projects are usually funding them poorly. Most states rely on sales or income taxes – paying for a new bridge or new bus system or expanded childcare all ultimately means raising such taxes – and raising those taxes can harm the economy, driving commerce away or underground. In extreme cases, the fear is that high earners, who disproportionately fund these projects with their taxes, will leave the state entirely. While federal income taxes create some of the same efficiency, the difficulty involved in packing up and leaving the country is far greater and so at least that fear is diminished – and of course the federal government can always lean on very low interest loans when taxes fall short.
Initially it seems like this should only handicap a relatively inefficient project – a truly transformative rail line or childcare center will more than pay for itself with new economic growth, far offsetting any tax increases, right? Well, possibly, but the problem is where those benefits go. A new train line might (indeed, probably will) lead to a jump in the value of land served by its stations. A childcare center in the neighborhood will make the whole neighborhood more desirable, driving up land values there. The problem is that States collect only a tiny portion of the benefits, perhaps as high as two but usually closer to one percent of the increased land value per year in property taxes. If landowners can generate even 5% of their land’s value in revenue every year – a very conservative figure – they end up enjoying 80% of the land value gains created by infrastructure spending. Meanwhile, even if workers find their incomes rising due to the greater productivity of the land, they are saddled with paying higher rents as well, making it politically difficult to collect taxes on their increased wages, either.
The solution is straightforward – fund such infrastructure problems primarily by capturing the land value that they create. This can be accomplished most easily in the US context by simply raising taxes on land value (ideally introducing a split rate so that the taxes on built property, the value of which is less impacted by the investment, the same). Worldwide, cities that funded their infrastructure investment via land value capture also had some of the best outcomes in terms of the valuable transit and educational systems they created – Singapore and Hong Kong, for instance, stand out by capturing most of their land value by municipal ownership and leasing of land, and are also notably for having among the best transportation and education systems in the world.
Economist Joseph Stiglitz has gone so far as to assert what he calls the Henry George Theorem: that under the right conditions, investment can pay for itself simply from land value capture. The best part? States doing so avoid many of the problems inherent in other funding mechanisms. The land, unlike high income individuals, isn’t going anywhere, and the tax can’t diminish the quantity of land available (a traditional property tax, however, can reduce the incentive to build new buildings – hence the value of split rate taxes).
No matter how the current situation in congress turns out, it is basically inevitable that some people will be disappointed to see their favored programs cut from the final deal. The nature of the bicameral system and our current polarization basically ensures disappointment on both sides. Fortunately, for many kinds of projects there is an alternative to federal funding – capturing the value of land created by infrastructure to help such projects ‘pay for themselves.’