Predetermined policy recommendations

Thoughts prompted by Saving, cost control, and infrastructure by Scott Sumner. This is an issue that is currently almost surpassingly complex, mostly because the payoffs are so large, the principles so fundamental, and the inducements for obfuscation so extreme.

When it comes to infrastructure, we don't even know what we are talking about much less whether or how to address it. Is there an infrastructure crisis in the US? I don't know. All my adult life, there have been cycles of bewailing the dilapidated state of our infrastructure. But is it really that dilapidated? What's the evidence? Anecdotally, living and working across much of the country and averaging all those places, I can't make a clear case that infrastructure is that much worse than it was three of four decades ago. Sure, specific issues in specific locations but not an overall systemic decline. In fact, based solely on personal experience, I would have to say that the general condition of infrastructure has improved, not declined. But anecdote is no substitute for empirical evidence objectively collected. Comprehensive, robust standardized empirical evidence seems in short supply.

Sumner's argument is summarized:

Places like Singapore have nice infrastructure because they have pro-saving public policies and effective cost controls on construction projects. America has neither. As long as this is the state of affairs, we will not have top-notch infrastructure, no matter how much money the federal government throws at the problem.

The first thing I would like to see is some sort of realistic comparison of apples-to-apples. It is easy to look at gleaming public infrastructure in Singapore and say they must be doing something right while we are doing something wrong. But you have to do a 360 degree view of the picture. What did it cost them? What have been the benefits? Who has benefited? Who has lost? Who has paid how much? What is being compared to what?

Singapore is, too all intents and purposes, a modern state. Literally. At the end of WWII Singapore was a small, devastated island of only 880,000 people with virtually no infrastructure. It now is home to some 5.5 million people with everything built from scratch in those seventy years. Most of it in the past thirty-five years. Of course, it is gleaming. It is all new. None of what I am trying to say is intended to take away from the accomplishments of Singapore. The point is that if Sumner wishes to make his case on comparisons, he needs to be comparing apples-to-apples. The US has older infrastructure because the US is an older country.

There is systems perspective as well. It's not just about how old or new is the infrastructure. It is easy to see that major construction companies, unions, and local governments all have an incentive to increase infrastructure building, particularly if some or all the costs can be off-loaded onto someone else (such as the federal government and taxpayers across the country). That's one side of the ledger, the proponents who benefit. But we also have to look at the other side of the ledger, other projects such as education and health, which do not get funded, property rights which get subverted, etc.

I periodically read about just how many dams, bridges, roadways, sewer lines, airports, sewer lines, etc. need to be upgraded or replaced. But is that true? What is the average life span of a well constructed and well maintained infrastructure project? Fifty years? A hundred years? Just what percentage of bridge collapses, dam failures, sewer line erosions occur in a year? Blessedly, relatively few.

I have lived and travelled in London, Paris, Stockholm, Sydney, Amsterdam, Tokyo, Hong Kong, Singapore, Atlanta, New York, San Francisco, Philadelphia, Chicago, Dallas, etc. Every one of those cities has a range of trade-offs that they make between politics, policy, state power, human rights, citizen rights, etc. None of those trade-offs are easy and few of them are optimal.

Catch them at the right time in a thirty year business cycle and the impression is much more positive than on the down side of a cycle. In the 1970s, Britain's transportation infrastructure was abysmal. By the 2000s it was much, much better but inordinately expensive and with a narrower range of options than virtually any city in the US. France does much better at public infrastructure but with the loss of citizen property rights compared to the US.

Infrastructure does have to be maintained. Over decades it does have to be upgraded or replaced. There is no denying those realities. But at what cost and at what rate?

There are so many other issues that factor into the discussion. In this article, Sumner is leading off with the fact that infrastructure (in this case subway tunnels) cost nearly seven times more to build in New York City than they do in places like Singapore, London, Paris, etc.

On that pitiful foundation (talk about weak infrastructure), Sumner leaps to a coercive policy recommendation that we should have compulsory savings accounts, the argument being that increased savings will decrease the cost of capital for major construction projects.

That's economically true, forced saving would increase the supply of capital and reduce its cost. However, is scarce, expensive capital what is driving the construction costs in New York? Is capital indeed scarce? It seems to me that the answer is no and no. The world has been awash with capital seeking returns for more than a decade. It is cheap already. Scarcity and expense of capital doesn't seem to be a constraint. And how much does the cost of capital actually affect the cost of construction? Some, certainly, but not all that much in the scheme of things.

If the cost of New York City construction is seven times that in Singapore, it is for reasons other than the cost of capital. Excess regulation, stronger property rights, conflicting policies (using more expensive union labor over non-union), stronger environmental protections, stronger court systems for protecting individual rights, etc. seem more likely candidates for the excess of costs in the US over those in other countries.

Bringing construction costs in the US down would likely entail weakening property rights, environmental protections, reducing policies that favor unions, weakening the powers of courts to delay projects, and basically weakening citizen rights in the face of State powers.

I am sympathetic to the argument that our protections are too strong or that the cost consequences of those protections are counterproductive to long term well-being. Fair enough. But that is the argument we need to have. Just how much protection is too much? Let's be open and honest that that is the debate.

There is no magic pot of gold that gives us the free lunch we want. Someone has to pay in some fashion.

Public infrastructure is a systems question. Costs and benefits, determinism and risk, payers and beneficiaries, subsidies and lost alternatives. You have to be asking systems questions not making your argument on apples and oranges comparisons and ignoring all the trade-offs and alternatives.

Saving, cost control, and infrastructure seems to be exactly what is wrong. Glib observations deriving from misperceptions based on shallow data leading to predetermined policy recommendations.