Scott Sumner  

Saving, cost control, and infrastructure

PRINT
Tabarrok on the Great Depressi... Cruise Ships and Private Plots...

There's been a lot of recent talk about increasing federal spending on infrastructure. In my view this misses the point. There are two effective ways to get more infrastructure; cost control and increased saving. Here's Matt Yglesias:

Mass transit construction costs in the United States appear to be far higher than what European countries pay for comparable projects.

The Second Avenue Subway in New York City, for example, is being built at a cost of nearly $1.7 billion per kilometer while new subway lines are being built in Paris, Copenhagen, and Berlin for about $250 million per kilometer.

Screen Shot 2017-05-27 at 12.41.14 PM.png
Unlike many pundits on the left, Yglesias draws the correct conclusion:

One popular school of thought holds that transit advocates essentially ought to circle the wagons and deny that there's a problem here. The Second Avenue Subway may be ungodly expensive, but it is a really valuable and useful project. The United States wastes plenty of money on highways, too, and there always seems to be enough money for another cruise missile or stealth bomber, so why should we nickel-and-dime transit projects?

I've come to think that this is fundamentally misguided. The reality is that if you want to build a lot of transit projects, it's really helpful to be able to build them at an affordable cost. Not only does that stretch a given pile of dollars further, but precisely because it lets you stretch further it means that your project touches more people's lives and can garner a broader coalition of political support. Paris's ability to build subways cheaply doesn't mean Paris has become stingy on its transit projects -- the ability to get a lot of bang for the buck is one reason they can do the enormous $25 billion Grand Paris Express expansion project.


Sir Roger Douglas and Robert MacCulloch have a very nice paper with a great title:

Welfare: Savings not Taxation

It describes how economies can be made more efficient by switching from mandatory taxation to mandatory savings accounts:

To our knowledge, showing how both a tax and welfare reform can be jointly designed to enable this transition to occur has not been done before. Our policy reform creates institutions that have features in common with Singaporean ones, especially for health-care. However there are also key differences. We present a new unified approach to the funding of health, retirement and risk-cover (for events like unemployment) through the establishment of a set of compulsory savings accounts. A case study of New Zealand is used as an illustration. The fiscal impact of our proposed reform on the government's current and future budgets is reported, as well as its effect on low, middle and high income individuals.
Of course higher savings rates will also lead to more investment, including infrastructure. A recent post over at Free Exchange expresses worry that there is too much saving:
There is, now, too much capital. Ben Bernanke, then chair of the Federal Reserve, first described a global savings glut in 2005. It is a well-studied and well-known phenomenon; here is Lawrence Summers of Harvard summing it up last year. And, as predicted in both Samuelson and Mankiw, there are not enough investments to accommodate the world's savings, and interest rates have dropped. Yields for both high-quality corporate bonds and long-term American federal debt are now lower than at any time since 1973 (see chart).

Members of Congress, however, seem to remember only that saving is good. As Washington considers rewriting the tax code, neither party is pushing to raise taxes on dividends, capital gains or estates, all of which could encourage more consumption now and drive down the total level of saving. Republicans have even signalled that they are considering a value-added tax, which discourages consumption.


This confuses several issues. The term 'savings glut' has no normative implications. It is simply a claim that long-term interest rates fell because of higher savings rates in Asia. Of course it's possible to have too much of anything, but as of now public policy is strongly biased against saving---especially the tax code, but also programs like college assistance and Medicaid. Congress should be trying to re-write the tax code in such a way as to promote more saving and less current consumption.

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.

PS. When I say "pro-saving" tax policies, I actually mean eliminating anti-saving policies, and switching to a neutral tax regime.


Comments and Sharing






COMMENTS (13 to date)
Rob L writes:

If you look at the American savings situation, there is no actual glut. The article at Free Exchange seems to be confusing the very liquid short term corporate and government debt markets with indicating too much savings.

It is even worse if you look at net domestic savings flows (ie after depreciation of the capital stock). The key savings flows supporting investment in America are nearly entirely from the trade deficit (capital account surplus).

Greenspan has been quite good to read / listen to on this topic.

As an Australian, I agree that you should look at compulsory savings accounts. Our 1:1 capital backed superannuation industry now outstrips the total value of equities on our publicly traded stock exchange.

However, there are still many regulatory restrictions and tax incentives that limit construction, development and investment that results in a large portion of savings just being poured into the secondary housing markets, not even new housing construction. This then effectively operates as home equity financed consumption in the short term.

Hopefully this can be addressed with reform, but that is a whole different discussion.

Scott Sumner writes:

Rob, Yes, loosening zoning laws would be the single best thing Australia could do to boost living standards. Having said that, Australia's overall public policy regime is one of the best in the world---I agree with you about the compulsory savings accounts.

Mark Brophy writes:

Ignoring high costs is a favorite political tactic among Democrats and Republicans not only in mass transit but in health care. It makes no sense to discuss ObamaCare and health insurance as long as the underlying corruption in the medical industry is ignored.

If health savings accounts were mandatory, shouldn't it be legal to buy health insurance with a $20,000 deductible as long as you have that amount in your account?

Andrew_FL writes:
When I say "pro-saving" tax policies, I actually mean eliminating anti-saving policies, and switching to a neutral tax regime.

Don't you favor an extremely progressive consumption tax? You have a very strange idea of what a neutral tax regime would be.

Scott Sumner writes:

Mark, Yes, I agree.

Andrew, I meant neutral in regard to saving/consumption. No tax regime is neutral with respect to everything.

Thaomas writes:

I very much agree with your recommendation, but would qualify it by saying that the shift should be done is such a way as to improve, not harm living standards of poorer citizens. We should 1) shift SS and Medicare funding from a wage tax to to a consumption tax. 2) Business taxes should shift to personal consumption taxes, which should be progressive.

John Thacker writes:
No tax regime is neutral with respect to everything.

Indeed. One of the strange (but unsurprising) things in the current corporate tax debate is that the House Republican leadership is proposing a neutral (with respect to foreign and domestic origin) tax on domestic consumption (which would be progressive due to the wage deduction.) It would not be neutral with respect to domestic versus foreign consumption, naturally.

In this, the DBCFT is in contrast to the current corporate tax regime, which is neutral with respect to domestic versus foreign consumption, but taxes domestic production more than foreign production. (It also taxes domestic-owned foreign production more once the money is repatriated.)

Of course no tax regime can be truly neutral with respect to everything, since we're never going to tax goods and services that are foreign-owned, foreign-produced, and foreign-consumed all at once.

Strangely, some people who claim to favor progressive consumption taxes oppose switching the corporate tax to a tax on all domestic consumption, neutral to where it is produced. This is partially because some people trying to sell the tax as well as those opposing it are arguing emphasizing that it will effectively lower taxes on exports (since those aren't consumed here) and raise taxes on imports (since those are consumed here.)

Andrew_FL writes:
Andrew, I meant neutral in regard to saving/consumption. No tax regime is neutral with respect to everything.

This seems to miss the point entirely. How is a consumption tax neutral with respect to the saving/consumption trade off?

John Thacker writes:

The gap between US infrastructure spending and other countries is really quite small compared to the gap in costs, and thus output. Indeed, given such differences in costs, willingness to spend less on infrastructure compared to other things is a fairly rational result.

People often look at the outputs in other countries, including rich ones, which have as much cost savings compared to the US as poor ones on infrastructure, to argue that obvious we should be building more. We should, but because we should be getting our costs closing to theirs.

Unfortunately, it's difficult to figure out exactly what makes costs high. It does seem like common law countries, with their torts, do make costs higher, but the USA is significantly more expensive than our English-speaking common law peers. There must be multiple factors, all coming out poorly. E.g., NEPA might be part of the problem, and Davis-Bacon might be as well, but nothing is the sole cause.

Khalil writes:

Scott -- if you have time, interested in your thoughts on the relationship between the relative weakness of US government agencies in terms of enforcement powers (compared with other Western countries) and how this impacts on costs for infrastructure projects.

Thinking specifically of the adversarial litigation that goes on between the US bureaucracy and private parties to resolve disputes/enforcement problems that in other countries would be solved with consultative processes.

In other words, the quite unique features of the US political and legal systems add to the procedural complexity, uncertainty and high transaction costs add significantly to public projects. Compare this with Singapore, a place in which protest against anything the Singaporean government decides is very limited.

Scott Sumner writes:

Thaomas, Keep in mind that a wage tax is a consumption tax.

John, Good points.

Andrew, A consumption tax taxes current and future consumption equally. Saving is basically future consumption.

Khalil, That may be important, but I don't think I'm qualified to offer an opinion.

BD writes:

Scott, are you aware of any literature out there on the "why/how" US infrastructure costs are so much higher than other rich countries?

Hazel Meade writes:

A large part of the problem is that infrastructure projects are widely used as jobs projects and stimulus spending by politicians. If that's how you think of infrastructure - not in terms of the utility of the infrastructure, but the stimulative effect of the spending, then a big project that costs more and employs more people is better, not worse.

When was the last time you heard a politican talk about the need to build infrastructure because of the benefit of the actual infrastructure, not the jobs it would create?

Comments for this entry have been closed
Return to top