Many people assume that, when we invest in infrastructure, "the government" gets back - by whatever means and at whatever level - cumulatively more money than it puts into a project. There are others who know that this is not remotely true, but believe that the money invested in infrastructure is going to a good cause that pays in other ways (job creation, quality of life, etc...). There are a growing number of people that believe that infrastructure spending, in its current form, is a financial sink hole fueled by political patronage and special interest agendas.

Regardless of which belief is correct, the cost benefit analysis is the mechanism whereby we justify the spending. Today we are going to have a theoretical discussion on how a cost benefit analysis should be done.

For a business in the private sector - let's assume it is a heartless corporation with profit as the sole motive for its existence - a cost/benefit analysis is simple. How much does an undertaking cost and what financial return does that yield? When the financial return exceeds the cost, this is an undertaking worth considering. Of course, a business will look at the amount of risk and volatility as well, but the basic return-on-investment calculation is the first step.

Private sector ROI: financial benefit > financial cost

The analysis gets a little more complicated for a non-profit endeavor like a charity hospital or a private school. For such an endeavor there is also a social motive beyond pure profit that comes into play. We'll call this "social value". While a charity hospital cannot ignore ROI - it must, after all, generate sufficient return to stay in operation - ROI is not always the overriding criteria for decisions.

For example, a charity hospital may increase ROI by having a cigarette vending machine in the emphysema ward, but this would seem rather counterproductive to the overall mission. Unless that cigarette vending machine provided the hospital with the revenue it needed to save the lives of premature babies born to impoverished mothers. Then maybe it is not such a black and white discussion.

In the end, though, a charity hospital would balance the relative costs and benefits associated with different actions and come up with an approach that allowed it to do the most good while still staying financially solvent.

Non-profit ROI: financial benefit + social benefit > financial cost + social cost

In concept, it is easy to see how governments would operate much like the non-profit charity hospital. When we collectively gather together for a common purpose for which we tax ourselves, the costs and the benefits are not simply financial. The questions this collective action raises are complex and largely unanswerable, at least in an absolute sense.

  • What is the best way to raise revenue?
  • Who should contribute to the costs and in what amount?
  • Should there be a means test where the better off pay more and the less fortunate pay less?
  • What is the best way to spend revenue?
  • How do we determine which projects have the greatest benefit?
  • Is there a minimum threshold of government service that everyone is entitled to?
  • How should all of this best be determined?

Certainly government can do bad while trying to do good. The classic utilitarian quandary of the "inhospitable hospital", where one patient is sacrificed and carved up so that their parts can be used to save many others, may be an action that would provide great benefit to a great many people. but it is done at a cost that we are not willing to pay. We collectively place some values - in that example it is the life and autonomy of an individual - above all others.

The constitution assumes that government (which, at its essence, is We the People), unless it is constrained, will overreach and that this act of overreaching, however good in its intention, will have impacts that are destructive. We can argue all day whether or not government has overreached and whether or not it is destructive. I don't think that gets us anywhere. What is critical to understanding how the government should conduct an ROI analysis is to understand that the social benefits and social costs are "constrained" by the rules we have established to govern ourselves.

As an example, the constitution gives everyone a right to legal representation. This is critical to our thinking that the state should not be able to deprive someone of their liberty without a fair process and a vigorous defense. So we have set a constraint and obligated ourselves to pay for the defense of the accused, if they cannot afford a defense. We have also continued to cut funding for public defenders, which in a practical sense constrains the vigor in which we defend the accused. These are difficult realities that go to the core of who we are.

From a practical standpoint, today we live under a number of constraints when it comes to funding infrastructure improvements.

Constraint #1: We have limited revenue for infrastructure spending. (Note: We acknowledge that one can argue that we can simply increase taxes or fees to add revenue, but the public has consistently shown that, where they support new revenue, it should go to priorities other than infrastructure spending.)

Constraint #2: We have massive, unfunded liabilities for maintaining existing infrastructure systems.

Constraint #3: We are highly indebted, both in the public sector and the private sector.

Constraint #4: We have huge obligations and priorities in other parts of our economy that will limit the amount that we can devote to infrastructure.

In the end, the government ROI analysis is very similar to the non-profit ROI analysis except that our social benefits and costs are constrained by the community rules and accepted practices we have established.

Government ROI: financial benefit + (constrained) social benefit > financial cost + (constrained) social cost

It is important to note here that financial benefit in the government analysis is actually tax revenue. It is not "jobs", which is a social benefit. I can hear the immediate argument that "jobs create government revenue", but in reality less than 2% of income tax revenue goes to infrastructure spending. In other words, "jobs" does not equate to revenue for paying back infrastructure investments. If our spending on infrastructure is actually designed to create jobs, then we should really be asking the question, "How do we create the most jobs at the least cost?" That is a completely different discussion that rarely finds its way into the public debate.

At this point I would like readers to briefly imagine a world where our collective government action did not take into account social costs and benefits but instead focused on strict financial ROI. In this imaginary world, when a road was built, it would need to generate enough tax revenue to ultimately pay for itself. When the government builds an overpass at $9.5 million, it would need to see tax revenue increase by $9.5 million, plus financing costs, to cover that expense.

In the real world, nothing close to this ever happens. For the most part we completely ignore an analysis of the real financial return. Our projects are largely done for social reasons, with tremendous spatial disconnect between the costs and the benefits.

Let me elaborate on that briefly. We pay gas tax and registration fees. These fees are collected centrally and then the money is distributed out for road projects. We don't look at a road project and say, "this project will create X amount of new gas tax revenue or registration fees". We just do the project and then lament that there is not enough money to do more. Everyone feels entitled to their project because everyone pays gas tax and registration. The only problem is that, collectively, we are not paying anything near the cumulative cost of our spending.

The establishment systems tend to look at every project exclusively for its social benefits. People can get to work more quickly. Public safety response is enhanced. More private land is not accessible for development. Etc, etc, etc... In a practical sense, our approach overlooks the financial return almost completely.

Government ROI: financial benefit + (constrained) social benefit > financial cost + (constrained) social cost

What this means is that, if we were honest with ourselves, we would be asking whether or not a given project produced enough social gain to justify the cost. This would open infrastructure projects up to comparisons with other social endeavors, such as education and medical research, and that is exactly what the general population is doing (to the bewilderment of the pro-infrastructure spending establishment).

Coming up, we will look at the analysis that Staples submitted for its TIGER II grant and show how the analysis for projects like this one are essentially rigged to convert social benefits to financial benefits, thus claiming tremendous financial gain when none actually exists.


Posts in this series: 

  • Part 1: Background on cost/benefit analyses
  • Part 2: Review of time savings benefit and distance benefit
  • Part 3: Review of safety, carbon reduction and O&M benefits
  • Part 4: Review of the project cost analysis 
  • Part 5: Finale

We've been told that the Strong Towns blog is like reading James Kunstler, only you can share it with your mom. You can leave a comment or join us for more Strong Towns content on Facebook and Twitter. If you are interested in having the Strong Towns message brought to your community, sign up for a Curbside Chat and we'll make plans to get together in a town near you.