# If the government builds a bridge, how do we know it's the best possible usage of real resources (i.e. steel, labor, etc) at the time?

I would assume the "goal" of any economic activity is to make the largest amount of people the "happiest" -- i.e. Pareto Efficiency or sum-total 'utility.'

How do we know if this occurs? Either before, during, or after construction?

• At the moment we don't. To know that we'd need more context on the government in question and the country, the market etc etc etc. As it stands your question needs more details – Maarten Punt Oct 6 '20 at 17:34
• Woah, woah, woah! Are you saying the government may not be omniscient and benevolent? – Giskard Oct 6 '20 at 17:43
• Also, note that "Pareto-efficiency" and maximized sum total utility are not the same thing, see "monotonic transformation". – Giskard Oct 6 '20 at 17:44
• @Giskard that made me laugh out loud. Also, fair enough if they're not the same thing. But I was just using terms loosely. I don't think it changes the thrust of my question. I'm just trying to figure out how we know if resources are allocated efficiently or not. – Davis Clute Oct 6 '20 at 17:45
• @DavisClute that’s why I said that it can be considered public good/ semi-public good or more specifically quasi-public good depending on situation. If in some area congestion is low bridge will be non-rivalrous making it already quasi public, and if there is not large enough customer base to support toll it becomes non-excludable. In real life a sort of litmus test for public project is whether their return would be higher than publicly decided required rate of return - that is not the same as being Pareto optimal or completely efficient but it’s common method applied to public infrastructure – 1muflon1 Oct 6 '20 at 18:09

A short answer is that in case of provision of public goods in real-life we can never be completely sure if they are provided in optimal quantity due sheer difficulty of quantifying all costs and benefits, measurement problems when it comes to underlaying utility, due to uncertainties involved and many other factors.

However, this does not mean that no evaluation of public projects can be done. Most developed countries will have some dedicated analytic unit. In the Netherlands where I live that would be Centraal Planbureau [Bureau for Economic Policy Analysis] (CPB) in the US it is Congressional Budget Office (CBO).

These units routinely engage in cost-benefit analyses of public projects and then make recommendations based on whether the projects net present value is positive. The net present value would be given as:

$$NPV = B_0 - C_0 + \frac{B_1-C_1}{1+r} + \frac{B_2-C_2}{(1+r)^2} + ... + \frac{B_T-C_T}{(1+r)^T}$$

Where $$B$$ are benefits of the project, $$C$$ are costs and $$r$$ is discount rate or you can interpret it as a required rate of return.

The benefits and cost try to capture actually underlaying changes in utility. They will be usually estimated using hedonic pricing or various discrete choice models that can tell us something about peoples preferences and utilities based on observing their actions. For example, from the way how housing prices vary in cities with high congestion we can estimate how much people value time lost in traffic. The required discount rate is usually set to be equal to interest rate on government/municipality bonds or rate that government could obtain if it had invested the money into wealth fund. Of course, the way how these parameters are estimated or assumed is always a point of contention as this can't be done with infinite precision and complete certainty (this holds even in cases when we do ex-post cost-benefit analysis as most benefits and costs are virtually never directly observable and have to be estimated).

The above is, of course, just a short overview as providing an exhaustive review of how public cost benefit analysis is done is unfortunately beyond scope of SE. A good and highly cited example of public cost-benefit analysis is: Leape, J. (2006). The London congestion charge. Journal of Economic Perspectives, 20(4), 157-176.

Furthermore, just because project passes cost-benefit analysis it does not necessary means that the outcome is pareto efficient or even that it maximizes utility. However, it is reasonably good guard against completely frivolous spending. This being said, advice given is not advice taken and such analyses are often ignored by politicians, as pointed out by Giskard public officials are not necessarily benevolent.

• Thanks for the answer, this looks good. Out of curiosity, do you know of any govt projects that were closed because of insufficient benefits vis-a-vis costs? It seems like we could only guarantee a project/building is "efficient," insofar as it has to cease to exist if it's not... – Davis Clute Oct 7 '20 at 1:48
• @DavisClute Yes I was actually working at the above mentioned CPB as RA. I was working in sector 1 public finance and partially sector 2 macro economics on forecasting inflation, but one of my classmates from public econ MSc was working in infrastructure unit and I know that in the case of project he was working on which was cost-benefit analysis for a bypass route in some city (I think it was Den Haag but cant remember) they decided not to construct the bypass - but cba's are usually done before municipality goes public with the project so usually cancelations don't get press-releases – 1muflon1 Oct 7 '20 at 8:33
• @DavisClute I talked to my friend and he said that this CPB report in the box on pp 37 called: The influence of the benefit–cost balance on decision-making provides short review of literature on how much government listens to the outcomes of the CBA's and that the literature cited therein has some concrete examples so if you are interested have at it: cpb.nl/sites/default/files/publicaties/download/… – 1muflon1 Oct 7 '20 at 8:47
• Whether or not the unit planning a capital investment project is the government or the private sector (markets) note very carefully that the cost benefit equation is subject to several modes of uncertainty associated with (1) the subjective utility of the community now and in the future; (2) projected cash flows; and (3) future changes in the discount rate. So one is comparing two or more alternatives in the present without knowledge of what will happen in the future. If a public project "fails" we have another political argument over government waste. Private failures resolved via bankruptcy. – SystemTheory Oct 12 '20 at 14:22
• @SystemTheory I agree hence the reason why the first paragraph of my answer says what it says. – 1muflon1 Oct 12 '20 at 14:32

Under the original question there are comments which imply a model or reasoning process stated as follows: markets validate investment decisions (e.g., bridge building) by rewarding the investors with profits. If this logic is used as the "yardstick" for measuring economic efficiency then it is clear that future spending decisions either validate or do not validate the actual cost outlays of capital spending projects.

The Austrian Business Cycle Theory (ABCT) describes malinvestment as the cause of economic business cycles and argues that the central bank is the proximate cause of malinvestment. This 12 page paper outlines the argument:

https://cdn.mises.org/qjae9_4_4.pdf

Hyman Minsky argues in the short paper below that uncertainty and financial relations are a feedback from the past into present spending decisions under the institutions of a particular capitalist economy:

https://digitalcommons.bard.edu/cgi/viewcontent.cgi?article=1023&context=hm_archive

These types of economic theory seem to argue that we do not know which projects will be validated before, during, or after construction, but if cash flows validate past debt structures and pay profits then in retrospect some economic observers argue that those investments were the more efficient.

• I appreciate the thoughtful answer. In a certain sense though -- perhaps profits can be the only yardstick? If you accept the premise that prices are an epistemological "interface" which express the supply/demand dynamics of different resources ...then by definition market-based exchange ratios (i.e. "prices") would be the only conceivable yardstick. However, don't jump on me too much for that logic lol -- I just wrote that off the top of my head. Simply trying to poke + prod your argument. – Davis Clute Oct 13 '20 at 4:13
• As an aside, I like your name -- I am a big fan of Systems Theory/Thinking as well. But I won't be on Econ Stack Exchange in the future for a variety of reasons. I'd love to debate more/hear more of your thoughts. You obviously know the topic well. I'm obviously using my real name, so you should reach out on LinkedIn / email. – Davis Clute Oct 13 '20 at 4:16
• Maybe I will reach out. The problem with the model where profits either validate or do not validate past investment projects has been investigated by (defunct) economist Hyman Minsky. If we consider one investment project in isolation then profits are the "democratic vote" which validates this use of assets (it is efficient because customers pay for service). But in the whole economy the debt finance of new investments is generating cash flow to validate prior investments. If debt and money markets crash then many investments are no longer validated by present cash flows. – SystemTheory Oct 13 '20 at 17:45
• Umm, I see what you're/Minsky is saying -- but that seems to be a very theoretical/edge-case objection. He's just saying: If the economy completely crashes, then it (arbitrarily) invalidates past investments? Two philosophical objections: 1) I think credit markets would dislocate BECAUSE of malinvestment. It's not like credit markets just happen to "freeze" one day arbitrarily... 2) I would think the bigger barometer of success isn't cash flow vis-a-vis debt financing. I would think the bigger determinant is increased cash flow vis-a-vis an increase in productivity. – Davis Clute Oct 13 '20 at 18:16
• I.e. I don't think an increase in debt (created by whom, btw?) is necessary to justify an earlier investment. I think of money as a veil over the real resources in the economy. I.e. it is neutral in the medium run. So I don't see an increase (or decrease) in M0, M1, etc as being terribly important to the question of: "Why does the pricing system allocate resources efficiently, and what does that even mean practically speaking?" – Davis Clute Oct 13 '20 at 18:20