A key question in corporate finance is, how should companies decide when to invest the funds at their disposal? The best answer provided by financial economics is to use the net present value (NPV) rule. Here I want to show that NPV is a specific case of the wider principle of cost-benefit analysis (CBA), which applies to any investment decision, including those in the public sector.
*
Summary of the NPV rule
One of the first things you learn in a course on corporate finance is that investment decisions by companies should be taken according to the net present value (NPV) rule. Based on the idea that rational agents like returns but don’t like risk, the rule contains four parts:
i) value is measured by monetary amounts – unless the benefits can be turned into money eventually, they are not of use to rational agents, as imagined by economics;
ii) the resource implications of any investment decisions can be divided into costs – those resources you need to invest to make the investment happen – and benefits – the returns that the investment is expected to generate;
iii) those costs and benefits typically are forecast to occur at different times in the future and we need to account for this by appropriate adjustments, called discounting – in particular, the further away in time that benefits or costs occur, the less they count in today’s money, so we discount them more; this is described as turning all monetary amounts into equivalent present values, that is values in today’s money;
iv) if the sum of the discounted present values (i.e. the net present value) is greater than zero, the project has a positive value and is worth investing in; if the NPV is negative, do not proceed.
At a later stage in learning corporate finance, we learn about the method of calculating the appropriate discount rate, which adds a further important consideration, namely the riskiness of the forecast benefits. If they are more risky, we should discount them at a higher rate. The main way of doing this is to use the capital asset pricing model.
NPV is a special case of cost-benefit analysis
Not everybody who takes a course in corporate finance plans to work in a company, or in the private sector more generally. So they might feel that the NPV framework is of little interest. But I want to show how net present value is really just a special case, a more narrow case, of a wider way of making decisions, which is called social cost benefit analysis. Hopefully I will convince you that this is a sensible way to think about decisions which have cost and benefits.
Let’s start with a public sector decision, and to make it specific, let’s imagine we’re in the UK and the government is thinking of building a bypass road around a small, old town. In Britain many roads are very old and they go through the centres of towns, because the towns usually grew up on the site of roads, especially road junctions. What that means is that as the traffic has built up over the years, those towns became very congested, noisy, polluted and dangerous. It’s understandable that the people who live there would rather have a new road built that takes the traffic around the town, so that those drivers who simply need to pass through the town can be safely diverted. These new roads that divert the traffic round the town are known in the UK as bypasses.
So you’re the government, you’re thinking about whether to build a new bypass round an old town that is being choked by traffic. Now a starting point would be to say, well, let’s consider all the benefits of building such a bypass and all the costs.
If this was simply a private sector project, the benefits would be very simple. It would be the revenue that would be generated by the road, which can be done if the road is a toll road, for example.
But if you’re a government, you may consider, indeed you should consider, a much wider range of benefits and costs. For example, if the road is built, it will reduce not just traffic congestion in the town, it will also reduce air pollution, assuming that the vehicles are still at least partly using internal combustion engines, and that will bring benefits to the people who live in the town. It will also reduce noise and will make the whole environment more pleasant.
It may also be safer. New roads tend to be built to higher standards than old roads, so there may be a lower level of traffic accidents.
Equally, a government should consider a wider range of costs. There’s the direct cost of actually constructing the road. There’s also the indirect cost that during the period of construction there will be disruption to the traffic, there will be delays and this will cause trouble and nuisance to people who use the road.
But there are other costs as well. If the road goes through, for example, an area of attractive natural landscape, it will reduce the value to people who go there as tourists, or even just passers-by. Perhaps there are rare or precious species of animals and birds that risk being displaced by the new road.
It may be that the construction requires cutting down trees. Trees are generally seen as a good thing, partly because they absorb carbon dioxide but also because they have health benefits for people, so cutting down trees has a social cost. There may be other costs and benefits as well. The point is that there’s a wide range of things that a government ought to take into account that go far beyond the pure or narrow private costs and benefits.
Let’s assume that we’ve itemized all of these costs and benefits and we have written them down on a piece of paper. On one side we have the benefits, and on the other side we have the costs. Maybe there are more benefits than costs, but that’s not a very good basis on which to decide whether to go ahead.
Firstly, we have to do we need to try to quantify all of these things. Now, some of this is fairly easy. The direct costs can be estimated by asking construction companies for tenders, using consultants and so forth. The benefits of traffic congestion can be reasonably well estimated by working out how many hours of sitting in traffic would be saved and then applying an estimate of the value of people’s time based for example on average wages.
Valuing things like the value of fewer accidents is more tricky. But there are ways of doing this. In the UK and in other countries, there are standard metrics that are used to assess, for example, the value of a life saved (*). It’s currently £1.8m, or $2.3m. That number is an estimate based on how we as a society appear to value safety, how much we’re willing to pay to reduce risk or how much extra pay is needed to compensate people for taking on riskier jobs.
To some people it’s unethical to reduce a person’s life to a financial amount. But the problem is, if we don’t put some number on it, we’re in effect valuing human life at zero, and we surely ought to take it into account somehow.
Similarly, we’re trying to value the damage done to the environment from cutting down trees, from noise and all the other unavoidable damage that’s done by building a road through an area that’s currently open countryside.
This is not straightforward, but again we must put something in, otherwise we’re valuing those costs at zero. There are ways to estimate or infer how people value open countryside. We can look at what people pay when they go on particular types of holiday. We can use the ticket prices from various country sites which charge admission. None of this is exact or perfect. But the important thing is to have some sort of estimate.
Converting all costs and benefits to present values: discounting
Now let’s assume we’ve done all of this. We’ve now turned our piece of paper with simply a list of items into a series of numbers and we could say, well, let’s add up the value of all the costs, let’s add up the value of all the benefits, and let’s see whether the number is net positive or negative.
That’s a lot better than the starting point where we simply counted the number of costs and benefits. But even allowing for the fact that some of the numbers are estimates and people may reasonably disagree about them, we’ve left out one further important consideration: timing.
Some of the costs, such as the construction costs, will be incurred right away. The benefits will mostly be delayed until the construction is finished. Let’s assume it takes three years, that’s realistic in the UK these days. The benefits will come after three years. Now, does this matter? Well, this is where we get into the question of the time value of costs and benefits, or equivalently, whether and how, by how much we should discount the future.
This is a very important question, both ethically and also practically.
The ethical question is whether we should ever discount the future, because we are in effect assuming that the value of benefits to people who live in the future are worth less than people today, and some people think that’s simply wrong.
Take action on climate change. Most of the costs of this are incurred in the short term, but the benefits will mainly be in the future. Does this matter? Should we discount the benefits because they will come to our children or even grandchildren? What if we expect them to be much richer than us, because of continued economic growth? These are serious questions that people can reasonably disagree on.
Having said that, the way most people behave in their lives is that they do discount the future. They do act as if a benefit now is worth more than a benefit in the future, partly because they may not be confident that the future benefit will actually happen. And partly because they may not be around in the future, there is always the possibility that a person will actually not survive and therefore it is reasonable, one could argue, to discount the future.
The rate at which we discount the future, the discount rate, is something that we’ll consider later on. It’s a very important topic and in the case of private corporate finance, we have a fairly clear way of doing this. It’s a bit more complicated in the public sector because there’s a wider range of considerations, but that’s not something we need to worry about now.
So what we’ve got here is a framework which is called Social Cost Benefit analysis for analysing when an investment should happen, in a rational, disciplined and hopefully open process so that people can challenge the assumptions and if need be, there can be a debate about it in an open society. Particularly in a democracy, all of these things should be transparent and therefore both the process and the assumptions are open to scrutiny, and this should avoid bad decisions being made.
Social cost benefit analysis is just a more inclusive version of what we call net present value. Net present value, as used in corporate decision making, is simply taking account of all the relevant costs and benefits, turning them into financial flows and then discounting them at an appropriate rate of discount. Then we see if the net value of those discounted financial costs and benefits is positive or negative.
The argument is that if we’ve done everything right, a project, an investment opportunity with a positive net present value is one that a company should go ahead with.
Private versus social costs and benefits: the importance of externalities
The key difference between net present value in this narrow sense and social cost benefit analysis is that the private corporation will typically only consider the private costs and benefits that are relevant to the company and to its investors. This is often just called maximizing shareholder value.
The problem is that it’s obvious that some of the costs and benefits that arise from a corporate investment decision may not directly influence the shareholders, but do influence other people, for example, the pollution that may result from an investment. These additional costs and benefits are called externalities. Most people may think that externalities should be included in investment decisions, but the company will not take them into account unless it’s forced to, through some form of regulation or tax.
Sometimes the corporate managers may decide that it is in the interest of shareholders to take, say, pollution into account because it represents a potential future liability, either in damaging the company’s reputation, or perhaps because of the risk of being sued by the people affected by the pollution.
But this is a more tricky approach for the managers to take. Some managers, who are paid to look out for the shareholders’ interests, will decide to simply ignore the pollution cost, which means that some projects will go ahead, which from a social point of view should not.
That’s not a criticism of the net present value framework. It’s a problem that needs to be handled by ensuring that corporate decision making somehow takes into account those external costs and benefits which society as a whole thinks should matter. That usually requires government action (**).
Beware of ignoring other ethical principles in decision making
A final thought. I hope by now I have persuaded you that social cost benefit analysis and net present value are useful ways for thinking about decision making, which indeed they are. But both depend on a fundamental ethical assumption, which is that we can evaluate decisions based purely on their outcomes, or consequentialism. A particular version of this, which is in effect what we use in most economics, is utilitarianism.
Consequentialism may seem a reasonable approach. But there are cases where our ethical instincts suggest this is not, or should not be, the whole story. Some actions may be considered simply wrong in principle, regardless of the consequences. A way to think about these questions is the thought experiment known as the trolley problem.
I think for the most part that utilitarianism is fine for many practical company and government decisions, but the really tricky cases usually involve matters of principle that appear to over-ride the consequentialist approach.
***
(*) The UK government refers to the value of a prevented fatality (VPF): “The Value of a Prevented Fatality (VPF) measures the social value of changes in risk to life. It is used to value small changes in fatality risks, where levels of human safety vary between options. This is not the value of a life, it is the value of a small change in the risk or probability of losing a statistical life. Not to value this in appraisal would effectively value human safety at zero.” The Green Book (2022) For a wider analysis of the international treatment of this topic see the October 2025 OECD report Mortality Risk Evaluation in Policy Assessment
(**) There are cases where private negotiation between the company causing the pollution and the people suffering from it can resolve the problem, perhaps through a legal process. But where there are many such people, the costs of action and coordination can become too high. If property rights are not well defined, legal processes can be ineffective, or just very expensive. In these cases some form of political solution may be the only way forward.
Leave a Reply