Tool/Decision Theory/No. 0218

Cost-Benefit Analysis

Cost-benefit analysis is a method in welfare economics that compares expected benefits and costs in money against an alternative. Also called benefit-cost analysis, it shows uncertain assumptions, unpriced effects and who gains or pays. Results depend on the values and assumptions used.

Also called Benefit-Cost Analysis

a tool: pick it up

Evidence: Well established
Read: 6 min
Links: 18 connections
Useful when: Deciding under uncertainty · Fairness and ethics · Money and investing · Running projects · Complex systems and policy

01You've seen this when…

in life
An insulation contractor quotes $8,000. You compare the next decade of heating savings with the bill, the installation disruption and warmer bedrooms.
at work
A team proposes software that saves analysts two hours a week. Finance adds the license fee; operations adds training, migration and the time someone will spend maintaining it.
out in the world
A council weighs a floodwall that protects downtown and increases flood risk downstream. The headline benefit exceeds the construction cost, while downstream residents carry much of the damage.

02The idea

A proposal uses resources and changes people’s lives. Cost-benefit analysis asks whether its expected gains justify what it consumes or harms, compared with a specified alternative.

The comparison starts with a baseline: what happens if this proposal goes ahead, and what happens otherwise? Insulation saves money relative to heating the same house without it. A floodwall protects buildings relative to the floods expected without the wall. Everything counted should be a change between those futures.

Benefits and costs are usually expressed in money so they can be compared. Some already have prices, such as electricity or construction materials. Others need estimated values: travel time, cleaner air, quieter streets or reduced injury risk. Consequences that resist credible pricing remain visible alongside the total.

A public analysis takes a wider view than the project’s bank account. It includes effects on people outside the organization, including externalities, and the opportunity cost of resources that could serve another purpose.

The resulting balance is conditional on the estimates and boundaries chosen. A positive total says the monetized benefits exceed the monetized costs. It leaves questions about fairness, rights and unpriced consequences to resolve.

03How to use it

Define the alternatives. Specify the proposal, the baseline and at least one feasible alternative. Include the time horizon. A baseline can involve repairs, worsening congestion or changing demand; spell out what you expect to happen.
Set the boundary. Decide whose gains and losses count. A business might examine its own returns; a public project should examine affected households, businesses and communities. Payments between groups inside the boundary can be transfers. Track both sides and count resource costs separately.
List the changes. Include setup, operation, maintenance, disruption and effects on others. Count resources at their next-best use. Keep unrecoverable past spending out of the forward comparison: this helps resist the sunk cost fallacy. Check for double counting, such as adding both a journey’s time savings and a property-price increase caused by those same savings.
Value what you can defend. Use relevant market prices and established appraisal values where suitable. Explain estimates for time, health or environmental benefits. For uncertain outcomes, calculate probability-weighted expected values when probabilities are defensible. Keep difficult-to-price effects in a separate, explicit record.
Put different years on the same footing. Discount future amounts into today’s equivalent value, then subtract discounted costs from discounted benefits. This gives net present value. State the discount rate and keep inflation treatment consistent: amounts in today’s prices need a rate that excludes inflation.
Test the result and show its distribution. Vary the assumptions most likely to change the decision. Sensitivity analysis can reveal the usage level, construction cost or benefit value at which the result changes sign. Show who gains, who pays and which effects remain unpriced. Present this record alongside the total.

04A worked example

In this invented example, a council considers a $500,000 footbridge. Upkeep costs $10,000 a year. The appraisal covers ten years and assumes no remaining asset value afterward. All amounts are in today’s dollars, with an illustrative annual discount rate of 3%.

The forecast is 40,000 crossings a year, each saving six minutes. That adds up to 4,000 hours annually. At an assumed value of $20 an hour, annual time savings equal $80,000.

What it looks like A large construction bill in exchange for a slightly shorter walk. Supporters emphasize convenience; opponents emphasize the upfront expense.

What’s actually going on The annual monetized benefit after upkeep is $70,000. Discounted over ten years, that is about $597,000. Subtract the construction cost and net present value is roughly $97,000.

Usage is decisive. If annual time savings reach only $60,000, net present value falls to about negative $73,000. The council also records step-free access to additional destinations and riverbank habitat disturbance separately, taking care to avoid counting access-related time savings twice.

What made it work The calculation exposed a forecast the decision depended on. Before committing, the council can count existing trips and investigate likely usage. It can also compare the bridge with improving the existing route and assess the habitat effects. The positive headline total becomes a testable case with visible gaps.

05When to reach for it

Costs and gains arrive at different times. Home improvements, infrastructure and preventive maintenance need a longer view than the next budget.
The sticker price dominates discussion. Training, maintenance, time savings and disruption deserve space beside it.
Several options pursue the same purpose. Comparing alternatives can uncover a cheaper route to most of the benefits.
Other people bear consequences. Public projects and regulation need an explicit account of effects beyond the decision-maker.

06When it misleads

Precise arithmetic hides weak inputs. A spreadsheet can produce a result to the nearest dollar from a demand forecast that varies twofold. Report a range and identify the assumptions that drive it.
Willingness to pay reflects purchasing power. A wealthy person’s willingness to pay can exceed a poorer person’s for the same improvement. Monetary totals therefore need a separate account of distribution and any equity weighting used.
The boundary leaves people out. Exporting pollution or displacement beyond the study area can make a project look attractive. Follow the effects far enough to capture consequential gains and harms.
A total settles questions it cannot answer. Legal duties, rights and ecological limits can constrain the options. A positive monetized balance also leaves any substantial unpriced damage unresolved.

Cost-effectiveness analysis compares cost per unit of a shared outcome, such as cost per additional vaccination. Cost-benefit analysis values benefits and costs in money. When several objectives resist a common valuation, multi-criteria decision analysis can make the trade-offs explicit.

07Roots

In 1844, French engineer Jules Dupuit was trying to work out what public works were worth. Bridges and roads created useful journeys, yet the toll receipts captured only part of their value. A free bridge could collect nothing while still saving travelers hours.

Dupuit’s solution was to examine how much people would be willing to pay to use a facility. Different travelers valued a crossing differently. Adding those values gave a way to estimate the benefit of a bridge and examine how tolls affected its use. His analysis helped lay the groundwork for what economists later called consumer surplus.

The method acquired institutional force as governments sought rules for choosing expensive projects. The U.S. Flood Control Act of 1936 made federal participation conditional on benefits, however distributed, exceeding estimated costs. Later public-economics and policy-analysis traditions extended appraisal across transport, health, environmental regulation and other investments. Contemporary guidance formalizes discounting, uncertainty and distribution, while preserving Dupuit’s central problem: estimating a public work’s value beyond the revenue it collects.

08How solid is this?

A long-established appraisal framework grounded in welfare economics and widely used in public investment and regulation. Results depend on the baseline, valuations, discount rate and treatment of distribution; monetized rankings carry no general guarantee of identifying the best policy.

10Origin and sources

Jules Dupuit developed an early economic analysis of public works in 1844. Public-economics and government policy-analysis traditions subsequently formalized the technique.

[1]Dupuit, J. (1844). De la mesure de l'utilité des travaux publics. Annales des ponts et chaussées, 2e série, 8, 332–375.
[2]Boardman, A. E., Greenberg, D. H., Vining, A. R., & Weimer, D. L. (2018). Cost-Benefit Analysis: Concepts and Practice (5th ed.). Cambridge University Press.
[3]HM Treasury. (2022). The Green Book: Central Government Guidance on Appraisal and Evaluation.
[4]Arrow, K. J., Cropper, M. L., Eads, G. C., Hahn, R. W., Lave, L. B., Noll, R. G., Portney, P. R., Russell, M., Schmalensee, R., Smith, V. K., & Stavins, R. N. (1996). Is There a Role for Benefit-Cost Analysis in Environmental, Health, and Safety Regulation? Science, 272(5259), 221–222.

Suggest an edit· Updated 2026-10-02