Cost/Benefit Analysis Part 1 – The Government Line

I’ll be writing a series of posts on Cost/Benefit Analysis.

First the Government line: baseline excerpts from government hosted publications regarding cost/benefit analysis, and links back to the publication.

OMB Circular A094 – Guidelines and Discount Rates for Benefit-Cost Analysis of Federal Programs:

…A program is cost-effective if, on the basis of life cycle cost analysis of competing alternatives, it is determined to have the lowest costs expressed in present value terms for a given amount of benefits.

…Cost-effectiveness analysis can also be used to compare programs with identical costs but differing benefits. In this case, the decision criterion is the discounted present value of benefits. The alternative program with the largest benefits would normally be favored.

…Analyses should include comprehensive estimates of the expected benefits and costs to society based on established definitions and practices for program and policy evaluation. Social net benefits, and not the benefits and costs to the Federal Government, should be the basis for evaluating government programs or policies that have effects on private citizens or other levels of government.

…Calculation of net present value should be based on incremental benefits and costs. Sunk costs and realized benefits should be ignored.

…Some Federal activities provide a mix of both Federal cost savings and external social benefits. For example, Federal investments in information technology can produce Federal savings in the form of lower administrative costs and external social benefits in the form of faster claims processing.

…Because taxes generally distort relative prices, they impose a burden in excess of the revenues they raise. Recent studies of the U.S. tax system suggest a range of values for the marginal excess burden, of which a reasonable estimate is 25 cents per dollar of revenue…(Therefore,) costs in the form of public expenditures should be multiplied by a factor of 1.25.

Department of Health and Human Services, Administration for Children and Families and Health Care Finance Administration Feasibility, Alternatives, And Cost/Benefit Analysis Guide:
Benefit/Cost Ratio is calculated for the status quo and each alternative by dividing the total present value benefits by the total present value costs. Where benefits equal costs, the ratio will be 1. For benefits exceeding costs, the ratio will be more than 1, again preferable. In fact, the larger the number (within reason), the more attractive the alternative. On the other hand, where costs exceed benefits, the ratio will be less than 1. Breakeven will not be reached.

The Value of IT Investments: It’s not just Return On Investment

Leave a Comment


Leave a Reply

Denise Hill


Yes, And…..Not passing judgement or attribution to any agency or problem..there is an underlying assumption that the baseline program cost are known. If a program has been around for a long time or a distributed cost structure in play then the cost/benefit becomes a SWAG. Does that make the investment any less cost effective?

Tim Constantine

Thanks for the comment Denise. It’s a great point, and one that I plan to address further in a future post about the fallacy of a strict cost/benefit analysis and moving beyond traditional cost/benefit analysis. SWAG – I don’t think I’ve heard that before, but I Googled it and I like it!

Marcus Peacock

It’s interesting the Circular focuses on ‘net present value’ while the Guide uses a benefit/cost ratio (similar to how the Corps of Engineers evaluates projects). I think the former is much better than the latter for making decisions or setting priorities. A ratio is pretty binary — the important thing is whether you are above “1” or not. Unfortunately the actual size of the ratio can be manipulated by, for instance, calling a cost that would normally go in the denominator a “negative benefit” and putting it in the numerator. You can’t get a project to go above or below “1” but you can sure get it closer or further away from “1” but just changing the way you do the math. You can’t do that with net present value. You get one number (with uncertainty) and you can compare that number to other options on an apples to apples basis. So if you have a bunch of ‘good’ projects but a limited budget (sound familiar?), you can use the net present value estimates to figure out which ones you should consider doing first.