Economic Analysis Primer

Benefit-Cost Analysis

LCCA is a useful economic tool for selecting among alternatives where benefits of the possible project alternatives are essentially identical. In many cases, however, alternatives that an agency is considering may not generate identical benefits. For instance, when reconstructing a road, an agency may wish to consider reconstructing it as is or with additional lanes. The appropriate economic tool for these instances is benefit-cost analysis (BCA), which considers life-cycle benefits as well as life-cycle costs (see box).

BCA attempts to capture all benefits and costs accruing to society from a project or course of action, regardless of which particular party realizes the benefits or costs, or the form these benefits and costs take. Used properly, BCA reveals the economically efficient investment alternative, i.e., the one that maximizes the net benefits to the public from an allocation of resources.

BCA is not the same thing as financial analysis. Financial analysis is concerned with how to fund a project over its lifespan and measures the adequacy of current and future funds and revenues to cover the cost of building, operating, and maintaining the project. While financial analysis is an important part of project management, the economic merit of the project as measured by BCA is generally not affected by how the project is financed.

The Benefit-Cost Analysis Process

In BCA, the analyst applies a discount rate to the benefits and costs incurred in each year of the project's life cycle. This exercise yields one or more alternative measures of a project's economic merit.

The BCA process (see box) begins with the establishment of objectives for an improvement to a highway facility, such as reducing traffic congestion or improving safety. A clear statement of the objective(s) is essential to reduce the number of alternatives considered. The next step is to identify constraints (policy, legal, natural, or other) on potential agency options and specify assumptions about the future, such as expected regional traffic growth and vehicle mixes over the projected lifespan of the improvement.

Having identified objectives and assumptions, the analyst (or analytical team) then develops a full set of reasonable improvement alternatives to meet the objectives. This process begins with the development of a "do minimal" option, known as the base case. The base case represents the continued operation of the current facility under good management practices but without major investments.³ Under these "do minimal" conditions, the condition and performance of the base case would be expected to decline over time. Reasonable improvement alternatives to the base case can include a range of options, from major rehabilitation of the existing facility to full-depth reconstruction to replacement by a higher volume facility. Such alternatives will often involve construction, but alternatives that improve highway operations (such as the use of intelligent transportation systems) or manage travel demand (such as incentives for off-peak travel) are suitable for consideration.

To ensure that the alternatives can be compared fairly, the analyst specifies a multiyear analysis period over which the life-cycle costs and benefits of all alternatives will be measured. The analysis period is selected to be long enough to include at least one major rehabilitation activity for each alternative.

Ideally, the level of effort allocated to quantifying benefits and costs in the BCA is proportional to the expense, complexity, and controversy of the project. Also, to reduce effort, the alternatives are screened initially to ensure that the greatest share of analytical effort is allocated to the most promising ones. Detailed analysis of all alternatives is usually not necessary.

When an alternative is expected to generate significant net benefits to users, particularly in the form of congestion relief, the analyst evaluates the effect that the alternative would have on the future traffic levels and patterns projected for the base case (see section on Forecasting Traffic for Benefit Calculations, page 27). Changes in future traffic flows in response to an alternative will affect the calculation of project benefits and costs.

The investment costs, hours of delay, crash rates, and other effects of each alternative are measured using engineering methods and then compared to those of the base case, and the differences relative to the base case are quantified by year for each alternative. The analyst assigns dollar values to the different effects (e.g., the fewer hours of delay associated with an alternative relative to the base case are multiplied by a dollar value per hour) and discounts them to a present value amount. Risk associated with uncertain costs, traffic levels, and economic values also is assessed (see section on Risk Analysis, page 30).

Any alternative where the value of discounted benefits exceeds the value of discounted costs is worth pursuing from an economic standpoint. For any given project, however, only one design alternative can be selected. Usually, this alternative will be the economically efficient one, for which benefits exceed costs by the largest amount.

Based on the results of the BCA and associated risk analysis, the analyst prepares a recommendation concerning the best alternative from an economic standpoint. It is good practice to document the recommendation with a summary of the analysis process conducted. In some cases, particularly for larger projects, this summary will include a discussion about the economic impact analysis conducted based on the results of the BCA (see section on Economic Impact Analysis, page 32).

Benefit and Cost Elements to Include

Table 2 lists the benefit and cost categories and elements that are generally included in BCA.

The user elements in Table 2 are labeled as "cost/benefit" rather than "cost" or "benefit" only. This is because improvement alternatives are being compared to the base case (the "do minimal" option), and each may have a different impact on users. For instance, one alternative may reduce crash rates (a benefit) relative to the base case; another alternative may increase crash rates (a cost, also called a negative benefit or disbenefit) relative to the base case. In BCA, most, if not all, agency and user elements will vary relative to the base case - thus, contrary to LCCA, all elements must typically be considered and quantified.

Note that toll receipts and other user fees are not listed as benefits or costs in Table 2. Rather, they represent transfers of some of a project's benefits from users to the agency operating the project (see box).

Many people are puzzled about how economists assign monetary values to highway project benefits and costs. For instance, how does one value an hour of travel time, or a crash? The valuation of each of the major elements listed in Table 2 is described below.

Agency costs. The assignment of monetary values to the design and construction of a project is perhaps the easiest valuation concept to understand. Engineers estimate these costs based on past experience, bid prices, design specifications, materials costs, and other information. Care must be taken to make a complete capital cost estimation, including contingencies and administrative expenses such as internal staff planning and overhead costs. A common error in economic analysis and budgeting is the underestimation of project construction and development costs. Particular care should be used when costing large or complicated projects.

Expenses associated with a project's financing, such as depreciation and interest payments, are not included in the BCA. The equivalent value of such expenses is already captured in the BCA through the application of the discount rate to the agency cost of the project. Adding depreciation or interest expense to agency costs in BCA would in most cases lead to double counting of costs.

Travel time and delay. An hour of travel associated with a business trip or commerce is usually valued at the average traveler's wage plus overhead - representing the cost to the traveler's employer. Personal travel time (either for commuting or leisure) is usually valued as a percentage of average personal wage and/or through estimates of what travelers would be willing to pay to reduce travel time. The U.S. Department of Transportation (USDOT) recommends that analysts value local personal travel time at 50 percent of average wage (see "Departmental Guidance for the Valuation of Travel Time in Economic Analysis," available on the internet, for additional guidance). The value of reduced travel time often accounts for the greatest share of a transportation project's benefits.

Crashes. The assignment of monetary values to changes in crash rates or severities can provoke controversy because crashes often involve injury or loss of life. The use of reasonable crash values is critical, however, to avoid underinvesting in highway safety. Economists often use the dollar amounts that travelers are willing to pay to reduce their risk of injury or death to estimate monetary values for fatalities and injuries associated with crashes. Medical, property, legal, and other crash-related costs are also calculated and added to these amounts. USDOT offers extensive guidance on this subject (see "Revision of Departmental Guidance on Treatment of the Value of Life and Injuries," and "The Economic Impact of Motor Vehicle Crashes," 2000 (DOT HS 809 446), available on the internet).

Vehicle operating costs. The costs of owning and operating vehicles can be affected by a project due to the changes that it causes in highway speeds, traffic congestion, pavement surface, and other conditions that affect vehicle fuel consumption and wear and tear. Accurate calculations of a project's effect on vehicle operating costs (VOC) require good information on the relationship of vehicle performance to highway conditions, and clear assumptions about future vehicle fleet fuel efficiency and performance. USDOT does not provide official guidance on estimating VOC, but useful information on the valuation of VOC (and other BCA elements) is provided in AASHTO's 1977 "Manual on User Benefit Analysis of Highway and Bus-Transit Improvements" and its successor document, and in the "Highway Economic Requirements System Volume IV: Technical Report" (FHWA-PL-00-028), Chapter 7. Benefits attributable to lower VOC are usually not a major component of a project's benefit stream.

Externalities. One of the more challenging areas of BCA is the treatment and valuation of the "externalities" of transportation projects. In economics, an externality is the uncompensated impact of one person's actions on the well-being of a bystander (see box). In the case of transportation investments, "bystanders" are the nonusers of the project. When the impact benefits the nonuser, this is called a positive externality. When the impact is adverse, this is called a negative externality.

Often, when there is talk about externalities of highways, the focus is on negative externalities. Negative externalities include undesirable effects of a project on air and water quality, noise and construction disruptions, and various community and aesthetic impacts. Positive externalities, however, also exist. A project may serve to reduce air or noise pollution from levels that would have otherwise prevailed without it.

Several methods exist for including externalities in BCA. In some cases, scientific and economic studies have revealed per-unit costs for air pollutants, for example, that can be incorporated directly into the BCA. Much uncertainty surrounds these valuations, however. Values can vary from project to project due to location, climate, and pre-existing environmental conditions. Risk analysis techniques (see section on Risk Analysis, page 30) can yield helpful information about the sensitivity of results to these uncertain values.

Externalities are specifically dealt with in environmental assessments required by the National Environmental Policy Act (NEPA). Where adverse impacts are identified, mitigation is required to avoid, minimize, or compensate for them. Required mitigation is part of the environmental decision, and the costs of mitigation will become "internalized" in the project's cost in the BCA. The BCA effort should be coordinated closely with the NEPA assessment (see box).

When an externality cannot be put into dollar terms, it can often be dealt with on a qualitative basis relative to other, monetized components of the BCA. If the measurable net benefits of a project are highly positive, the presence of minor unquantified externalities can be tolerated from an economic standpoint even if they are perceived to be negative. On the other hand, if the net benefits are very low, then the existence of significant unquantified negative externalities may tip the economic balance against the project.

Comparing Benefits to Costs

Once the analyst has calculated all benefits and costs of the project alternatives and discounted them, there are several measures to compare benefits to costs in BCA. The two most widely used measures are described below.

• Net present value (NPV). NPV is perhaps the most straightforward BCA measure. All benefits and costs over an alternative's life cycle are discounted to the present, and the costs are subtracted from the benefits to yield a NPV. If benefits exceed costs, the NPV is positive and the project is worth pursuing. Where two or more alternatives for a project exist, the one with the highest NPV over an equivalent analysis period should usually be pursued. Policy issues, perceived risk, and funding availability, however, may lead to the selection of an alternative with a lower, positive NPV.
• Benefit-cost ratio (BCR):The BCR is frequently used to select among projects when funding restrictions apply. In this measure, the present value of benefits (including negative benefits) is placed in the numerator of the ratio and the present value of the initial agency investment cost is placed in the denominator. The ratio is usually expressed as a quotient (e.g., $2.2 million/$1.1 million = 2.0). For any given budget, the projects with the highest BCRs can be selected to form a package of projects that yields the greatest multiple of benefits to costs (see box, page 24).

FHWA recommends the use of either the NPV or BCR measures for most economic evaluations. Other BCA measures are available and may be used, however, depending on agency preference. For example, the equivalent uniform annual value approach converts the NPV measure into an annuity amount. The internal rate of return measure represents the discount rate necessary to yield an NPV of zero from a project's multiyear benefit and cost stream.

Misunderstandings

BCA is a powerful, informative tool available to assist planners, engineers, and decision makers. Agencies often avoid or underutilize BCA due to misconceptions about it.

In some cases, agency personnel are skeptical about the accuracy of BCA due to perceived uncertainties in measuring or valuing costs and benefits. In reality, there is much more substance to economic analysis techniques and values than is generally understood. Where uncertainty does exist, it can usually be measured and managed. It is helpful to remember that sound economic analysis reduces uncertainty. Not doing the analysis only serves to hide uncertainty from decision makers.

Another concern is that the workload involved in BCA may be excessive relative to agency resources. Once the engineering and economic capabilities are in place, however, BCA workloads diminish markedly. BCA level of effort should also reflect project cost, complexity, and controversy - routine projects may be analyzed with minimal effort.

Finally, some agencies are concerned that the results of BCA could conflict with preferred or mandated outcomes. In any situation, an objective and independent assessment of a project's economic consequences can contribute valuable information to the decision process. There are, however, valid reasons why decision makers may choose to override or constrain economic information. For example, if there are concerns that BCA results would disproportionately favor projects in urban areas, policy makers can initially apportion funds between urban and rural areas based on equity considerations. Urban projects would then compete based on their economic merits for the urban funds; rural projects would similarly compete for the rural funds.

Avoiding Pitfalls

As with any analytic method, BCA can give erroneous results if it is misused. Perhaps the foremost cause of error in BCA is the selection of an unrealistic base case. The base case must be premised on intelligent use and management of the asset during the analysis period. For instance, allowances should be made for traffic diversion and changing peak periods as congestion builds in the base case (the broader importance of accurate traffic forecasts is discussed in the next section, page 27). Failure to do this can lead to overly pessimistic estimates of delay levels in the base case, to which by comparison any alternative would look attractive. BCA results can also be biased by the comparison of only one design alternative to the base case, even though less costly alternatives exist. Proper BCA considers a full range of reasonable alternatives.

Another common BCA problem involves the evaluation of a "project" that is actually a combination of two or more independent or separable projects. In such cases, the net benefits of one project may hide the net costs of the other, or vice versa. Both of the projects would either be built or rejected if incorrectly joined together, when in fact one should be built and the other rejected.

BCA results can be erroneous if they do not include the correct cost or benefit elements or amounts associated with a project. This problem happens most often with the omission of user costs or major externalities (if present). In some cases, an agency may focus only on local costs and benefits, failing to include those that accrue outside its jurisdiction. Care must also be taken not to include "benefits" that are simply restatements of other benefits (or costs) measured elsewhere in the BCA. This latter error, a form of double counting, can occur when employment, business, or land use effects measured using economic impact analysis are added to the travel time saving, safety, and vehicle operating cost benefits of a project. A more thorough discussion of this latter issue is provided later in this primer, in the section on Economic Impact Analysis (page 32).

Tools

Many tools that can accommodate BCA are available. The majority of tools capture benefits and costs at the project level only, but some tools can estimate the net benefits of projects at the program level.

In the United States, perhaps the best-known BCA tool for highways is that presented in the 1977 AASHTO "Manual on User Benefit Analysis of Highway and Bus-Transit Improvements," referred to as the 1977 Redbook. This guide is being updated under a National Cooperative Highway Research Program contract and will be reissued. The Texas Transportation Institute developed the MicroBENCOST model to implement the guidance in the 1977 Redbook. A few States have developed their own BCA models.

Some software applications are specific to subsets of highway investments. For instance, the Federal Railroad Administration's Gradedec software is specific to the BCA evaluation of upgrades, separations, and closures of highway-rail grade crossings.

In 2000, FHWA released a State-level version of its Highway Economic Requirements System (HERS-ST). HERS-ST is a computer model that applies BCA to section-level highway data to predict system-wide investment requirements. HERS-ST considers capital improvements directed at correcting pavement, geometric, or capacity deficiencies. HERS-ST can determine the program funding levels required to achieve desired highway performance goals in a cost-beneficial way. Alternatively, the model can estimate the highway system performance that would result from various program-funding levels. An abundance of material concerning HERS-ST is available on the Office of Asset Management Web site, http://www.fhwa.dot.gov/infrastructure/asstmgmt/invest.htm.

³ The base case is sometimes called the "do nothing" option, but this term can be mistaken to mean that future management of the facility is not responsive to change. The term "do minimal" conveys the notion of ongoing managerial discretion to adjust to changing circumstances in the base case scenario.