A comprehensive Cost/Benefit Analysis provides managers, users, and designers with the information necessary to evaluate alternative system development, enhancement, or improvement approaches. The analysis provides the estimated costs of developing and operating each feasible alternative and the benefits to be derived from each.

Therefore, Cost/Benefit Analysis is not simply a method of determining the least cost alternative, but a means of determining the most cost effective alternative.

Each Cost/Benefit Analysis must begin with the determination of the operational costs of the installed system - the "status quo" alternative. This critical step sets the stage for comparing alternatives against baseline costs.

Costs for the current environment are expressed in terms of total system operational costs, including State costs, projected over the systems life in accordance with the approved State plan and previously approved APDs, if applicable. This requires that States measure current costs and project anticipated costs over a period of time matching the systems life of the project.

Most "status quo" cost worksheets will reflect primarily recurring costs, such as systems support personnel and monthly lease expenses.

However, non-recurring costs may also apply. For example, the State may have approval for contract support services to meet special year-end processing demands.

Special problems arise when the status quo is no longer a satisfactory solution. For example, if the State does not redesign the current system for distributed processing, current processing workload growth will require a major systems augmentation (non-recurring capital expenditure for equipment purchase) and new systems facility (either a non-recurring capital expenditure or recurring site lease expenses, or both) - and may still fail to meet Federal program requirements.

In such instances, the status quo reinforces the need to develop new systems. By costing in the capital expenditure on the status quo alternative, the true systems life cost of the status quo is revealed. By claiming the "cost avoidance" as a benefit of the alternatives, achievement of the cost avoidance will later be measured, as the State compares actual to projected costs and benefits.

A State may at times be required to determine whether to characterize an element as a cost or a benefit. If so, the State should consider the following. If a "benefit" can be depicted as directly affecting a cost element of the status quo and can be predicted with virtual certainty to occur, then it should be factored into the cost side of the analysis. An example would be higher staff costs when a reduced personnel ceiling has been approved for the new system.

On the other hand, cost elements to be characterized as benefits are normally costs or savings which are more indirect or hidden, which may or may not accrue, and which must be monitored. Benefits can derive from elements such as greater productivity, reduced training costs, less overtime, and reduced staffing not planned and approved at the time of the cost/benefit analysis.

Following baseline costing, total systems life costs, including State costs, are defined for each alternative of the two to four found viable in the Feasibility Study's alternatives analysis.

For each alternative approach, costs are calculated for both the system developmental and operational stages. Non-recurring costs will dominate the early months and years of the developmental process. However, as the alternative is tested and moves into implementation, recurring costs will again predominate the cost profile.

In order to develop cost profiles, States must first identify all categories of costs that apply to the status quo and each alternative in each stage of the development and operational cycles. Cost categories include such diverse expenses as those for equipment, personnel, travel, training, utilities, supplies, conversion, site preparation, space, and overhead.

Next, those costs must be identified as either recurring or non-recurring costs.

Recurring costs are those which apply over a range of time - either months or throughout the systems life. Recurring costs will predominate in the "status quo" alternative and in the later years of alternatives as the solutions are implemented.

Non-recurring costs are one-time costs, frequently capital expenditures, expected to occur at a point in the future. Non-recurring costs may be overlooked if future workload growth is not considered. States should be especially careful to consider the effects of rising caseloads and peak transaction volumes on equipment and software capacities, by providing for scheduled upgrades or enhancements to the operational system if required.

(See Tables 3-1 and 3-2 on pages 3-4 and 3-5 for descriptions of recurring and non-recurring costs.)

Finally, to aid in the development of the recurring cost figures, each cost category should be identified as either a fixed, adjusted, or variable cost factor. Adjusted cost factors are those which increase over time, tied (for example) to contractual obligations. Variable factors are those which are volume sensitive.

The Cost Profile Worksheet (page 3-7) can be used to determine cost categories, types, and factors applicable to both the status quo and each feasible alternative. Representative cost categories are included in the worksheet.

Each alternative should be evaluated from the point of view of its developmental and operational costs, and separate worksheets prepared.

In projecting future costs, States should determine whether the analysis will be based on constant (real) dollars or current (nominal) dollars.

Constant dollar costs and benefits are costs and benefits which reflect the prices of the base year of the systems life. Constant dollars do not consider the effect of inflation, are normally used in cost/benefit analyses, and do not require justification to ACF. Constant dollars are then adjusted by present value discounting, described in a following section.

Current dollar costs and benefits are costs and benefits which have been adjusted to reflect the effect of inflation on prices. Current dollars are normally used in budget projections.

The forecasting of future benefits and costs becomes complicated when there has been an appreciable and persistent rise in inflation, significant enough to affect investment considerations. If current dollars are used, a three-step (rather than two-step) approach is required:

• Projecting constant dollar costs and benefits;
• Converting constant dollars to current dollars by factoring in inflation based on price indices, such as the Consumer Price Index or the Producer Price Index; and
• Applying present value to convert future dollars to today's dollars.

• Purchase
• Site Preparation/Modification
• Other

• ADP
• Data Communications
• Environ. Conditioning
• Security
• Other

• Operating System
• Applications
• Utilities
• Other

• Data
• Software
• Services

• Cost of Planning
• Cost of Conducting

• Salaries
• Benefits
• Contract Support Services
• Extraordinary Personnel Costs

• Development
• Trainee Expenses
• Trainer Expenses

• Project and Technical
• Management
• Incremental
• Lost Productivity

• Lease
• Maintenance Fees
• Other

• ADP
• Data Communications
• Environ. Conditioning
• Security
• Other

• Operating System
• Applications
• Utilities
• Other

• Salaries
• Benefits

• Contract
• Detailed/Tasked

• Primary Facilities
• Back-up Facilities

If a State decides to use current dollars, the decision and the proposed inflation factor should be described and justified in submissions to ACF.

The use of present value analysis, which involves the discounting of cash flows, should not be confused with the treatment of the estimated effects of inflation. Present value considers the effect of interest on money over time - not the effect of inflation on money over time. Present value is used in cost/benefit analysis, against both constant and current dollars.

Once costs have been identified and characterized, the challenge is to quantify the factors. Four methods, or a combination, are typically used:

• Estimation - sometimes referred to as the bottom-up method, in which each organization involved in system development, operation, and use estimates, averages, and projects its costs;
• Comparison - in which current costs on comparable systems are used as a baseline for the new system;
• Simulation - in which the process is analyzed and simulated to obtain costs; and/or
• Observation - in which processes are measured and recorded to provide estimates.

If there is a secret to successfully developing costs, it is to rationally and reasonably identify, apply, and project the costs for each alternative. Not all costs will apply - or apply the same way - to each alternative.

For each alternative (including the status quo) and for each year, costs should be developed using a format such as the Annual Cost Worksheet on page 3-9.

The top of the worksheet identifies the alternative, whether the costs are developmental or operational, whether dollars are constant or current, and the year of the cost estimates.

Separate areas in the worksheet provide space for both recurring and non-recurring costs, by month and for the year.

The costs (and later the benefits) of each alternative should be expressed in terms of their "present value." This allows the conversion of benefits and costs occurring at different times in the future to their current (that is, present) value, reflecting the time-value of money. Present value calculations equalize the comparison of alternatives when expenses are distributed unequally over time.

Present value calculations deal not with inflation, but with interest. Present value discounting is the inverse of compounding interest; it shrinks tomorrow's dollars to today's dollars by the difference of the compounded interest. Present value reflects the opportunity cost of money.

Present value analysis is based on two principals:

• Benefits accruing in the future are worth less than the same level of benefits that accrue now; and
• Costs that occur in the future are less burdensome than costs that occur now.

The current year establishes the time reference point for present value calculations.

Present value is calculated by multiplying costs by a predetermined factor (called discounting) based on the established discount rate and time period. The discount rates are published in tables, which factor in the amount of interest earned by the dollar invested today until the future dollar is spent.

For example, if we assume a 7 percent interest rate, $107 in estimated costs or projected benefits for next year would be worth $100 today - the present value.

Therefore, present value calculations discount - that is, reduce - costs or benefits projected to occur in future years to a common point in time so they can be compared.

For consistency in the Federal review and approval processes, ACF requires States to use a 7% present value factor in their submissions.

The year-end discount factors presented in the table assume end-of-year, lump sum costs and returns. When costs and returns occur in a steady stream, applying the mid-year discount factors may be more appropriate.

Once the annual costs for each alternative are identified, totaled, and discounted for present value, the systems life cost profiles (page 3-12) can be prepared.

These system life cost profiles are built from the totals calculated on each yearly cost worksheet. Present value totals are inserted: no recalculation is required.

Once the cost profiles have been developed, the State must identify the categories of benefits that apply to the status quo and each feasible alternative over the systems life. In addition, the State should verify that the benefits are properly categorized and are not, in fact, better described as costs.

Benefits should relate directly to the system objectives defined during the Feasibility Study, such as:

• Reduced Error Rates,
• Increased Collections,
• Reduced Costs,
• Reduced Staffing,
• Improved Security, and
• Improved Access or Interface.

Examples of system objectives are listed in the table on page 2-4.

Benefits, listed for the status quo and for each alternative, should be categorized as either quantitative or qualitative benefits. The Benefit Profile Worksheet on page 3-14 may be used. Note that some benefit categories, such as "more timely reporting" are listed under both quantitative and qualitative benefits, both or either of which might apply to a given situation.

Once the cost profiles have been developed, the State must identify the categories of benefits that apply to the status quo and each feasible alternative over the systems life. In addition, the State should verify that the benefits are properly categorized and are not, in fact, better described as costs.

Benefits should relate directly to the system objectives defined during the Feasibility Study, such as:

• Reduced Error Rates,
• Increased Collections,
• Reduced Costs,
• Reduced Staffing,
• Improved Security, and
• Improved Access or Interface.

Examples of system objectives are listed in the table on page 2-4.

Benefits, listed for the status quo and for each alternative, should be categorized as either quantitative or qualitative benefits. The Benefit Profile Worksheet on page 3-14 may be used. Note that some benefit categories, such as "more timely reporting" are listed under both quantitative and qualitative benefits, both or either of which might apply to a given situation.

Quantitative benefits are those for which a reasonable valuation may be predicted and projected. For example, the State should be able to calculate the value of reduced staffing or increased collections.

The objective of the benefit valuation process is to document the State's experience (the status quo), develop assumptions (such as average overpayment), and present a reasoned prediction of the value of the benefit to the State and Federal governments.

The standard of analysis and documentation should be to analyze, develop, substantiate, and present a logical argument supporting the apparent validity of the predicted value of the benefit - sufficient so that the worksheets would be capable of convincing a third party that the prediction is justified and meritorious.

Quantifying benefits is usually more difficult than estimating and predicting costs. Four methods, or a combination, are typically used:

• Estimation - in which each organization involved in system development, operation, and use estimates and projects the value of benefits, using averaging to reduce the potential for error;
• Comparison - in which current benefit values on comparable systems are used as a baseline;
• Simulation - in which the anticipated benefit is analyzed and simulated to obtain costs and values; and/or
• Observation - in which benefit processes are measured and recorded to provide estimates.

If there is a secret to valuing benefits, it is to rationally and reasonably identify, breakdown, apply, and project the costs and values for each alternative. Not all costs and values will apply - or apply the same way - to each alternative.

States will make their most effective arguments for Federal funding through quantified costs and benefits.

The Annual and Systems Life Quantified Benefits Worksheets on page 3-16 and 3-17 may be used to document the value of quantifiable benefits. Note that space has been allotted on the worksheets for the results of three critical determinations:

• Assumptions - on which the numbers and predictions are based;

• Basis for numbers - such as average value of overpayments during the last complete fiscal year; and
• Source for numbers - which specifies the documents (and their locations) from which numbers or predicted increases or decreases were pulled or calculated.

In most cases, addendum sheets should be attached with narrative and numbers explaining in more detail the origin, derivation, and calculation of the actual and predicted values. In addition, critical documents on which the numbers and predictions are based should be attached to the State's master copy of the Cost/Benefit Analysis.

Since the State is now required by regulation to accrue actual costs and benefits over time and since ACF will oversee cost/benefit actuals, care should be taken to document the process in full. The documentation should be sufficiently detailed so that personnel changes will not affect the State's ability to accrue, measure, and explain costs and benefits.

Once the State has prepared a set of benefit worksheets, the calculated benefits for the status quo and each alternative can be listed on the Annual Benefits Worksheet (see page 3-19).

As with costs, benefits of each alternative should be expressed in terms of their present value. Present value is calculated by multiplying the benefit values by a factor based on the established discount rate and time period. (See page 3-11 for a discount factor table based on 7%.)

Once the annual benefit values for each alternative are identified, totaled, and discounted for present value, the system life benefit profiles can be prepared. (See page 3-20.) These benefits profiles are built from the totals calculated on each yearly benefit worksheet. Present value totals are inserted: recalculation is not required.

Errors commonly made in analyzing benefits and costs include double counting, counting sunk costs, omitting costs, and applying cost categories unevenly.

Double counting occurs when the same expense is charged to more than one category. This could occur, for example, if the same or overlapping expenses to "clean up" and convert information and software were charged to the categories of conversion, studies, and in-house and contract personnel.

Including sunk costs is also an error in calculating cost/benefits. "Sunk costs" represent money already spent and such costs have no bearing on the analysis. Cost/benefit analysis addresses current and future costs, not past expenditures.

Omitting costs is a common error which can occur from overlooking overhead costs or hidden costs. Overhead costs may include charges for space, electricity, personnel benefits, and project or management overhead. Hidden costs are usually indirect costs or support activities existing elsewhere in the organization and sometimes charged at a standard rate by an internal billing system.

The error of uneven application of cost categories occurs when costs are applied to one alternative and not another to which it applies. For example, if an upgrade to accommodate workload growth is applied to the status quo - but not to the alternatives also needing additional future capacity - an error has been made.

Despite the preponderant weight given quantified benefits, qualitative benefits are also important in the evaluation of alternatives, gaining weight as the cost differential between alternatives narrows.

Contrasted to the ease of valuing such benefits as reduced staffing, assigning values to the benefits of improved security against unauthorized access or improved access to system information may lead to numbers which are far from unassailable.

If numbers cannot be reasonably defended, they should not be set forth as quantified benefit values. Instead, the State should acknowledge the benefits as qualitative - sometimes called intangible - benefits.

Qualitative benefits are linked to factors other than numbers. For example, qualitative benefits may be coupled with legislative mandates, socio-political edicts, or technical, functional, or programmatic considerations.

Benefits associated with legislative mandates are the most powerful. For example, an alternative which would modify a child support management information system to achieve enhanced "compatibility among the systems of different jurisdictions to permit periodic screening" would support the objectives of Public Law 96-265, which amended the Social Security Act with respect to provisions related to AFDC and Child Support programs.

Qualitative benefits, identified previously on Benefit Profile Worksheets (see page 3-14) for each alternative and the status quo, should now be:

• Listed in order of their relative importance,
• Linked to system objectives, and
• Assessed as to the measure of effectiveness of the benefit in meeting the system objective.

For example, the benefit of PC-based distributed systems might fulfill the "access to information" system development objective more effectively than a terminal-to-mainframe link. Note that in this example the State might derive quantitative benefits, based on access and wait times. However, qualitative benefits - such as increased worker satisfaction and enhanced access - may apply as well.

The State may use the Qualitative Benefits Worksheet on page 3-24 to develop the intangible benefits profile for the status quo and each alternative.

Four levels of effectiveness are included on the worksheet: very effective, effective, minimally effective, and not effective.

The latter is included so that the State has the option of pairing benefits and objectives in standard sets against which the alternatives can be measured. For example, the status quo might be rated "not effective" in the benefit/objective combination of work satisfaction/access to information.

Since the State's actual benefits will be monitored and measured against projected benefits, it is critical:

• Ed???? prediction of benefits' values.

After the initial cost/benefit profiles have been developed, the State should assign an independent team of reviewers to critique the profiles and to propose measurement and monitoring strategies.

The first step in comparing alternatives is to develop a Cost/Benefit Profile (see page 3-26) for the status quo and each alternative. The profile builds on information already developed during the analysis, including sections for:

• System Life Cost Profile,
• System Life (Quantitative) Benefits Profile,
• Cumulative Costs and Benefits, and
• Qualitative Benefits.

The profile aggregates information on systems life and cumulative costs and benefits used in the next stage of analysis - the comparison of alternatives. The system life and cumulative costs and benefits derive from the systems life profile worksheets (pages 3-12 and 3-20), while the qualitative benefits derive from the Qualitative Benefits Worksheet (page 3-24).

The Cost/Benefit Profile provides on a single sheet the most essential data pertinent to that alternative.

The State is now ready to compare the systems life cost and benefit values for the status quo and each alternative, transferring the key information from each Cost/Benefit Profile (page 3-26) to the Comparison of Alternatives worksheet (page 3-27).

Three methods are typically used by the States to compare alternatives:

Net Benefit (Cost) is calculated for the status quo and each alternative by subtracting the total present value costs from the total present value benefits. Where benefits exceed costs, the result will be a positive number, which is, of course, preferable. However, for the status quo and perhaps some alternatives, costs may exceed benefits and result in a negative number, suggesting no (quantifiable) payback on the alternative.

Net Benefit (Cost), sometimes referred to as Net Present Value, is the most straightforward comparison, showing which alternative is the most economical based on present value dollars.

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 Benefit/Cost Ratio provides a relative measure of an alternative's value - that is, a measure of the benefits obtained per dollar spent. If the ratio calculated for an alternative is 1.08, then for each dollar spent, the State estimates a return of $1.08 in benefits.

Therefore, the Benefit/Cost Ratio shows which alternative provides the largest return relative to the costs - as does the Net Benefit factor.

Breakeven or Payback is the calculation of how many months it will take for cumulative benefits to equal (then exceed) cumulative costs. These may be presented in two ways, numerically and graphically.

In the mathematical calculation, cumulative total costs are compared against cumulative total benefits to determine the month of breakeven or payback. Note that projected, cumulative numbers - not present value numbers - are used.

The breakeven point can be determined from the worksheets by a two-step process. First, the year of breakeven is determined by comparing the cumulative total projected costs and benefits on the system life worksheets. Then, the month of breakeven is established by comparing the cumulative total projected costs and benefits on the annual worksheets of the breakeven year.

Table 3 15: Graph: Systems Life Breakeven

In the graph above, the cumulative benefits breakeven at $5,000,000 each in the fourth year. This is readily apparent from the chart. When cumulative costs are equal to cumulative benefits, the lines intersect - the more cumulative benefits exceed cumulative costs, the wider the gap after intersection.

In some cases, it may be important to recover the initial costs of a project as quickly as possible. In those instances, the breakeven calculation may become the most important. Note, however, that alternatives which deliver the earliest breakeven may not have the most favorable benefit/cost ratio and net present value benefit.

The Comparison of Alternatives worksheet (page 3-27) is used to set the numbers side-by-side for the status quo and each alternative. Space is provided for:

• Total present value benefits,
• Total present value costs,
• Net benefit (cost),
• Benefit / Cost ratio, and
• Breakeven.

Breakeven should be stated in terms of months, as discussed above, based on the cumulative figures from the Annual Cost Worksheet (page 3-9) and the Annual Benefits Worksheet (page 3-19). Note that breakeven is based on projected costs - not present value costs.

Now that all quantitative facts about each alternative are side-by-side, the art of cost/benefit analysis comes into play. The term "art" is used because cost/benefit analysis is not simply a mathematical formula which dictates a decision, but a managerial decision-making tool.

Therefore, the decision should not be limited to an exercise of dividing or subtracting to determine which alternative has the highest net benefit, largest benefit/cost ratio, or the shortest breakeven period. In fact, an analysis which results in an unusually low breakeven should be cause for concern rather than excitement. Such a payback would be unlikely to occur, and payback must be monitored by the State.

Therefore, the first step in comparing alternatives should be assessing the range in the numbers among the alternatives. The manager / analyst should consider such questions as:

• How close are the numbers? Do they make sense?
• Is there an apparent winner? Does it seem a sensible selection?
• Do the results suggest that the assumptions and projections are reasonable?
• If the numbers are close, should the assumptions be reassessed and the numbers checked for validity?
• If the range is wide, should the assumptions be reassessed and the numbers checked for validity?
• Do I believe this payback will occur?

Next, the numbers - or quantitative part of the analysis - should be checked against the qualitative or intangible benefits. The Cost/Benefit Profiles should be placed side-by-side. For the qualitative part of the analysis, the manager / analyst should consider similar questions:

• How close are the benefit profiles? Are there appreciable differences among them?
• Is there an apparent winner, based on qualitative factors? Does it also win in numbers?
• If the benefits are close, should the assumptions be reassessed?
• If the benefits vary widely, should the assumptions be reassessed?
• Do I believe these benefits will occur?

Significant differences in qualitative benefits, which require consideration in the final analysis, should be noted on the Comparison of Alternatives worksheet. In fact, the manager can develop an evaluation scheme for the qualitative factors to aid in the analysis. Techniques include:

• Ranking - which involves ranking benefits by their relative importance and determining the degree to which each alternative achieves the benefits, and
• Weighting - which involves scoring each alternative on the extent of benefits projected or assigning values to benefits.

With these methods, numbers are assigned in the form of ranks or scores. Although this lends the appearance of a quantified process, the determinations are, by their nature, subjective.

Once this is done, the State should "step back and look at the big picture." This process should test the validity of the numerical results in light of the benefits. For example, if there is a very low cost differential between two alternatives, offset by a very large intangible benefits differential, then the alternatives certainly deserve a closer examination. Questions to be asked include:

• Would I select the same alternative considering quantitative and qualitative factors separately?
• Can I attach a value to a benefit that I was unable to cost before?
• If I am unable to quantify the benefit, can I prepare a reasonable justification that the superior benefit is worth the expense?
• Can I apply probability analysis to refine my determination?

If serious questions arise, the cost and benefit worksheets, beginning with the assumptions and including all ensuing calculations, should be redone.

The reconsideration of assumptions is sometimes referred to as "sensitivity analysis" - that is, testing the sensitivity of the results to changes in the assumptions. Such analysis considers how vulnerable the results are to a change in assumptions.

Generally, sensitivity analysis centers on the factors expected to have the most effect on the net present value determination. Such factors can include:

• Cost estimates,
• Workload projections,
• Project implementation schedule, or
• Offsetting benefit values.

There are four steps in testing the sensitivity of a factor and its effect on the cost/benefit determination:

• Select the factor to be tested;
• Hold all other factors in the analysis constant;
• Rework the analysis, varying the estimates for the factor under consideration; and
• Check the results to see if the ranking of alternatives is materially affected.

Qualitative benefits are subjective determinations and balance the objective determinations related to quantified benefits. Both objective and subjective determinations contribute to effective decision-making.

Simply because a benefit is listed as a qualitative factor does not mean that it lacks cost implications. It simply means a different logical approach is needed, because the value is either unpredictable or difficult to establish and defend.

For example, suppose in comparing two alternatives, a manager determines that a $10,000 cost differential separates two alternatives. But the more costly alternative delivers enhanced compatibility between State benefits systems, improved public assistance, improved management information, and improved security. Although the manager could not directly apply values to those benefits earlier in the analysis, the manager may now be in a position to argue that the $10,000 cost differential is worth the increased benefits. In fact, at this point, some calculations and assumptions might be made to support the decision.

The point is - qualitative benefits have values, but they cannot be analyzed in the same manner as quantified benefits.

Nonetheless, States should keep in mind - the more objective (quantified) the profile, the easier the decisions for States in selecting systems alternatives and for ACF in approving Federal funding support.

After analyzing both the quantitative and qualitative factors, the results of the analysis should be formally verified, and the best alternative selected. Finally, the decision-making process - especially the basis for selection of the chosen alternative - should be thoroughly described and documented.