Health & Medical Capital Budgeting at Cleveland Clinic Discussion

CHAPTER PROJECT RISK ANALYSIS 12 Learning Objectives Copyright 2020. AUPHA/HAP Book. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. After studying this chapter, readers should be able to • describe the three types of risk relevant to capital budgeting decisions, • discuss the techniques used in project risk assessment, • conduct a project risk assessment, • discuss several types of real options and their impact on a project’s value, and • explain how risk is incorporated into the capital budgeting process. Introduction Chapter 11 covered the basics of capital budgeting, including cash flow estimation, breakeven analysis, and profitability measures. This chapter extends the discussion of capital budgeting to include risk analysis, which is composed of three elements: (1) defining the type of risk relevant to the project, (2) measuring the project’s risk, and (3) incorporating that risk assessment into the capital budgeting decision process. Although risk analysis is a key element in all financial decisions, the importance of capital investment decisions to a healthcare organization’s success makes risk analysis vital. The higher the risk associated with an investment, the higher its required rate of return. This principle is just as valid for healthcare businesses that make capital expenditure decisions as it is for individuals who make personal investment decisions. Thus, the ultimate goal in project risk analysis is to ensure that the cost of capital used as the discount rate in a project’s profitability analysis properly reflects the riskiness of that project. The corporate cost of capital, which is covered in detail in chapter 9, reflects the cost of capital to the organization on the basis of its aggregate risk—that is, the riskiness of the business’s average project. In project risk analysis, a project’s risk is assessed relative to the firm’s average project: Does the project have average risk, below-average risk, or 459 EBSCO Publishing : eBook Collection (EBSCOhost) – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY AN: 2329791 ; George H. Pink, Paula H. Song.; Gapenski’s Understanding Healthcare Financial Management, Eighth Edition Account: shapiro.main.eds 460 G a p en s k i’s U n d e r s ta n d i n g H e a l th c a re F inanc ial Managem ent above-average risk? The corporate cost of capital is then adjusted to reflect any differential risk, resulting in a project cost of capital. In general, high-risk projects are assigned a project cost of capital that is higher than the corporate cost of capital, average risk projects are evaluated at the corporate cost of capital, and low-risk projects are assigned a discount rate that is less than the corporate cost of capital. (Note that when capital budgeting is conducted at the divisional level, the adjustment process is handled in a similar manner but the starting value is the divisional cost of capital.) Types of Project Risk Three types of project risk can be defined and, at least in theory, measured: 1. Stand-alone risk, which views the risk of a project as if it were held in isolation and hence ignores portfolio effects in the firm and among equity investors 2. Corporate risk, which views the risk of a project in the context of the business’s portfolio of projects 3. Market risk, which views a project’s risk from the perspective of the business’s owners, who are assumed to hold a well-diversified portfolio of stocks1 The type of risk that is most relevant to a particular capital budgeting decision depends on the business’s ownership and the number of projects the business operates. Stand-Alone Risk Stand-alone risk is present in a project whenever there is a chance of a return that is less than the expected return. A project is risky whenever its cash flows are not known with certainty because uncertain cash flows mean uncertain profitability. Furthermore, the greater the probability of a return far below the expected return, the greater the risk. Stand-alone risk can be measured by the standard deviation of the project’s profitability (return on investment [ROI]), as measured typically by net present value (NPV) or internal rate of return (IRR). Because standard deviation measures the dispersion of a distribution about its expected value, the larger the standard deviation, the greater the probability that the project’s profitability (NPV or IRR) will be far below that expected. An alternative measure of stand-alone risk is the project’s coefficient of variation, which is the standard deviation divided by the project’s expected NPV. Conceptually, stand-alone risk is relevant in only one situation: when a EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use C hap ter 12: Proj ec t Risk A naly sis not-for-profit firm is evaluating its first project. In this situation, the project will be operated in isolation, so no portfolio diversification is present; that is, the business does not have a collection of different projects, nor does it have stockholders who hold diversified portfolios of stocks. Corporate Risk In reality, businesses usually offer many different products or services and thus can be thought of as having a large number (perhaps even hundreds) of individual projects. For example, MinuteMan Healthcare, a New England HMO (health maintenance organization), offers healthcare services to a large number of diverse employee groups in numerous service areas, and each different group can be considered a separate project. In this situation, the stand-alone risk of a project (service line) under consideration by MinuteMan is not relevant because the project will not be held in isolation. The relevant risk of a new project to MinuteMan is its contribution to the HMO’s overall risk—the impact of the project on the variability of the overall profitability of the business. This type of risk, which is relevant when the project is part of a not-for-profit business’s portfolio of projects, is called corporate risk. A project’s corporate risk depends on the context (i.e., the firm’s other projects), so a project may have high corporate risk to one business but low corporate risk to another, particularly when the two businesses operate in widely different industries. Market Risk Market risk is generally viewed as the relevant risk for projects being evaluated by investor-owned businesses. The goal of shareholder (owner) wealth maximization implies that a project’s returns as well as its risk should be defined and measured from the owners’ perspective. The riskiness of an individual project to a well-diversified owner is not the risk the project would have if it were owned and operated in isolation (i.e., stand-alone risk), nor is it the contribution of the project to the riskiness of the business (i.e., corporate risk). Most business owners hold a large diversified portfolio of stocks of many firms, which can be thought of as a large diversified portfolio of individual projects. Thus, the risk of any single project to a for-profit business’s owners is its contribution to the riskiness of their well-diversified stock portfolios. 1. What are the three types of project risk? 2. How is each type of project risk measured, both in absolute and relative terms? SELF-TEST QUESTIONS EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use 461 462 G a p en s k i’s U n d e r s ta n d i n g H e a l th c a re F inanc ial Managem ent Relationships Among Stand-Alone, Corporate, and Market Risks After discussing the three types of project risk and the situations in which each is relevant, it is tempting to say that stand-alone risk is almost never important because not-for-profit businesses should focus on a project’s corporate risk and investor-owned businesses should focus on a project’s market risk. Unfortunately, the situation is not that simple. First, it is almost impossible in practice to quantify a project’s corporate or market risk because it is extremely difficult—some practitioners would say impossible—to estimate the prospective return distributions for given economic states for either the project, the firm as a whole, or the market. If these return distributions cannot be estimated, the appropriate beta cannot be estimated, and hence a project’s corporate or market risk cannot be quantified. Fortunately, as demonstrated in the next section, it is possible to get a rough idea of the relative stand-alone risk of a project. Thus, managers can make statements such as “project A has above-average risk, project B has below-average risk, and project C has average risk,” all in the stand-alone sense. After a project’s stand-alone risk has been assessed, the primary factor in converting stand-alone risk to corporate or market risk is correlation. If a project’s returns are expected to be highly positively correlated with the firm’s returns, high stand-alone risk translates to high corporate risk. Similarly, if the firm’s returns are expected to be highly correlated with the stock market’s returns, high corporate risk translates to high market risk. The same relationships hold when the project is judged to have average or low standalone risk. Most projects will be in a firm’s primary line of business and hence will be in the same line of business as the firm’s average project. Because all projects in the same line of business are generally affected by the same economic factors, such projects’ returns are usually highly correlated. When this situation exists, a project’s stand-alone risk is a good proxy for its corporate risk. Furthermore, most projects’ returns are also positively correlated with the returns on other assets in the economy; that is, most assets have high returns when the economy is strong and low returns when the economy is weak. When this situation holds, a project’s stand-alone risk is a good proxy for its market risk. Thus, for most projects, the stand-alone risk assessment also provides good insights into a project’s corporate and market risk. The only exception is a situation in which a project’s returns are expected to be independent of, or negatively correlated to, the business as a whole. In these situations, considerable judgment is required because the stand-alone risk assessment will over-state the project’s corporate risk. Similarly, if a project’s returns EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use C hap ter 12: Proj ec t Risk A naly sis 463 are expected to be independent of or negatively correlated to the market’s returns, the project’s stand-alone risk will overstate its market risk. An additional problem arises with investor-owned healthcare businesses. Finance theory specifies that investor-owned businesses should focus on market risk when making capital budgeting decisions. However, most healthcare businesses (even proprietary ones) have corporate goals that focus on the provision of quality healthcare services in addition to owner (shareholder) wealth maximization. Furthermore, a proprietary healthcare business’s stability and financial condition, which primarily depend on corporate risk, are important to all the firm’s other stakeholders: its managers, physicians, patients, community, and so on. Some financial theorists even argue that stockholders, including those that are well diversified, consider factors other than market risk when setting required returns. This point is especially meaningful for small businesses because their owners and managers are not well diversified in their relationship to the business. Considering all the factors, it may be reasonable for managers of investor-owned healthcare businesses, particularly small ones, to be just as concerned about corporate risk as are managers of not-for-profit businesses. Fortunately, in most real-world situations, a project’s risk in the corporate sense will be the same as its risk in the market sense.2 1. Name and define the three types of risk relevant to capital budgeting. 2. How are these risks related? 3. Should managers of investor-owned providers focus exclusively on a project’s market risk? SELF-TEST QUESTIONS Risk Analysis Illustration To illustrate project risk analysis, consider Ridgeland Community Hospital’s evaluation of a new MRI (magnetic resonance imaging) system presented in chapter 11. Exhibit 12.1 contains the project’s cash flow analysis. If all of the project’s component cash flows were known with certainty, the project’s projected profitability would be known with certainty and hence the project would have no risk. However, in most project analyses, future cash flows— and hence profitability—are uncertain and, in many cases, highly uncertain, so risk is present. The component cash flow distributions and their correlations with one another determine the project’s profitability distribution and hence the On the web at: ache.org/HAP/ PinkSong8e EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use Profitability measures: Net present value (NPV) = $82,493. Internal rate of return (IRR) = 11.1%. 1. System cost ($1,500,000) 2. Related expenses) (1,000,000) 3. Gross revenues 4. Deductions 5. Net revenues 6. Labor costs 7. Maintenance costs 8. Supplies 9. Incremental overhead 10. Depreciation 11. Operating cash flow 12. Taxes 13. Net operating cash flow 14. Depreciation 15. Net salvage value 16. Net cash flow ($2,500,000) 0 $1,050,000    262,500 $   787,500 52,500 157,500 31,500 10,500 350,000 $    185,500 0 $    185,500 350,000 $  535,500 $   510,000 2 $ 562,275 $1,102,500   275,625 $  826,875 55,125 165,375 33,075 11,025 350,000 $   212,275 0 $   212,275 350,000 3 Annual Cash Flows $1,000,000    250,000 $   750,000 50,5000 150,000 30,000 10,000 350,000 $    160,000 0 $    160,000 350,000 1 EXHIBIT 12.1 Ridgeland Community Hospital: MRI Site Cash Flow Analysis $ 590,389 $1,157,625 289,406 $  868,219 57,881 173,644 34,729 11,576 350,000 $ 240,389 0 $ 240,389 350,000 4 $1,215,506   303,876 $    911,630 60,775 182,326 36,465 12,155 350,000 $  269,908 0 $  269,908 350,000 750,000 $1,369,908 5 464 G a p en s k i’s U n d e r s ta n d i n g H e a l th c a re F inanc ial Managem ent C hap ter 12: Proj ec t Risk A naly sis 465 project’s risk. In the following sections, three quantitative techniques for assessing a project’s risk are discussed: (1) sensitivity analysis, (2) scenario analysis, and (3) Monte Carlo simulation. In a later section, we present a qualitative approach to risk assessment. 1. What condition creates project risk? 2. What makes one project riskier than another? 3. What type of risk is initially assessed? SELF-TEST QUESTIONS Sensitivity Analysis Historically, sensitivity analysis has been classified as a risk assessment tool. In reality, it is not very useful in assessing a project’s risk. However, it does have significant value in project analysis, so we discuss it in some detail here. Many of the variables that determine a project’s cash flows are subject to some type of probability distribution, not known with certainty. If the realized value of such a variable is different from its expected value, the project’s profitability will differ from its expected value. Sensitivity analysis indicates exactly how much a project’s profitability—NPV, IRR, or modified internal rate of return (MIRR)—will change in response to a given change in a single input variable, with all other input variables held constant. Sensitivity analysis begins with the base case developed using expected values (in the statistical sense) for all uncertain variables. For example, assume that Ridgeland’s managers believe that all of the MRI project’s component cash flows—except for weekly volume and salvage value—are known with relative certainty. The expected values for these variables (volume = 40, salvage value = $750,000) were used in exhibit 12.1 to obtain the base case NPV of $82,493. Sensitivity analysis is designed to provide managers with the answers to such questions as, What if volume turns out to be more or less than the expected level? What if salvage value turns out to be more or less than expected? (Typically, more than two variables would be examined in a sensitivity analysis. We use only two to keep the illustration manageable.) In a sensitivity analysis, each uncertain input variable typically is changed by a fixed percentage amount above and below its expected value, while all other variables are held constant at their expected values. Thus, all input variables except one are held at their base case values. The resulting NPVs (or IRRs or MIRRs) are recorded and plotted. Exhibit 12.2 contains the NPV sensitivity analysis for the MRI project, assuming that there are two uncertain variables: (1) volume and (2) salvage value. On the web at: ache.org/HAP/ PinkSong8e EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use 466 G a p en s k i’s U n d e r s ta n d i n g H e a l th c a re F inanc ial Managem ent EXHIBIT 12.2 MRI Project Sensitivity Analysis Net Present Value Change from Base Case Level (%) Volume Salvage Value –30 –20 –10 0 +10 +20 +30 ($814,053) (515,193) (216,350) 82,493 381,335 680,178 979,020 ($ 57,215) (10,646) 35,923 82,493 129,062 175,631 222,200 Note that the NPV is a constant $82,493 when there is no change in either of the uncertain variables because a 0 percent change recreates the base case. The values in exhibit 12.2 give managers a feel for which input variable will have the greatest impact on the MRI project’s profitability—the larger the NPV change for a given percentage input change, the greater the impact. Considering only these two variables, we see that the MRI project’s NPV is affected by changes in volume to a much greater degree than it is by changes in salvage value. Often, the results of sensitivity analyses are shown in graphical form. For example, the exhibit 12.2 sensitivity analysis is graphed in exhibit 12.3. Here, the slopes of the lines show how sensitive the MRI project’s NPV is to changes in each of the uncertain input variables—the steeper the slope, the more sensitive the NPV is to a change in the variable. Note that the sensitivity lines intersect at the base case values—0 percent change from base case level and $82,493. Also, spreadsheet models are ideally suited for performing sensitivity analyses because such models automatically recalculate NPV when an input value is changed and facilitate graphing.3 Exhibit 12.3 illustrates that the MRI project’s NPV is very sensitive to volume and only mildly sensitive to changes in salvage value. A sensitivity plot that has a negative slope indicates that increases in the value of that variable decrease the project’s NPV. If two projects were being compared, the one with the steeper sensitivity lines would be regarded as riskier because a relatively small error in estimating a variable—for example, volume—would produce a large difference in the project’s realized NPV. Thus, a realized volume that turns out to be smaller than that expected means that the project’s actual NPV will be far less than that expected. If information were available on the sensitivity of NPV to input changes to Ridgeland’s average project, similar judgments regarding the riskiness of the MRI project could be made, but they would be relative to the firm’s average project. EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use C hap ter 12: Proj ec t Risk A naly sis 467 EXHIBIT 12.3 Sensitivity Analysis Graph Although sensitivity analysis historically has been thought of as a risk assessment tool, it has severe limitations in this role. For example, suppose that Ridgeland had a contract with an HMO that guaranteed a minimum MRI volume at a fixed reimbursement rate. In that situation, volume would not contribute to project risk at all, despite the sensitivity analysis showing NPV to be highly sensitive to changes in volume. In general, a project’s stand-alone risk depends on the sensitivity of its profitability to changes in key input variables and the ranges of likely values of these variables. Because sensitivity analysis considers only the first factor, its results can be misleading. Furthermore, sensitivity analysis does not consider interactions among the uncertain input variables; it considers each variable independently. Despite its shortcomings in risk assessment, sensitivity analysis does provide managers with valuable information. First, it provides some breakeven information about the project’s uncertain variables. For example, exhibits 12.2 and 12.3 show that just a small decrease in expected volume makes the project unprofitable, whereas the project remains profitable even if salvage value falls by more than 10 percent. Although somewhat rough, this breakeven information is clearly valuable to Ridgeland’s managers. (The breakeven points can be easily refined by using Excel’s Goal Seek capability.) Second, and perhaps more important, sensitivity analysis helps managers identify which input variables are most critical to the project’s profitability and hence to the project’s financial success. In this MRI example, volume is EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use 468 G a p en s k i’s U n d e r s ta n d i n g H e a l th c a re F inanc ial Managem ent clearly the key input variable of the two that were examined, so Ridgeland’s managers should ensure that the volume estimate is the best possible. The concept here is that Ridgeland’s managers have a limited amount of time to spend on analyzing the MRI project, and sensitivity analysis enables them to focus on what’s most important. The ability to identify the critical input variables is also useful postaudit. If the project is performing poorly and changes must be made, such changes will have the greatest positive impact if they are made to one of the critical variables. In our illustration, if the MRI project is initiated but its profitability is not meeting forecasts, it clearly is better to focus on increasing volume than on increasing the salvage value. SELF-TEST QUESTIONS 1. Briefly describe sensitivity analysis. 2. What type of risk does it attempt to measure? 3. Is sensitivity analysis a good risk assessment tool? If not, what is its value in the capital budgeting process? Scenario Analysis On the web at: ache.org/HAP/ PinkSong8e Scenario analysis is a stand-alone risk-analysis technique that considers (1) the sensitivity of NPV or another profitability measure to changes in key variables, (2) the likely range of variable values, and (3) the interactions among the variables. To conduct a scenario analysis, managers pick a “bad” set of circumstances (e.g., low volume, low salvage value), an average or “most likely” set, and a “good” set (e.g., high volume, high salvage value). The resulting input values are then used to create a probability distribution of NPV. For an illustration of scenario analysis, assume that Ridgeland’s managers regard a drop in weekly volume below 30 scans as very unlikely; they also feel that a volume above 50 is also improbable. On the other hand, salvage value can be as low as $500,000 or as high as $1 million. The most likely values are 40 scans per week for volume and $750,000 for salvage value. Thus, a volume of 30 and a $500,000 salvage value define the lower bound (or worst-case scenario), while a volume of 50 and a salvage value of $1 million define the upper bound (or best-case scenario). Ridgeland can now use the worst-, most likely, and best-case values for the input variables to obtain the NPV corresponding to each scenario. Ridgeland’s managers used a spreadsheet model to conduct the analysis, and exhibit 12.4 summarizes the results. The most likely case results in a positive EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use C hap ter 12: Proj ec t Risk A naly sis 469 NPV, the worst case produces a large negative NPV, and the best case results in an even larger positive NPV. These results, along with each scenario’s probability of occurrence, can now be used to determine the expected NPV and standard deviation of NPV. Suppose that Ridgeland’s managers estimate that there is a 20 percent chance that the worst case will occur, a 60 percent chance that the most likely case will occur, and a 20 percent change that the best case will occur. Of course, it is difficult to estimate scenario probabilities with any confidence, and, in most situations, the probabilities used will not be symmetric. For example, in an environment of increasing managed care penetration and increasing competition among providers, the probability may be higher for the worst-case scenario than for the best-case scenario. Exhibit 12.4 contains a discrete distribution of returns, so the expected NPV can be found as follows: Expected NPV = (0.20 × [−$819,844]) + (0.60 × $82,493) + (0.20 × $984,829) = $82,493. The expected NPV in the scenario analysis is the same as the base case NPV— $82,493. The results are consistent because, when coupled with the scenario probabilities, the values of the uncertain variables used in the scenario analysis—30, 40, and 50 scans for volume and $500,000, $750,000, and $1 million for salvage value—produce the same expected values that were used in the exhibit 12.1 base case analysis. If inconsistencies exist between the base case NPV and the expected NPV in the scenario analysis, the two analyses have inconsistent input assumptions. In general, such inconsistencies should be identified and removed to ensure that common assumptions are used throughout the project risk analysis. However, remember that our purpose here is to conduct a risk assessment, not to measure profitability. Ultimately, Scenario Worst case Most likely case Best case Expected value Standard deviation Probability of Outcome Volume Salvage Value NPV 0.20 0.60 0.20 30 40 50 $ 500,000 750,000 1,000,000 ($ 819,844) 82,493 984,829 40 $ 750,000 $ 82,493 $ 570,688 EXHIBIT 12.4 MRI Project Scenario Analysis EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use 470 G a p en s k i’s U n d e r s ta n d i n g H e a l th c a re F inanc ial Managem ent we will use the base case (expected value) cash flows to reassess the project’s profitability when we have completed the risk assessment. The standard deviation of NPV, as shown here, is $570,688: σNPV = [0.20 × (−$819,844 − $82,493)2 + 0.60 × ($82,493 − $82,493)2 + 0.20 × ($984,829 − $82,493)2]1/2 = $570,688, while the coefficient of variation (CV) of NPV is 6.9: CV = σ NPV $570, 688 = = 6.9. Expected NPV $82, 493 The MRI project’s standard deviation and coefficient of variation measure its stand-alone risk. Suppose that when a similar scenario analysis is applied to Ridgeland’s aggregate cash flows (average project), the result is a coefficient of variation of NPV in the range of 2.5 to 5.0. Then, on the basis of its stand-alone risk measured by coefficient of variation, along with subjective judgments, Ridgeland’s managers might conclude that the MRI project is riskier than the firm’s average project, so it would be classified as a high-risk project. Scenario analysis can also be interpreted in a less mathematical way. The worst-case NPV—a loss of about $800,000—is an estimate of the worst possible financial consequences of the MRI project. If Ridgeland can absorb such a loss in value without much impact on its financial condition, the project does not pose significant financial danger to the hospital. Conversely, if such a loss would mean financial ruin for the hospital, its managers might be unwilling to undertake the project, regardless of its profitability under the most likely and best-case scenarios. Note that the risk of the project is not changing in these two situations. The difference is in the organization’s ability to bear the risk inherent in the project. While scenario analysis provides useful information about a project’s stand-alone risk, it is limited in two ways. First, it considers only a few discrete states of the economy and hence provides information on only a few potential profitability outcomes for the project. In reality, an almost infinite number of possibilities exist. Although the illustrative scenario analysis contained only three scenarios, it can be expanded to include more states of the economy— say, five or seven. However, there is a practical limit on how many scenarios can be included in a scenario analysis. Second, scenario analysis—at least as normally conducted—implies a definite relationship among the uncertain variables involved. For example, our analysis assumed that the worst value for volume (30 scans per week) EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use C hap ter 12: Proj ec t Risk A naly sis would occur at the same time as the worst value for salvage value ($500,000) because the worst-case scenario is defined by combining the worst possible value of each uncertain variable. Although this relationship (all worst values occurring together) may hold in some situations, it may not hold in others. If volume is low, for example, maybe the MRI will withstand less wear and tear and hence be worth more after five years of use. The worst value for volume, then, should be coupled with the best salvage value. Conversely, poor volume may be symptomatic of poor medical effectiveness of the MRI and hence lead to limited demand for used equipment and a low salvage value. Scenario analysis tends to create extreme profitability values for the worst and best cases because it automatically combines all worst and best input values, even if these values have only a remote chance of occurring together. This problem can be mitigated, but not eliminated, by assigning relatively low probabilities to the best and worst cases. The next section describes a method of assessing a project’s stand-alone risk that deals with these two problems. 1. Briefly describe scenario analysis. 2. What type of risk does it attempt to measure? 3. What are its strengths and weaknesses? SELF-TEST QUESTIONS Monte Carlo Simulation Monte Carlo simulation, so named because it developed out of work on the mathematics of casino gambling, describes uncertainty in terms of continuous probability distributions, which have an infinite number of outcomes rather than just a few discrete values. Thus, Monte Carlo simulation provides a more realistic view of a project’s risk than does scenario analysis and can be installed on personal computers as an add-on to a spreadsheet program. Because most financial analysis today is done with spreadsheets, Monte Carlo simulation is now accessible to virtually all health services organizations, both large and small. The first step in a Monte Carlo simulation is to create a model that calculates the project’s net cash flows and profitability measures, just as was done for Ridgeland’s MRI project. The relatively certain variables are estimated as single, or point, values in the model, while continuous probability distributions are used to specify the uncertain cash flow variables. After the model has been created, the simulation software automatically executes the following steps: EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use 471 472 G a p en s k i’s U n d e r s ta n d i n g H e a l th c a re F inanc ial Managem ent 1. The Monte Carlo program chooses a single random value for each uncertain variable on the basis of its specified probability distribution. 2. The values selected for each uncertain variable, along with the point values for the relatively certain variables, are combined in the model to estimate the net cash flow for each year. 3. Using the net cash flow data, the model calculates the project’s profitability—for example, as measured by NPV. A single completion of these three steps constitutes one iteration, or run, in the Monte Carlo simulation. 4. The Monte Carlo software repeats these steps many times (e.g., 5,000). Because each run is based on different input values, each run produces a different NPV. The ultimate result of the simulation is an NPV probability distribution based on a large number of individual scenarios, which encompasses almost all of the likely financial outcomes. Monte Carlo software usually displays the results of the simulation in both tabular and graphical forms and automatically calculates summary statistical data such as expected value, standard deviation, and skewness.4 For an illustration of Monte Carlo simulation, again consider Ridgeland’s MRI project. As in the scenario analysis, the illustration has been simplified by specifying the distributions for only two key variables: (1) weekly volume and (2) salvage value. Weekly volume is not expected to vary by more than ±10 scans from its expected value of 40 scans. Because this situation is symmetrical, the normal (bell-shaped) distribution can be used to represent the uncertainty inherent in volume. In a normal distribution, the expected value ±3 standard deviations will encompass almost the entire distribution. Thus, a normal distribution with an expected value of 40 scans and a standard deviation of 10 ÷ 3 = 3.33 scans is a reasonable description of the uncertainty inherent in weekly volume. A triangular distribution was chosen for salvage value because it specifically fixes the upper and lower bounds, whereas the tails of a normal distribution are, in theory, limitless. The triangular distribution is also used extensively when the input distribution is nonsymmetrical because it can easily accommodate skewness. Salvage value uncertainty was specified by a triangular distribution with a lower limit of $500,000, a most likely value of $750,000, and an upper limit of $1 million. The basic MRI model containing these two continuous distributions was used, plus a Monte Carlo add-on to the spreadsheet program, to conduct a simulation with 5,000 iterations. The output is summarized in exhibit 12.5, and the resulting probability distribution of NPV is plotted in exhibit EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use C hap ter 12: Proj ec t Risk A naly sis 473 12.6. The mean, or expected, NPV ($82,498) is about the same as the base case NPV and expected NPV indicated in the scenario analysis ($82,493). In theory, all three results should be the same because the expected values for all input variables are the same in the three analyses. However, some randomness exists in the Monte Carlo simulation that leads to an expected NPV that is slightly different from the others. The more iterations that are run, the more likely the Monte Carlo NPV will be the same as the base case NPV, assuming that the assumptions are consistent. The standard deviation of NPV is lower in the simulation analysis because the NPV distribution in the simulation contains values within the Expected NPV $ 82,498 Minimum NPV ($ 951,760) Maximum NPV $ 970,191 Probability of a positive NPV Standard deviation Skewness EXHIBIT 12.5 Simulation Results Summary 62.8% $256,212 0.002 EXHIBIT 12.6 NPV Probability Distribution EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use 474 G a p en s k i’s U n d e r s ta n d i n g H e a l th c a re F inanc ial Managem ent entire range of possible outcomes, while the NPV distribution in the scenario analysis contains only the most likely value and the best-case and worst-case extremes. In this illustration, one value for volume uncertainty was specified for all five years; that is, the value chosen by the Monte Carlo software for volume in year 1—for example, 40 scans—was used as the volume input for the remaining four years in that iteration of the simulation analysis. As an alternative, the normal distribution for year 1 can be applied to each year separately, which would allow the volume forecasts to vary from year to year. Then, the Monte Carlo software might choose 35 as the value for year 1, 43 as the year 2 input, 32 for year 3, and so on. This approach, however, probably does not do a good job of describing real-world behavior; high usage in the first year presumably means strong acceptance of the MRI system and hence high usage in the remaining years. Similarly, low usage in the first year probably portends low usage in future years. The volume and salvage value variables were treated as independent in the simulation; that is, the value chosen by the Monte Carlo software from the salvage value distribution was not related to the value chosen from the volume distribution. Thus, in any run, a low volume can be coupled with a high salvage value and vice versa. If Ridgeland’s managers believe that high utilization at the hospital indicates a strong national demand for MRI systems, they can specify a positive correlation between these variables. A positive correlation would tend to increase the riskiness of the project because a low-volume pick in one iteration cannot be offset by a high–salvage value pick. Conversely, if the salvage value is more a function of the technological advances that occur over the next five years than local utilization, it may be best to specify the variables as independent, as was done. As in scenario analysis, the project’s simulation results must be compared with a similar analysis of the firm’s average project. If Ridgeland’s average project were considered to have less stand-alone risk when a Monte Carlo simulation was conducted, the MRI project would be judged to have above-average (high) stand-alone risk. Monte Carlo simulation has two primary advantages over scenario analysis: (1) All possible input variable values are considered, and (2) correlations among the uncertain inputs can be incorporated into the analysis. However, there is a downside to these two advantages: Although it is mechanically easy to input the probability distributions for the uncertain variables as well as their correlations into a Monte Carlo simulation, it is much more difficult to determine what those distributions and correlations are. The problem is that the more information a risk-analysis technique requires, the harder it is to develop the data with any confidence; hence, managers are left with an elegant result of questionable value. EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use C hap ter 12: Proj ec t Risk A naly sis 1. Briefly, what is Monte Carlo simulation? 2. What type of risk does it attempt to measure? 3. What are its strengths and weaknesses? SELF-TEST QUESTIONS Qualitative Risk Assessment In some situations, it may be difficult to conduct a quantitative risk assessment because the input variable estimates are nebulous. In other situations, a quantitative assessment may be possible, but a verification of results provides managers with additional confidence. More and more healthcare organizations are using qualitative risk assessment techniques to confirm quantitative assessment results or as the sole basis for the risk assessment. Qualitative risk assessment is based on the answers to a set of questions. For example, one large healthcare clinic uses these questions: • Does the project require additional market share or represent a new service initiative? • Is the project outside the scope of current management expertise? • Does the project require difficult-to-recruit physicians, nurses, or technical specialists? • Will the project pit the organization against a strong competitor? • Does the project involve new, unproven technology? Each “yes” answer is assigned one point (while each “no” answer receives zero points). If the total point count for the project is zero, it is judged to have low risk; one or two points indicate moderate risk, and three or more points indicate high risk. Although such a subjective approach appears to have little theoretical basis, a closer examination reveals that each question in the list seen earlier is tied to cash flow uncertainty. The greater the number of “yes” answers, the greater the cash flow uncertainty and hence the greater the stand-alone risk of the project. The value of using the qualitative risk assessment approach in conjunction with a quantitative risk assessment is that it forces managers to think about project risk in alternative frameworks. If the quantitative and qualitative assessments do not agree, the project’s risk assessment requires more consideration. After some discussion, Ridgeland’s managers concluded that the MRI project’s qualitative risk assessment score was 3. Thus, the quantitative EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use 475 476 G a p en s k i’s U n d e r s ta n d i n g H e a l th c a re F inanc ial Managem ent and qualitative assessments reached the same conclusion: The project has high risk. SELF-TEST QUESTIONS 1. Describe qualitative risk assessment. 2. Why does a qualitative risk assessment work? 3. Assume a quantitative risk assessment has been conducted on a project. Is a qualitative risk assessment necessary? Incorporating Risk into the Decision Process Thus far, the MRI illustration has demonstrated that a project’s riskiness is difficult to quantify. It may be possible to reach a general conclusion that one project is more or less risky than another or to compare the riskiness of a project with the business as a whole, but it is difficult to develop a good measure of project risk. This lack of precision in meaHow Many Scenarios in a suring project risk adds to the difficulties Scenario Analysis? involved in incorporating differential risk into the capital budgeting decision. In the scenario analysis of Ridgeland’s MRI project, we used three scenarios. However, three is There are two methods for incorno magic number, given that the more scenarios porating project risk into the capital budused, the more information is obtained from the geting decision process: (1) the certainty analysis. Furthermore, more scenarios lessen equivalent method, which adjusts a projthe problem associated with extreme values ect’s expected cash flows to reflect project because the best- and worst-case scenarios can risk, and (2) the risk-adjusted discount rate be assigned low probabilities (which are probably realistic) without causing the risk inherent in the method, which deals with differential risk project to be understated. by changing the cost of capital. Although Although more scenarios add additional realmost businesses use the risk-adjusted disism and provide more information for decision count rate method, there are some theomakers, a greater number of scenarios increases retical advantages to using the certainty forecasting difficulty and makes the analysis more equivalent method. Furthermore, it raises time-consuming. Furthermore, the greater the number of scenarios, the more difficult it is to interpret some interesting issues related to the riskthe results. Thus, the entire process is easier if adjustment process. three scenarios are used rather than, say, nine. What do you think? Are three scenarios sufficient or should more be used? How many scenarios are too many? Is it better to have an odd number than an even number of scenarios? Is there an optimal number of scenarios? Certainty Equivalent Method The certainty equivalent (CE) method directly follows the economic concept of utility.5 Under the CE approach, managers must first evaluate a cash flow’s risk EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use C hap ter 12: Proj ec t Risk A naly sis and then specify how much money, with certainty, would be required for an individual to be indifferent between the riskless (certain) sum and the risky cash flow’s expected value. For example, suppose that a rich eccentric offered someone the following choices: • Flip a coin. If it is a head, the individual receives $1 million; if it is a tail, the individual receives nothing. The expected value of the gamble is (0.5 × $1,000,000) + (0.5 × $0) = $500,000, but the actual outcome will be either zero or $1 million, so the gamble is highly risky. • Do not flip the coin. Simply pocket $400,000 in cash. If the individual is indifferent to the two alternatives, $400,000 is defined to be her CE amount for this particular risky expected $500,000 cash flow. The riskless $400,000 provides that individual with the same satisfaction (utility) as the risky $500,000 expected return. In general, investors are risk averse, so the CE amount for this gamble will be something less than the $500,000 expected value. Each individual would have his own CE value—the greater the individual’s degree of risk aversion, the lower the CE amount. The CE concept can be applied to capital budgeting decisions, at least in theory, in this way: • Convert each net cash flow of a project to its CE value. Here, the riskiness of each cash flow is assessed, and a CE cash flow is chosen on the basis of that risk. The greater the risk, the greater the difference between the expected value and its lower CE value. (If a cash outflow is being adjusted, the CE value is higher than the expected value. The unique risk adjustments required on cash outflows will be discussed in a later section.) • Once each cash flow is expressed as a CE, discount the project’s CE cash flow stream by the risk-free rate to obtain the project’s differential risk adjusted NPV.6 Here, the term “differential risk-adjusted” implies that the unique riskiness of the project, as compared to the overall riskiness of the business, has been incorporated into the decision process. The risk-free rate is used as the discount rate because CE cash flows are analogous to risk-free cash flows. • A positive differential risk-adjusted NPV indicates that the project is profitable even after adjusting for differential (project-specific) risk. The CE method is simple and neat. Furthermore, it can easily handle differential risk among the individual cash flows. For example, the final year’s CE cash flow might be adjusted downward an additional amount to account EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use 477 478 G a p en s k i’s U n d e r s ta n d i n g H e a l th c a re F inanc ial Managem ent for salvage value risk if that risk is considered to be greater than the risk inherent in the operating cash flows. Unfortunately, there is no practical way to estimate a risky cash flow’s CE value. No benchmarks are available to inform the estimate, so each individual would have her own estimate, and they can vary significantly. Also, the risk assessment techniques—for example, scenario analysis—focus on profitability and hence measure the stand-alone risk of a project in its entirety. This process provides no information about the riskiness of individual cash flows, so there is no basis for adjusting each cash flow to reflect its own unique risk. Risk-Adjusted Discount Rate Method In the risk-adjusted discount rate (RADR) method, expected cash flows are used in the valuation process, and the risk adjustment is made to the discount rate (the opportunity cost of capital). All average-risk projects are discounted at the business’s corporate cost of capital, which represents the opportunity cost of capital for average-risk projects; high-risk projects are assigned a higher cost of capital; and low-risk projects are discounted at a lower cost of capital. One advantage to using the RADR method is that it has a starting benchmark: the business’s corporate cost of capital. This discount rate reflects the riskiness of the business in the aggregate, or the riskiness of the firm’s average project. Another advantage is that project risk-assessment techniques identify a project’s aggregate risk—the combined risk of all of the cash flows—and the RADR applies a single adjustment to the cost of capital rather than attempts to adjust individual cash flows. However, the disadvantage is that, typically, there is no theoretical basis for setting the size of the RADR adjustment, so the amount of adjustment remains a matter of judgment. There is one additional disadvantage to using the RADR method. RADR combines the factors that account for time value (the risk-free rate) and the adjustment for risk (the risk premium): Project cost of capital = Differential risk-adjusted discount rate = Risk-free rate + Risk premium. The CE approach, on the other hand, keeps risk adjustment and time value separate—time value in the discount rate and risk adjustment in the cash flows. By lumping together risk and time value, the RADR method compounds the risk premium over time, just as interest compounds over time. This compounding of the risk premium means that the RADR method automatically assigns more risk to cash flows that occur in the distant future, and the farther into the future, the greater the implied risk. Because the CE method assigns risk to each cash flow individually, it does not impose assumptions regarding the relationship between risk and time. The RADR model is one method used to incorporate risk in the capital budgeting decision process. It is based on the following concept: EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use C hap ter 12: Proj ec t Risk A naly sis Key Equation 12.1: Risk-Adjusted Discount Rate (RADR) Theoretical Model Project cost of capital = Risk-free rate + Risk premium. The idea here is that the risk-free rate accounts for the time value of money, while the risk premium accounts for the unique (below average, average, or above average) risk of the project. The RADR method as it is normally used—with a constant discount rate applied to all cash flows of a project—implies that risk increases with time. This implication imposes a greater burden on long-term projects, so short-term projects tend to look better financially than do long-term projects. For most projects, the assumption that risk increases over time is probably reasonable because cash flows are more difficult to forecast the farther one moves into the future. However, managers should be aware that the RADR approach automatically penalizes distant cash flows, and an additional explicit penalty based solely on cash flow timing is not warranted unless some specific additional risk can be identified. 1. What are the differences between the CE and RADR methods for risk incorporation? 2. What assumptions about time and risk are inherent in the RADR method? 3. How do most businesses incorporate differential risk into the capital-budgeting decision process? SELF-TEST QUESTIONS Final Risk Assessment and Incorporation for the MRI Project In most project risk analyses, it is impossible to assess the project’s corporate or market risk quantitatively, and managers are left with only an assessment of the project’s stand-alone risk. However, like the MRI project, most projects being evaluated are in the same line of business as the firm’s other projects, and the profitability of most firms is highly correlated with the overall economy. Thus, stand-alone, corporate, and market risk are usually highly correlated, which suggests that managers can get a feel for the relative risk of most projects on the basis of the quantitative and qualitative analyses conducted to assess the project’s stand-alone risk. In Ridgeland’s case, its managers concluded that the MRI project, with its above-average stand-alone risk, also EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use 479 480 G a p en s k i’s U n d e r s ta n d i n g H e a l th c a re F inanc ial Managem ent had above-average corporate risk, which is the risk most relevant to not-forprofit organizations; hence, the project was categorized as a high-risk project. The business’s corporate cost of capital provides a basis for estimating a project’s differential RADR—average-risk projects are discounted at the corporate cost of capital, high-risk projects are discounted at a higher cost of capital, and low-risk projects are discounted at a rate below the corporate cost of capital. Unfortunately, there is no good Uncertainty in Initial Cash Outflows way of specifying exactly how much higher In many capital budgeting situations, the initial or lower these discounts rates should be; cost of the project—especially when occurring given the present state of the art, risk only at time 0—is assumed to be known with adjustments are necessarily judgmental certainty. The idea here is that, in most cases, and somewhat arbitrary. bids have already been received from vendors, Ridgeland’s standard procedure is so the initial cost can be predicted with relative precision. However, in some circumstances, there to add 4 percentage points to its 10 can be substantial uncertainty in initial costs. For percent corporate cost of capital when example, there can be a great deal of uncertainty evaluating high-risk projects and to subin the cost of a building that will not be contract 2 percentage points when evaluating structed for several years. Or there can be uncerlow-risk projects. Thus, to estimate the tainty in the cost of a major construction project high-risk MRI project’s differential riskthat will take several years to complete. When there is uncertainty in initial cost, how adjusted NPV, the project’s expected (base should that risk be incorporated into the analycase) cash flows shown in exhibit 12.1 sis? If the entire cost (or even the major portion) are discounted at 10% + 4% = 14%. This occurs at time 0, the discount rate is not applied rate is called the project cost of capital, as to the cash flow, so the RADR method will not get opposed to the corporate cost of capital, the job done. because it reflects the risk characteristics What do you think? Can the CE method be used? Assume that time 0 costs on a project could of a specific project rather than the aggrebe $100,000 or $150,000 with equal profitability, gate risk characteristics of the business. so the expected initial cost is $125,000. What is The resultant NPV is −$200,017, so the your estimate of the CE cash flow? (Hint: Rememproject becomes unprofitable when the ber that risk adjustments to cash outflows are the analysis is adjusted to reflect its high risk. opposite of those applied to inflows.) Ridgeland’s managers may still decide to go ahead with the MRI project, but at least they know that its expected profitability is not sufficient to make up for its riskiness. The RADR method is implemented as follows: Key Equation 12.2: Risk-Adjusted Discount Rate (RADR) Implementation Model Project cost of capital = Corporate cost of capital + Risk adjustment. EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use C hap ter 12: Proj ec t Risk A naly sis Here, the corporate cost of capital is used as the base rate (starting point), and a risk adjustment is applied if the project has non-average risk. For aboveaverage risk projects, the risk premium is added to the base rate, while the risk premium is subtracted for those projects judged to have below-average risk. To illustrate, assume a project having above-average risk is being evaluated. The corporate cost of capital is 10 percent, and the standard adjustment amount is 3 percentage points. With these assumptions, the project cost of capital is 13 percent: Project cost of capital = Corporate cost of capital + Risk adjustment = 10% + 3% = 13%. 1. How did Ridgeland’s managers translate the MRI project’s standalone risk assessment into a corporate risk assessment? 2. How was risk incorporated into the MRI project decision process? 3. Is the risk adjustment objective or subjective? 4. What is a project cost of capital? SELF-TEST QUESTIONS Incorporating Debt Capacity into the Decision Process Just as different businesses have different optimal capital structures, so do individual projects. In any business, the overall optimal capital structure, which is reflected by the weights used in the corporate cost of capital estimate, is an aggregation of the optimal capital structures of the business’s individual projects. However, some projects support only a little debt, while other projects support a high level of debt. The proportion of debt in a project’s, or a business’s, optimal capital structure is called the project’s, or business’s, debt capacity. One mistake often made when considering a project’s debt capacity is to look at how the project is actually financed. For example, even though Ridgeland may be able to obtain a secured loan for the entire cost of the MRI equipment, the MRI project does not have a debt capacity of 100 percent. The willingness of lenders to furnish 100 percent debt capital for the MRI project is based more on Ridgeland’s overall creditworthiness than on the financial merits of the MRI project because all of the hospital’s operating cash flow, less interest payments on embedded debt, is available to pay the lender. Think of it this way: Would lenders provide 100 percent financing if Ridgeland were a start-up business and the MRI project was its sole source of income? EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use 481 482 G a p en s k i’s U n d e r s ta n d i n g H e a l th c a re F inanc ial Managem ent The logical question here is whether debt capacity differences should be taken into consideration in the capital budgeting process. In theory, if there are meaningful debt capacity differences between a project and the business, capital structure differentials—as well as risk differentials—should be taken into account in the capital budgeting process. For example, an academic health center might be evaluating two projects: one involves research and development (R&D) of a new surgical procedure, and the other involves building a primary care clinic in a local upscale residential area. The R&D project would have relatively low debt capacity because it has high business risk and no assets suitable as loan collateral. Conversely, the clinic project would have relatively high debt capacity because it has low business risk and involves real estate suitable as collateral. Incorporating capital structure differentials is mechanically easy. We merely change the weights used to compute the corporate cost of capital to reflect project debt capacity rather than use the standard weights that reflect the business’s target capital structure. Projects with higher-than-average debt capacity would use a relatively high value for the weight of debt and a relatively low value for the weight of equity and vice versa. However, a problem arises when attempting to make debt capacity adjustments. We know from chapter 10 that increased debt usage raises capital costs, so both the cost of debt and the cost of equity must increase as more debt financing is used. This dependency of capital costs on capital structure means that as the weights are changed in the cost-of-capital calculation, so should the component costs. However, it is very difficult, if not impossible, to estimate individual project costs of debt and equity that correspond to the project’s optimal capital structure. Thus, capital structure adjustments quickly become a somewhat futile guessing game, so most businesses do not make such adjustments unless there are specific benchmark values that can be used for both a project’s unique debt capacity and the corresponding capital costs.7 SELF-TEST QUESTION 1. Discuss the advantages and disadvantages of incorporating debt capacity differences into the capital budgeting decision process. Adjusting Cash Outflows for Risk Although most projects are evaluated on the basis of profitability, some are evaluated solely on the basis of costs. Such evaluations are done when it is impossible to allocate revenues to a particular project or when two competing projects will produce the same revenue stream. For example, suppose that EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use C hap ter 12: Proj ec t Risk A naly sis 483 Ridgeland must choose one of two ways of disposing of its medical waste. There is no question about the necessity of the project, and neither method will affect the hospital’s revenue stream. In this situation, the decision will be based on the present value of expected future costs; the method with the lower present value of costs will be chosen. Exhibit 12.7 lists the projected annual costs associated with each method. A large expenditure would be required at year 0 to upgrade the hospital’s current disposal system, but the yearly operating costs would be relatively low. Conversely, if Ridgeland contracts for disposal services with an outside contractor, it will have to pay only $25,000 up front to initiate the contract. However, the annual contract fee would be $200,000 a year. Note that inflation effects are ignored in this illustration to simplify the discussion. If both methods were judged to have average risk, Ridgeland’s corporate cost of capital—10 percent—would be applied to the cash flows to obtain the present value (PV) of costs for each method. Because the PVs of costs for the two waste disposal systems—$784,309 for the in-house system and $783,157 for the contract method—are roughly equal at a 10 percent discount rate, Ridgeland’s managers would be indifferent as to which method should be chosen if they were basing the decision on financial considerations only. However, Ridgeland’s managers actually believe that the contract method is much riskier than the in-house method. They know the cost of modifying the current system to the dollar, and they can predict operating costs fairly well. Furthermore, the in-house system’s operating costs are under the control of Ridgeland’s management. Conversely, if the hospital relies on Cash Flows Year In-House System Outside Contract 0 1 2 3 4 5 ($500,000) (75,000) (75,000) (75,000) (75,000) (75,000) ($ 25,000) (200,000) (200,000) (200,000) (200,000) (200,000) Present value of costs at a discount rate of: 10% ($784,309) 14% –   6% –   ($ 783,157) ($ 711,616) ($ 867,473) EXHIBIT 12.7 Ridgeland Community Hospital: Waste Disposal Analysis EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use 484 G a p en s k i’s U n d e r s ta n d i n g H e a l th c a re F inanc ial Managem ent the contractor for waste disposal, it more or less will have to continue the contract because it will lose in-house capability. Because the contractor was willing to guarantee the price only for the first year, perhaps the bid was lowballed and large price increases will occur in future years. The two methods have about the same PV of costs when both are considered to have average risk—so which method should be chosen if the contract method is judged to have high risk? Clearly, if the costs are the same under a common discount rate, the lower-risk in-house project should be chosen. Now, try to incorporate this intuitive differential risk conclusion into the quantitative analysis. Conventional wisdom is to increase the corporate cost of capital for high-risk projects, so the contract cash flows would be discounted using a project cost of capital of 14 percent, which is the rate that Ridgeland applies to high-risk projects. However, at a 14 percent discount rate, the contract method has a PV of costs of only $711,616, which is about $70,000 lower than that for the in-house method. If the discount rate on the contract method’s cash flows were increased to 20 percent, an even greater amount, it would appear to be $161,000 cheaper than the inhouse method. Thus, the riskier the contract method is judged to be, the better it looks. Something is obviously wrong here! For a cash outflow to be penalized for higher-than-average risk, it must have a higher present value, not a lower one. Therefore, a cash outflow that has higher-than-average risk must be evaluated with a lower-than-average cost of capital. Recognizing this, Ridgeland’s managers applied a 10% − 4% = 6% discount rate to the highrisk contract method’s cash flows. The result is a PV of costs for the contract method of $867,473, which is about $83,000 more than the PV of costs for the average-risk in-house method. The appropriate risk adjustment for cash outflows is also applicable in other situations. For example, the city of Detroit offered Ann Arbor Health Care Inc. the opportunity to use a city-owned building in a blighted area for a walk-in clinic. The city offered to pay to refurbish the building, and all profits made by the clinic would accrue to Ann Arbor. However, after ten years, Ann Arbor would have to buy the building from the city at the then-current market value. The market value estimate that Ann Arbor used in its analysis was $2 million, but the realized cost could be much greater, or much less, depending on the economic condition of the neighborhood at that time. The project’s other cash flows were of average risk, but this single outflow was high risk, so Ann Arbor lowered the discount rate that it applied to this one cash flow. This action created a higher present value for the $2 million cost (outflow) and hence lowered the project’s NPV. EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use C hap ter 12: Proj ec t Risk A naly sis The bottom line here is that risk adjustment for cash outflows is the opposite of adjustment for cash inflows. When cash outflows are being evaluated, higher risk calls for a lower discount rate.8 1. Why are some projects evaluated on the basis of present value of costs? 2. Is there any difference between the risk adjustments applied to cash inflows and cash outflows? Explain your answer. 3. Can differential risk adjustments be made to single cash flows, or must the same adjustment be made to all of a project’s cash flows? SELF-TEST QUESTIONS Real (Managerial) Options According to traditional capital budgeting analysis techniques, a project’s NPV is the present value of its expected future cash flows when discounted at an opportunity cost rate that reflects the riskiness of those flows. However, as discussed in chapter 11 in the section on strategic value, such valuations generally do not incorporate the value inherent in additional actions that the business can take only if the project is accepted. In other words, traditional capital budgeting can be likened to playing roulette: A bet is made (the project is accepted) and the wheel is spun, but nothing can be done to influence the outcome of the game. In reality, capital projects are more like draw poker: Chance does play a role, but the players can influence the final result by discarding the right cards and assessing the other players’ actions. The opportunities that managers have to change a project in response to changing conditions or to build on a project are called real, or managerial, options. These terms denote that such options arise from investments in real, rather than financial, assets and that the options are available to managers of businesses as opposed to individual investors. To illustrate the concept of real options, we introduce decision tree analysis. Although risk analysis is an integral part of capital budgeting, managers are at least as concerned (or maybe more concerned) about managing risk than they are about measuring it. One way of managing risk is to structure large projects as a series of decision points that provide the opportunity to reevaluate decisions as additional information becomes available, and possibly to cancel—or once it begins, to abandon—the project if events take a turn for the worse. EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use 485 486 G a p en s k i’s U n d e r s ta n d i n g H e a l th c a re F inanc ial Managem ent Projects that are structured as a series of decision points over time are evaluated using decision trees. For example, suppose Medical Equipment International (MEI) is considering the production of a new and innovative intensive care monitoring system. The net investment for this project is broken down into three stages, as set forth in exhibit 12.8. If the go-ahead is given for stage 1 (year 0), the firm will conduct a $500,000 study of the market potential for the new monitoring system, which will take about one year. If the results of the study are unfavorable, the project will be canceled, but if the results are favorable, MEI will (at year 1) spend $1 million to design and fabricate several prototype systems. These systems will then be tested at two hospitals, and MEI will base its decision to proceed with full-scale production on their medical staffs’ reactions to them. If their reactions are positive, MEI will establish a production line for the monitoring systems at one of its plants at a net cost of $10 million. If this stage is reached, MEI’s managers estimate that the project will generate net cash flows over the following four years that will depend on the vitality of the hospital sector at that time and the overall performance of the system. A decision tree such as the one in exhibit 12.8 often is used to analyze such multistage, or sequential, decisions. Here, for simplicity, let’s assume that one year goes by between decisions. Each circle represents a decision point or stage. The dollar value to the left of each decision point represents the net investment required to go forward at that decision point, and the cash flows under the t = 3 to t = 6 headings represent the cash inflows that would occur if the project is carried to completion. Each diagonal line represents the beginning of a branch of the decision tree, and each carries a probability that MEI’s managers estimate on the basis of the information available to them today. For example, management estimates that there is a probability of 0.8 that the initial study will produce favorable results, which would lead to the expenditure of $1 million at stage 2, and a 0.2 probability that the initial study will produce unfavorable results, which would lead to cancellation after stage 1. The joint probabilities shown in exhibit 12.8 give the probability of occurrence of each final outcome—that is, the probability of moving completely along each branch. Each joint probability is obtained by multiplying together all the probabilities along a particular branch. For example, if stage 1 is undertaken, the probability that MEI will move through stages 2 and 3 and that a strong demand will produce $10 million in net cash flows in each of the next four years is 0.8 × 0.6 × 0.3 = 0.144 = 14.4%. The NPV of each final outcome is also given in exhibit 12.8. MEI has a corporate cost of capital of 11.5 percent, and its management assumes initially that all projects have average risk. For example, the NPV of the top branch (the most favorable outcome) is about $15,250 (in thousands of dollars): EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use ($500) t=0 0.8 1 Stop ($1,000) t=1 2 0.6 Stop 3 0.3 ($10,000) t=2 0.144 $ 2,000 $ 2,000 ($ 2,000) ($ 2,000) 0.200 1.000 0.320 0.192 $ 4,000 0.144 (1,397) (14,379) 436 $15,250 NPV (447) (2,701) 84 $ 2,196 Product: Prob. × NPV (500) (100) Expected NPV = ($ 338) σNPV= $7,991 Joint Probability 0.4$ 4,000 $ 4,000 $ 4,000 t=6 $10,000 $10,000 $10,000 t=5 $10,000 t=4 t=3 Time EXHIBIT 12.8 Decision Tree Analysis (in thousands of dollars) C hap ter 12: Proj ec t Risk A naly sis 0.3 0.4 0.2 EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use 487 488 G a p en s k i’s U n d e r s ta n d i n g H e a l th c a re F inanc ial Managem ent $1, 000 $10, 000 $10, 000 $10, 000 − + + (1.115)1 (1.115)2 (1.115)3 (1.115)4 $10, 000 $10, 000 + + (1.115)5 (1.115)6 = $15, 250. NPV = −$500 − Other NPVs are calculated similarly. The last column in exhibit 12.8 indicates the product of the NPV for each branch and the joint probability that that branch will occur; the sum of the NPV products is the expected NPV of the project. Considering the expectations set forth in exhibit 12.8, and assuming a cost of capital of 11.5 percent, we determine that the monitoring equipment project’s expected NPV is −$338,000. Because the expected NPV is negative, it appears that this project would be unprofitable and hence should be rejected by MEI unless other considerations prevail. However, this initial judgment may not be correct. MEI must now consider whether this project is more, less, or about as risky as the firm’s average project. The expected NPV is a negative $338,000, and the standard deviation of NPV is $7,991,000, so the coefficient of variation of NPV is $7,991,000 ÷ $338,000 = 23.6, which is quite large. (Note that the negative sign for NPV does not enter into the calculation.) The value for the coefficient of variation suggests that the project is highly risky in terms of stand-alone risk. Note also that there is a 0.144 + 0.320 + 0.200 = 0.664 = 66.4% probability of incurring a loss. On the basis of these findings, the project appears to be unacceptable financially unless it has some embedded real options that will increase its value or reduce its risk. The Real Option of Abandonment On the web at: ache.org/HAP/ PinkSong8e Abandonment, which is discussed in chapter 11 in connection with estimating a project’s economic life, is one type of real option that many projects possess. For an illustration of this real option’s impact, suppose that MEI is not contractually bound to continue the project once production has begun. Thus, if sales are poor during year 3 (t = 3), if MEI experiences a cash flow loss of $2 million, and if similar results are expected for the remaining three years, MEI can abandon the project at the end of year 3 rather than continue to suffer losses. In this situation, low first-year sales signify that the monitoring equipment is not selling well, so future sales will also be poor, and MEI can act on this new information when it becomes available. EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use C hap ter 12: Proj ec t Risk A naly sis MEI’s ability to abandon the project changes the branch of the decision tree that contains the series of $2 million losses in exhibit 12.8. It now appears as follows (in thousands of dollars): 3 0.3 ($2,000) 4 Stop Joint Probability NPV Product: Prob. × NPV 0.144 ($10,883) ($1,567) Changing this branch to reflect abandonment eliminates the $2 million cash losses in years 4, 5, and 6 and thus causes the NPV for the branch to be higher, although still negative. This change increases the project’s expected NPV from −$338,000 to about $166,000 and lowers the project’s standard deviation from $7,991,000 to $7,157,000. Thus, the abandonment real option changes the project’s expected NPV from negative to positive and also lowers its stand-alone risk as measured either by standard deviation or by coefficient of variation of NPV. We can use the data just developed to estimate the value of the abandonment option. The NPV with the abandonment option is $166,000, while the NPV without this option is −$338,000, so the value of the real option is $166,000 − (−$338,000) = $504,000. However, this value understates the true value of the option because the ability to abandon the project also lowers the riskiness of the project. With lower risk, the difference between the two NPVs is greater than that calculated, although the added value of risk reduction would be relatively small in this illustration as well as difficult to quantify with confidence. Because of this and similar complications, discounted cash flow techniques (when they can be used to value real options) generally will not produce an accurate estimate of the option’s value. Here are some additional points to note concerning decision tree analysis and abandonment: • Managers can reduce project risk if they can structure the decision process to include several decision points rather than just one. If MEI were to make a total commitment to the monitoring equipment project at t = 0 and sign contracts that would require completion of the project, it might save some money and accelerate the project, but doing so would substantially increase the project’s riskiness. • Once production or service begins, a business’s ability to abandon a project can dramatically reduce the project’s risk. • The cost of abandonment generally is reduced if the firm has alternative uses for the project’s assets. If MEI can convert the EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use 489 490 G a p en s k i’s U n d e r s ta n d i n g H e a l th c a re F inanc ial Managem ent abandoned monitoring equipment production line to a different, more productive use, the cost of abandonment would be reduced and the monitoring equipment project would become more attractive. Finally, note that capital budgeting is a dynamic process. Virtually all inputs to a capital budgeting decision change over time, and firms must periodically review both their expenditure plans and their ongoing projects. In the MEI example, conditions might change between decision points 1 and 2; if they do, this new information should be used to revise the probability and cash flow estimates. If a capital budgeting decision can be structured with multiple decision points, including abandonment, and if the firm’s managers have the fortitude to admit when a project is not working out as initially planned, risks can be reduced and expected profitability can be increased. Other Real Options The MEI monitoring system project demonstrates that the real option of abandonment can add value to a project. In addition to abandonment, there are many other types of real options. Flexibility Options The flexibility option allows managers to switch inputs between alternative production or service processes. For example, by training clinical personnel to perform multiple tasks, individuals hired for a new service can potentially be used productively in other parts of the business. Thus, labor costs associated with the new service can be easily reduced if demand estimates are not met. This flexibility option reduces costs in poor utilization scenarios and hence increases the value of the project. Capacity Options The capacity option allows businesses to manage their productive capacity in response to changing market conditions. If a project can be structured so that its operations can be reduced or suspended if warranted rather than completely shut down, the value of the project increases. The option to expand new services from a relatively small scale to a large scale also adds value. New Service Options It is easy to envision a situation in which a negative NPV project is accepted because embedded in it is an option to add complementary services or successive “generations of services.” A managed care organization’s first move into a new geographic area and the introduction of transplant services at a hospital are two examples. In such situations, the first project may not be profitable, but it can lead to additional opportunities that are. EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use C hap ter 12: Proj ec t Risk A naly sis Timing Options In our examples thus far, new projects brought with them embedded real options that could be exercised in the future and hence added value to the project. Timing options can be somewhat different, in that in some circumstances they involve extinguishing existing real options. Timing options were first analyzed in situations involving natural resources, such as when to harvest a forested area or how much oil to pump out of a well. By harvesting or pumping now, the project can produce immediate cash flows, but doing so eliminates the opportunity to obtain future cash flows from the same resource. Of most interest to healthcare businesses is the option to delay, which is another type of timing option. If a project can be postponed, it might be more valuable in the future because, for example, managed care power is diminishing, technology is advancing, or information that will decrease the project’s risk is expected to become available. Of course, the option to delay is valuable only if it is worth more than the costs of delaying, which include time value of money costs, costs associated with competitor actions, and patient satisfaction costs. Thus, in general, the option to delay is most valuable to businesses that have proprietary technology or some other barrier to entry that lessens the costs associated with postponement. 491 On the web at: ache.org/HAP/ PinkSong8e Valuation of Projects That Have Real Options In general, the true value of a project with real options can be thought of as the discounted cash flow (DCF) NPV plus the value of the real options: True NPV = DCF NPV + Value of real options. In most healthcare situations, a dollar value cannot be placed on any real options associated with a project. However, managers should still think about the value of many projects in terms of this equation. Here are some points to consider: • Real options can add considerable value to many projects, so failure to consider such options leads to downward-biased NPVs and thus to systematic underinvestment. • In general, the longer a real option lasts before it must be “exercised,” the more valuable it is. For example, suppose the real option is to expand into related services, such as expanding rehabilitative services into sports medicine services. The longer the expansion can be delayed and still retain its value, the more valuable the option. • The more volatile the value of the underlying source of the real option, the more valuable the option. Thus, the more return volatility there is EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use 492 G a p en s k i’s U n d e r s ta n d i n g H e a l th c a re F inanc ial Managem ent in the return on sports medicine services, the greater the value of a real option to expand into such services. • The higher the cost of capital (the higher the general level of interest rates), the more valuable the real option. This point is not intuitive, but we explain the rationale in chapter 18 (available online) in our discussion of stock options. SELF-TEST QUESTIONS 1. How can the possibility of abandonment affect a project’s profitability and stand-alone risk? 2. What are the costs and benefits of structuring large capital budgeting decisions in stages rather than in a single decision? 3. Why might DCF valuation underestimate the true value of a project? 4. What are some different types of real options? 5. How does the presence of real options influence capital budgeting decisions? An Overview of the Capital Budgeting Decision Process The discussion of capital budgeting thus far has focused on how managers evaluate individual projects. For capital planning purposes, healthcare managers also need to forecast the total number of projects that will be undertaken and the dollar amount of capital needed to fund these projects. The list of projects to be undertaken is called the capital budget, and the optimal selection of new projects is called the optimal capital budget. While every healthcare provider estimates its optimal capital budget in its own way, some procedures are common to all businesses. We use the procedures followed by CALFIRST Health System to illustrate the process: • The chief financial officer (CFO) estimates the system’s corporate cost of capital. As discussed in chapter 9, this estimate depends on market conditions, the business risk of CALFIRST’s assets in the aggregate, and the systemwide optimal capital structure. • The CFO then scales the corporate cost of capital up or down to reflect the unique risk and capital structure features of each division. Assume that CALFIRST has three divisions: LRD (low-risk division), ARD (average-risk division), and HRD (high-risk division). EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use C hap ter 12: Proj ec t Risk A naly sis • Managers in each of the divisions evaluate the riskiness of the proposed projects to their divisions by categorizing each project as LRP (low-risk project), ARP (average-risk project), or HRP (high-risk project). These project risk classifications are based on the riskiness of each project relative to the other projects in the division, not to the system in the aggregate. • Each project is then assigned a project cost of capital that is based on the divisional cost of capital and the project’s relative riskiness. As discussed previously, this project cost of capital is then used to discount the project’s expected net cash flows. From a financial standpoint, all projects with positive NPVs are acceptable, while those with negative NPVs should be rejected. Subjective factors are also considered, and these factors may prompt a decision that differs from the one established solely on the basis of financial considerations. Exhibit 12.9 summarizes CALFIRST’s overall capital budgeting process. Here, the corporate cost of capital, 10 percent, is adjusted upward to 14 percent in the HRD and downward to 8 percent in the LRD. The same adjustment—4 percentage points upward for HRPs and 2 percentage points downward for LRPs—is applied to differential risk projects in each division. The end result is a range of project costs of capital in CALFIRST that runs from 18 percent for HRPs in the HRD to 6 percent for LRPs in the LRD. The result is a financial analysis process that incorporates each project’s debt capacity, at least at the divisional level, and riskiness. However, managers also must consider other possible risk factors that may not have been included in the quantitative analysis. For example, could the project being evaluated significantly increase the business’s liability exposure? Conversely, does the project have any real option value, social value, or other attributes that could affect its profitability or riskiness? Such additional factors must be considered, at least subjectively, before a final decision can be made. (A framework for considering multiple decision factors—the project scoring approach—is discussed in chapter 11.) Typically, if the project involves new products or services and is large (in capital requirements) relative to the size of the business’s average project, the additional subjective factors will be important to the final decision; one large mistake can bankrupt a firm, so “bet-the-firm” decisions are not made lightly. On the other hand, a decision on a small replacement project would be made mostly on the basis of numerical analysis. Ultimately, capital budgeting decisions require an analysis of a mix of objective and subjective factors such as risk, debt capacity, profitability, medical staff (patient) needs, real option value, and social value. The process is not precise, and often there is a temptation to ignore one or more important factors because they are so nebulous and difficult to measure. Despite this EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use 493 494 G a p en s k i’s U n d e r s ta n d i n g H e a l th c a re F inanc ial Managem ent EXHIBIT 12.9 CALFIRST: Divisional and Project Costs of Capital High-risk project HRD cost of capital = 14% Average-risk project Low-risk project High-risk project Corporate cost of capital = 10% ARD cost of capital = 10% Average-risk project Low-risk project High-risk project LRD cost of capital = 8% Average-risk project Low-risk project 18% 14% 12% 14% 10% 8% 12% 8% 6% imprecision and subjectivity, a project’s risk, as well as its other attributes, should be assessed and incorporated into the capital budgeting decision process. SELF-TEST QUESTIONS 1. Describe a typical capital budgeting decision process. 2. Are decisions made solely on the basis of quantitative factors? Explain your answer. Capital Rationing Standard capital budgeting procedures assume that businesses can raise virtually unlimited amounts of capital to meet capital budgeting needs. Presumably, as long as a business is investing the funds in profitable (i.e., positive NPV) projects, it should be able to raise the debt and equity needed to fund all such projects. In addition, standard capital budgeting procedures assume that a business raises the capital needed to finance its optimal capital budget EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use C hap ter 12: Proj ec t Risk A naly sis roughly in accordance with its target capital structure and at an average cost equal to the estimated corporate cost of capital. This picture of a business’s capital financing and capital investment process is probably appropriate for large investor-owned firms in most situations. However, not-for-profit firms and small investor-owned businesses typically do not have unlimited access to capital. Their ability to raise equity capital often is limited, and their debt capital is constrained to the amount supported by the equity capital base. Thus, such businesses will likely face periods in which the capital needed for investment in worthwhile new projects will exceed the amount of capital available. This situation is called capital rationing. If capital rationing exists (i.e., a business has more acceptable projects than capital), from a financial perspective the business should accept the set of capital projects that maximizes aggregate NPV and still meets the capital constraint. This approach can be called “getting the most bang for the buck” because it picks projects that have the most positive impact on the business’s financial condition. Another ROI measure—the profitability index (PI)—is useful in a capital rationing situation. The PI is defined as the PV of cash inflows divided by the PV of cash outflows. Thus, for Ridgeland’s MRI project discussed earlier in the chapter, PI = $2,582,493 ÷ $2,500,000 = 1.03. The PI measures a project’s dollars of profitability per dollar of investment, all on a PV basis. The MRI project promises three cents of profit for every dollar invested, which indicates it is not very profitable. (The PI of 1.03 is before adjusting for risk. After adjusting for risk, the project’s PI is less than 1.00, indicating that the project is unprofitable.) In a capital rationing situation, the optimal capital budget is determined by first listing all profitable projects in descending order of PI. Then, projects are selected from the top of the list downward until the capital available is used up. Of course, in healthcare businesses, priority may be assigned to some low or even negative NPV projects, which is fine as long as these projects are offset by the selection of profitable projects, which would prevent the lowprofitability priority projects from eroding the business’s financial condition. 1. What is capital rationing? 2. From a financial perspective, how are projects chosen when capital rationing exists? 3. What is the profitability index, and why is it useful in a capital rationing situation? SELF-TEST QUESTIONS EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use 495 496 G a p en s k i’s U n d e r s ta n d i n g H e a l th c a re F inanc ial Managem ent Chapter Key Concepts This chapter discussed project risk definition, assessment, and incorporation. Here are its key concepts: • There are three types of project risk: (1) stand-alone risk, (2) corporate risk, and (3) market risk. • A project’s stand-alone risk is the risk the project would have if it were the sole project of a not-for-profit firm. It is measured by the variability of profitability, generally by the standard deviation or coefficient of variation of NPV. Stand-alone risk often is used as a proxy for corporate and market risk because (1) corporate and market risk are often impossible to measure and (2) the three types of risk are usually highly correlated. • Corporate risk reflects a project’s contribution to the overall riskiness of the business. Corporate risk ignores stockholder diversification and is relevant to not-for-profit firms. • Market risk reflects the contribution of a project to the overall riskiness of the owners’ well-diversified investment portfolios. In theory, market risk is relevant to investor-owned firms, but many people argue that corporate risk is also relevant to owners, especially the owners and managers of small businesses, and it is certainly relevant to a business’s other stakeholders. • Three quantitative techniques are commonly used to assess a project’s stand-alone risk: (1) sensitivity analysis, (2) scenario analysis, and (3) Monte Carlo simulation. • Sensitivity analysis shows how much a project’s profitability—for example, as measured by NPV—changes in response to a given change in an input variable such as volume, other things held constant. • Scenario analysis defines a project’s best, most likely, and worst possible outcomes and then uses these data to measure its standalone risk. • Whereas scenario analysis focuses on only a few possible outcomes, Monte Carlo simulation uses continuous distributions to reflect the uncertainty inherent in a project’s component cash flows. The result is a probability distribution of NPV, or IRR, that provides a great deal of information about the project’s riskiness. (continued) EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use C hap ter 12: Proj ec t Risk A naly sis (continued from previous page) • In addition to quantitative risk assessment techniques, the qualitative approach uses the answers to yes-or-no questions to assess project risk. • Projects that require capital outlays in stages over time often are evaluated using decision trees. The branches of the tree represent different outcomes, and, when subjective probabilities are assigned to the outcomes, the tree provides the profitability distribution for the project. • In addition to the DCF-calculated NPV, some projects have additional value in the form of embedded real (managerial) options. • One type of real option is the ability to abandon a project once operations have begun. This option can both increase a project’s dollar return and decrease its riskiness and thus has a twofold positive effect on value. • There are two methods for incorporating project risk into the capital budgeting decision process: (1) the certainty equivalent (CE) method, which adjusts a project’s expected cash flows to reflect project risk, and (2) the risk-adjusted discount rate (RADR) method, which deals with differential risk by changing the cost of capital. • Projects are generally classified as high risk, average risk, or low risk on the basis of their stand-alone risk assessment. High-risk projects are evaluated at a discount rate greater than the firm’s corporate cost of capital, average-risk projects are evaluated at the corporate cost of capital, and low-risk projects are evaluated at a rate less than the corporate cost of capital. In a business with divisions, the risk-adjustment process often takes place at the divisional level. • In the evaluation of risky cash outflows, the risk adjustment process is reversed—that is, lower rates are used to discount more risky cash flows. • Ultimately, capital budgeting decisions require an analysis of a mix of objective and subjective factors such as risk, debt capacity, profitability, medical staff needs, real option value, and social value. The process is not precise, but good managers do their best to ensure that none of the relevant factors are ignored. (continued) EBSCOhost – printed on 11/14/2023 6:45 PM via SOUTHERN NEW HAMPSHIRE UNIVERSITY. All use subject to https://www.ebsco.com/terms-of-use 497 498 G a p en s k i’s U n d e r s ta n d i n g H e a l th c a re F inanc ial Managem ent (continued from previous page) • In a capital rationing situation, the business has more profitable projects than investment capital. In such cases, the profitability index (PI) is a useful measure of profitability (ROI). This concludes our discussion of capital budgeting. In chapters 13 and 14, we discuss financial and operating analyses and financial forecasting. Chapter Models, Problems, and Minicases The following ancillary resources in spreadsheet format ar…
Purchase answer to see full attachment