UltraLight Aircraft, Inc. A Capital Budgeting Decision "So, what do you think, Cody? Isn’t she a beaut?" Cheryl Morrison, founder and majority stockholder of UltraLight Aircraft, Inc., wants a reaction from her production manager, Cody Hill, regarding a prototype of the company’s latest aircraft, the SilkStream. Ultralight airplanes are typically constructed from aluminum alloy tubes, chromalloy steel, and stainless steel rivets. These items are assembled to form a skeleton to which a wing and an engine are attached. The single-seat planes weigh less than 250 pounds and are designed for one person. The maximum fuel capacity is 5 gallons and the planes have a top speed of 64 miles per hour. The Federal Aviation Administration (FAA) restricts flying to daylight hours and over unpopulated areas. The pilot does not have to have a license to fly and the activity is considered recreational. With the financial backing of her parents, Cheryl founded UltraLight Aircraft in 2003, two short years after flying her first ultralight aircraft. Beginning with three employees and working out of a 1920s red brick warehouse in Tupelo, Mississippi, Cheryl began the business with little experience, money, or managerial training. However, the company became profitable in its second year of operations and, through an effective web site and advertising campaign, has established a reputation as a manufacturer of safe, dependable micro-airplanes. Cheryl met Cody Hill at an ultralight aircraft show in Albuquerque, New Mexico, in 2001. Both developed a love for the sport and frequently flew on weekends together. When Cindy founded her company in 2003, she hired Cody to be the production manager due to his background in mechanical engineering and his love of flying ultralight aircraft. Cody quickly developed a production plan, questioned other manufacturers about their best practices, hired a couple of workers, and began producing a limited number of ultralight planes. Today, the company employs 16 employees and has distributors (sales outlets) in several states. The Production Facilities UltraLight Aircraft currently houses its production facilities in a rented 5,000 square-foot warehouse in Tupelo, Mississippi. The monthly rent for the entire building is $3,000 (i.e., $0.60 per square foot). The company’s employees assemble aluminum tubing in a truss-like fashion that forms the skeleton of the airplane. The wings are attached to the skeleton, the engine (purchased externally) is mounted behind the seat (with the propeller facing rearward), and the frame is covered with sheet metal to improve airflow and to reduce wind resistance. The company currently purchases the wings from an outside supplier. The wings are then smoothed to remove burrs, coated to produce an aerodynamic surface, and aluminum and stainless steel brackets are attached so that the wings can be attached to the skeleton of the plane. UltraLight Aircraft expects to produce 80 planes next year. Sales are expected to increase by 10% per year for the following five years. UltraLight Aircraft currently pays the outside supplier $2,000 for each wing that is ordered. (Each plane uses only one wing, which is located above the pilot’s seat.) This contract is up for renewal and this supplier is willing to lock in this price for the next six years if a long-term contract is signed. UltraLight Aircraft Capital Budgeting Case Page 2 of 8 The Decision to Produce or Buy Cheryl and Cody are considering whether to bring the wing manufacturing process "in-house" and produce it as part of its manufacturing process. Since the wing is one continuous piece with a span of 30 feet, the stress on the wing while in flight is considerable. Whenever the plane changes direction suddenly, almost 100% of the wind forces are exerted upon the wing. For that reason, it must be made to rigorous manufacturing standards. The company has invested $20,000 to date on testing scale models of wing designs in wind-tunnel tests. Most of this money was paid to a capital goods manufacturer (Peachtree Engineering, Inc. of Atlanta, Georgia) which employs design and aerospace engineers. If UltraLight makes the decision to proceed, it will place an order with Peachtree to make a machine that will produce full-scale versions of the wings. If it purchases the new machine, the company will need to use 500 square-feet of its warehouse space that is currently unused. (This space was reserved for storage of parts in the future as the company expands.) The cost of the newly designed machine will be $121,000. The cost of shipping, installation, and technical adjustments add an additional $4,000. The equipment will be depreciated over a 5-year period (using MACRS) and will have an estimated salvage value of $16,000 at the end of the expected economic life of six years. The company will depreciate the entire cost of the machine (i.e., it will not deduct the salvage value from the cost when calculating the depreciable basis). If purchased, the new machine will produce the wings in one continuous extrusion process, which requires three workers to oversee the operation. Cody has determined that he is currently slightly “over-staffed” on workers in the assembly area and that he can transfer one worker from there to work on the wing without hurting the assembly process. (If he is not transferred, he will retain his current duties, which include some “make work” duties like sweeping and landscaping.) The workers will earn $9.00 per hour plus FICA/Medicare benefits. (FICA/Medicare employer contributions are 7.65% of the employees’ pay.) Training costs for all three are minimal. In addition, the company contributes $250 per month for medical insurance for each employee. The company’s projections are that the company will run only one 8-hour shift per day; employees will work eight hours per day, five days per week, fifty weeks per year (with two weeks of paid vacation when the plant is closed during the first two weeks of July each year). Cody has traveled to Atlanta twice to discuss the project with Peachtree officials, to tour the manufacturing facilities, and discuss changes in the design as they occurred. The total cost of the trips was $3,200. If the project is dropped, there will be no other travel expenses. If UltraLight Aircraft makes the decision to have the machine manufactured, one more trip will be required to sign the contracts and discuss the manufacturing process, at a cost of $1,000. The new machine will generate energy expenses of $6.00 per hour of operation. Energy costs are expected to increase 7% per year in subsequent years. The new machine will be maintenance-free and any repairs over the next six years are covered by the manufacturer’s warranty for the first two years and its service contract for the next four years. (The service contract is included in the purchase price). The machine will run continuously during the 8-hour day since the start-up and clean-up/shut-down time is complicated. The variable cost of materials used in the wing’s manufacture will be $25.00 per linear foot of wingspan (for the first year of production). Material costs are expected to increase at the rate of 5% per year. Working capital requirements (i.e., inventories necessary to support the maintenance/operation of the machine) are expected to be $6,000 at the time of the machine’s purchase and, after that, will be 6% of the following year’s material costs. (For example, at the end of year 1, the company will increase its working capital by an amount equal to 6% of the 2nd year’s material cost.) All working capital expenditures will be recovered at the end of the project’s life. If the new machine is purchased, the company can get rid of a machine used to coat the wings since the new machine uses a material that does not need coating. The company bought the machine three years ago for $18,000, UltraLight Aircraft Capital Budgeting Case Page 3 of 8 is depreciating the machine on a 5-year MACRS basis, and can sell it today for $7,778. (To simplify the issues, assume that any operating costs associated with this machine are negligible.) Neither Cheryl nor Cody are well-trained in financial analysis, but they decide to conduct the capital budgeting analysis for the new machine anyway. They are both very intelligent, have read a lot to prepare themselves, and have spoken with other firms about the procedures that those firms follow when analyzing similar proposals. As Cheryl and Cody discuss the proposed acquisition, it becomes quite clear that they have several differences of opinion. These differences begin to emerge as they begin to calculate the hurdle rate for UltraLight. Hurdle Rate Cheryl: The first thing that we need to do is determine the appropriate hurdle rate that we should use for evaluating the purchase of the new machine. That is, the company’s cost of capital plus an appropriate risk premium. Cody: Yes, let’s start with the cost of capital. We have no debt at the moment, so we’re currently financed entirely with equity. Our investors are happy with an 18% rate of return and we’ve been able to generate that for them in the past. So there’s an argument that we could use our current cost of capital of 18%. On the other hand, in all likelihood, we will raise all of the money for the project through a bank loan. We could use the interest rate on that loan (10.0% pre-tax, or 7.00% after-tax) as the cost of capital for any new projects that are financed with those funds.1 After all, part of the purpose of doing this analysis is to make sure that we recover our financing cost for the project, right? If we use only debt to finance the project, then the interest on that debt IS our financing cost – and therefore the appropriate cost of capital to use is the after-tax interest rate on the loan, i.e., 7%. Cheryl: No, everything that I’ve read says that we should use the overall cost of capital, which is a weighted average of all of the sources of financing. We don’t use preferred stock, so we would use a weighted average of debt and equity. We have $875,000 of equity now being used to finance the company. I’m guessing that this machine is going to require an initial outlay of between $120,000 and $130,000. Using $125,000 as the amount that we’ll need to borrow, we will have a cost of capital of 16.63%. Proportion % Cost Weighted Average Long-term Bank Note (after-tax) 0.125 7.0% 0.88% Common Equity 0.875 18.0% 15.75% Cost of Capital 1.000 16.63% Cody: O.K., I can see that. But I think that all of this misstates the cost anyway. We’re ignoring the effect that the additional debt has on the rate of return that our investors will want. Assuming that we raise the money in the form of a bank loan, the higher leverage will increase the risk of the company. As we’ve discussed in the past, our common stockholders seem to use the Capital Asset Pricing Model (CAPM) to estimate a fair rate of return to them. The Treasury note rate is 2% per year and I’m guessing that the stock market’s overall return will be 10% a year for the foreseeable future. The higher leverage will have the effect of increasing the beta from the current 2.0, up to a new level of somewhere around 2.5, so the CAPM indicates that 22.0% will be required to satisfy our shareholders at the new risk level. This means our cost is: Proportion % Cost Weighted Average Long-term Bank Note (after-tax) 0.125 7.0% 0.88% Common Equity 0.875 22.0% 19.25% Cost of Capital 1.000 20.13% 1 The cost of debt of 10.0%, but the interest is tax-deductible. In essence, by making the interest payments tax-deductible, the government is paying part of the interest. Since UltraLight is in the 30% tax bracket, the true cost of the debt is 10.0% * (1 - 0.30), or 7.0%. UltraLight Aircraft Capital Budgeting Case Page 4 of 8 Cheryl: Wow! That’s expensive. Cody: Yes, it is. Cheryl: So far, we’ve determined that the appropriate cost of capital is either … (jotting down numbers quickly)… 7.00%, 16.63%, and 20.13%. Gosh – my head is spinning! Appropriate Risk Premium Cody: Let’s leave the cost of capital for the moment and focus on the appropriate risk premium for the project. We should decide whether the rate should be different for different types of projects, such as replacing existing equipment for our current product line or purchasing equipment that we have no expertise in using (such as for a new product line). Cheryl: Well, as you know, a couple of our competitors always use a 3.0% risk premium for an analysis of all of their capital budgeting projects. Cody: I know - but that bothers me. The purpose of the risk premium is to make sure that we are fairly compensated for accepting the risk in the cash flow estimates. In other words, the “iffier” the cash flow estimates, the higher the risk premium should be. That seems fairly apparent to me. Take this project, for example - we have some relevant experience in using machines similar to this one. That means that we should be able to estimate the labor, energy, and materials costs associated with the new machine pretty accurately. In other words, the error in our cash flow estimates should be fairly small. If this were a machine being used to produce a brand-new product, then that would be totally different - we would have no experience to base our estimates on, plus we would have to estimate the revenues for the new product. I can see that we might be off considerably on our estimated costs and revenues in that case. The risk would be higher, so we would be justified in using a larger risk premium. I just don't see the justification at all for using a flat 3% rate for all capital budgeting projects. Cheryl: Perhaps you’re right. But the 3% number appears to be just high enough that it prevents the other firms from investing in projects that would have ended up losing money for them. And isn't that the overall purpose of the risk premium - to give us an extra “safety cushion” and to prevent us from investing in projects that will be unprofitable if our estimated savings don’t materialize? Cody: Your point is well taken. But we do have quite a bit of experience in with this type of equipment, so we’re not likely to have any big surprises. Why don’t we shave the competitors’ risk premium in half and use a 1.50% risk premium for a machine of this type? That reduces our cushion for making errors but I really believe that this is appropriate for this type of machine. Cindy: I still like the 3%, but maybe I’m just too conservative. Let’s see, I’m certain that the appropriate hurdle rate is one of the following combinations. I’m just not sure which one is the most accurate. Using 3.0% as the appropriate risk premium: Using 1.5% as the appropriate risk premium: 7.00% + 3.00% = 10.00% 7.00% + 1.50% = 8.50% 16.63% + 3.00% = 19.63% 16.63% + 1.50% = 18.13% 20.13% + 3.00% = 23.13% 20.13% + 1.50% = 21.63% UltraLight Aircraft Capital Budgeting Case Page 5 of 8 Cash Flows Cody: Why don’t we move to the cash flows associated with the machine? The initial investment (in “year 0”) is pretty easy – most of the costs are apparent (machine cost, shipping, etc.), except I’m not sure what to do with the cost of the trips that I took to Atlanta. Cheryl: It’s not that hard – the trips were necessary. We would never have considered buying the machine without your going to Atlanta to discuss and plan for it, and the costs are clearly associated with purchasing the machine. Those costs are similar to a feasibility study – you had to see if the new machine was a viable option for our company. So the trips’ costs should be allocated to the machine. It’s clearly a necessary cost to get the machine set up and working, so it’s therefore part of our cost of acquiring the machine. Cody: I suppose that you’re right. That does make it clearer. Let’s move to the remaining years’ cash flows. Cheryl: I’ve read that this formula should be used for estimating the operating cash flows: Annual cash flow = Savings * (1 – tax rate) + Change in depreciation * tax rate Our savings will be the “net” savings, i.e., we save money by not paying our current supplier but we generate additional costs by using the machine. Our net savings are simply our savings on the current contract minus our additional production costs. We also pick up some additional depreciation by buying the new machine and bringing that process in-house. That’s going to result in a substantial tax shield for us. As for the financial cash flows, we are likely going to finance this project with a 6-year bank loan that has an interest rate of 10.00%. Again assuming that we borrow $125,000, the annual payments on this loan will be $28,700. Let’s not forget to include them. Cody: Do we really need to include the payments on the debt? I thought that they were included already in the hurdle rate. After all, the hurdle rate includes the financing cost of the project. Cheryl: I think they should be included. After all, they are cash outflows associated with the project. And we should identify and include all the relevant cash flows. Cody: I’m also concerned about the effect of inflation on our cash flows. After all, we’re going to use three workers on the new machine. We’re paying them $9.00 per hour now, but you know that we have always tried to give annual percentage raises that are at least as large as the inflation rate. That way, our workers are able to maintain their purchasing power. In estimating our savings, should we consider only the current labor rate of $9.00 per hour or should we use the hourly rate that we actually expect to pay them? The inflation rate for labor over the next six years is expected to average 4% per year. And the company’s contributions for medical insurance will likely increase by 12% per year. Cheryl: If you're going to do that, you’ll need to make the same type of adjustment for the energy costs associated with the machine. After all, energy prices are expected to go up by 7% per year. But I don't believe that we need to do that. There is an inflation premium built into the hurdle rate already. That takes into account the effects of inflation, so we shouldn't have to adjust the cash inflows. We should use the current rate of $9.00 per hour as the labor rate over the next six years and the energy cost of $6.00 per hour, without any adjustment for inflation. The average increase in all of our costs (labor, energy, etc.) is already built into the hurdle rate and therefore the discount rate 2 already takes these inflationary changes into account without any further adjustments being necessary. 2 The discount rate is the percentage rate used for determining the present value factors, i.e., the hurdle rate. UltraLight Aircraft Capital Budgeting Case Page 6 of 8 Capital Budgeting Techniques Cody: O.K., so that lets us identify both the cash inflows and cash outflows. We’ll have the usual terminal cash flows in there as well. Once we perform the after-tax estimations, we’ll resolve the discount rate issues, determine the NPV,3 and we’ll be done. Cheryl: I prefer using the IRR4 method instead of the NPV. I’m more used to dealing with percentage rates of return rather than dollar amounts. After all, if this project generates a 30% rate of return, I can compare that to other options that we have to invest in: 5% in a money market account, 10.0% for paying off a bank note, etc. It's just an easier number to use. It's hard for me to visualize what NPV measures. Cody: A project with an NPV of $20,000 simply means that making this investment is equivalent to someone giving you $20,000 in cash today. It's the best measure of the profitability of a project and is theoretically superior to the IRR. Cheryl: Yes, but the IRR is used more frequently by businesses. After all, isn't there a scale effect associated with the NPV? Assume that I have two options for investing my money: 1. Option A has an initial investment of $100,000 and an NPV of $30,000, 2. Option B has an initial investment of $1,000,000 and an NPV of $31,000. Are you going to tell me that you would accept Option B because it has a higher NPV? Option A would have a higher IRR – I believe that it is the more accurate of the two methods. Cody: But that’s only for mutually exclusive projects – and we are only considering one machine here. But the IRR has its own unique set of problems. Sometimes, depending on the cash flows, there is more than one solution when using the IRR method. And there is a question as to how realistic its assumption about reinvesting cash flows is. If the project has an IRR of 80%, you are assuming that the project's cash flows can be reinvested to earn an 80% rate of return per year. Is this really realistic? Cheryl: O.K., O.K., so both methods may not be perfect. How about this? In my research into this, I've read that the MIRR5 overcomes the problems associated with the IRR method. It only has one solution and we can set the reinvestment rate to be any number that we want. We could use a reinvestment rate of 5%, which we would surely be able to earn. What if we calculate both the NPV and the MIRR and see what they recommend? If both methods suggest that we buy (or not buy), great – we’ll go with that. If they differ in their recommendation, then we’ll sit down and decide which to use to make our decision. Cody: That sounds fine to me. Adjusting for risk Cheryl: There’s still one final issue that we need to consider: a risk analysis. We are considering the risk of the cash flows already by adding an appropriate risk premium to the cost of capital. Here’s a copy of an article that appeared in Manufacturing Digest on this subject. (She hands the article to Cody.) As it turns out, there are other ways of adjusting for the risk that might be useful: 1. vary the discount rate for each year – This would be easy to do. Instead of having one hurdle rate, we could have several. That is, we could use a different discount rate for each year's cash flow. After all, the farther into the six-year period we go, the more difficult it will be to arrive at an accurate estimate. In other words, the risk of being wrong on our estimates increases as we forecast farther out into the future. Following the risk-return trade-off principle, we can require a higher rate of return (i.e., hurdle rate) for the later years' cash flows. How much should the discount rates change from year to year? I suppose that the average of the six discount rates should be 3 NPV = Net Present Value 4 IRR = Internal Rate of Return 5 MIRR = Modified Internal Rate of Return UltraLight Aircraft Capital Budgeting Case Page 7 of 8 approximately the same as the hurdle rate that we use under the original method (i.e., hurdle rate = cost of capital + the risk premium). 2. a scenario analysis – This is kind of a worst-case, most-likely, best-case type of analysis. For example, we can calculate the NPV using our lowest estimated savings, our most-likely level of savings, and our most optimistic estimate of savings. If we have an opinion as to the probabilities of these outcomes, we can assign probabilities to each of the outcomes and calculate an expected NPV (by multiplying the probabilities of these savings occurring times the NPVs for each of the levels of savings). Or we can simply consider the range of possible outcomes without using probabilities. Microsoft Excel’s Scenario Manager can help us to model this.6 3. a sensitivity analysis – We can determine the NPV over a range of possible input values (i.e., key input drivers) at relatively small intervals. For example, we could calculate our expected after-tax cash inflows for years 1-6, then express each year’s savings as a percent of the 1 st year’s cash flow (i.e., 100%, 94%, 90%, etc.). We could then vary the 1st year’s cash flow over a dollar range from the lowest level expected to the highest level expected and then use Microsoft Excel’s Data Table feature to conduct the sensitivity analysis. 4. A certainty-equivalent analysis - This is very similar to the worst-case part of the scenario analysis, but giving weight to the fact that the cash flows are certain. This means that the risk premium for the discount rate should be set to the risk-free rate (since there is no risk in the cash flows). 5. A Monte Carlo analysis - We can identify a range of possible values for a key input driver (such as the total annual savings associated with the machine) and let the Monte Carlo computer program randomly select values within these ranges. The computer will calculate the NPV for each set of inputs, and repeat this process for 1,000 times or more. We can then calculate a probability distribution to measure the probability that the project will have an NPV that is less than zero. Cody: Surely it isn’t necessary for us to conduct a test using all of these methods; after all, they do test the same thing, right? However, it should be relatively easy for us to choose one of these - say, method #2 on your list (a scenario analysis), since the computations are relatively easy using an electronic spreadsheet. Using this in addition to the one that we had planned already (adding a risk premium to the cost of capital) should give us a relatively complete view of the risks that are inherent in the project. Cheryl: I agree. Let’s do that. 6 There is a video on the website that explains the setup for Excel’s Scenario Manager feature. UltraLight Aircraft Capital Budgeting Case Page 8 of 8 A Suggested Procedure for the Analysis As you have seen, Cheryl and Cody make a number of errors. You should identify and resolve these errors; then, in a typed report, you should address the following tasks and questions. You should attach a well-organized, clearly labeled spreadsheet to analyze the problem and to support your analysis. You may provide the end-result (or summary) of your calculations in the write-up and leave the details to the spreadsheet, but the write-up should be self-contained and completely address the major issues, specifically: 1. Choose the appropriate hurdle rate that should be used in the initial analysis. Justify your decision. 2. Identify any sunk costs and irrelevant cash flows that may be inherent in the decision.7 3. Determine the correct manner to treat inflation and whether or not you should inflate the cash flows. 4. Explain why the loan payments should or should not be included in the relevant cash flows. 5. Determine the “year 0” cash outflows. (Include the present value of the future working capital cash flows.) 6. Determine the net after-tax cash flows for years 1 – 6. (Keep in mind this question: “If we make the decision to buy the machine, what cash flows will change?”) Make sure that you include the terminal cash flows for year 6. 7. Calculate the NPV and MIRR of the proposed investment; use a 5% reinvestment rate for cash flows in the MIRR method. (If the two methods disagree, state which of the methods you should use and the reason why.) 8. Conduct a risk analysis of the proposed investment, as proposed in the case. Make sure that you include this analysis in your spreadsheet. (You may choose any reasonable numbers that you want to use for the worst case and best case cash flows.) 9. Make a clear recommendation as to whether UltraLight Aircraft should purchase the machine or renew the existing contract with the outside supplier. Justify your recommendation. MACRS Depreciation Table Year 3-year 5-year 7-year 10-year 1 0.333 0.200 0.143 0.100 2 0.445 0.320 0.245 0.180 3 0.148 0.192 0.175 0.144 4 0.074 0.115 0.125 0.115 5 0.115 0.089 0.092 6 0.058 0.089 0.074 7 0.089 0.066 8 0.045 0.066 9 0.065 10 0.065 11 0.033 Formulas (if needed for your Excel calculations): The formulas for determining the present value factor (PVF) and the future value factor (FVF) for a single year is: 1 PVF where r is the required rate of return and n is the time period (1 r) n FVF (1 r) n 7 Sunk costs are always irrelevant, but not all irrelevant costs are sunk costs. There may be some costs that will occur in the future but they are not relevant to the decision and should not be included in the analysis.