Appendix A
Economics of Project Analysis
How Industry Decides to Invest
Much of a company's planning centers on maximizing profit from capital investment. Capital investments are selected in priority order with consideration of ongoing burdens of depreciation, insurance, taxes, and maintenance costs; reduced flexibility; and increased risk and liability. Often, decision makers forecast profits for several action/no-action pairs. Only a positive difference between options justifies increased investment. Investments must also meet an acceptable return on investment to warrant continuation and completion of the project. Some of the factors influencing the profitability of capital investments are shown in Table AA-1.
Table AA-1. Factors Influencing Profitability
|
Economic life span
Depreciation life and method
Construction period
Cost stream (insurance-disposal) over technology life
Salvage value (usually small)
Inflation
Tax rate
Product revenues or production cost savings
Lost production during installation
Risk |
Industry also has qualitative considerations, which, at management's discretion, can supersede quantitative cost considerations. A recent Austrian study of pollution-prevention investments revealed that 30 percent of the companies that made such investments did not even attempt to calculate the economic costs and benefits. Other non-cost reasons for investment include production problems (reliability, quality, regulatory compliance); public image (zero discharge, "green" issues, toxics discharge list); and business outlook (future liability, international markets).
Estimating the Costs and Benefits of Technology Projects
To determine the economic and other benefits of individual projects, a comprehensive set of economic data are required, including costs-savings associated with end use in the industrial process. Table AA-2 provides a list of the data most commonly required for an initial analysis. In the early stage of a project many of these costs are estimated at a fairly low level of detail. As projects near pilot-stage or commercialization, more complex data is required to more accurately apses profitability.
The data shown in Table AA-2 can be used to determine the economic attractiveness for both the developer and end-user. The developer is the firm involved in the research, demonstration, or manufacturing stages of the project. The end-user is the firm utilizing the technology in its manufacturing operation. Often the developer and end-user are the same entity.
Table AA-2. Cost Data Requirements
Acquisition investment
- Capital cost of equipment
- Installation cost
Useful life of equipment (years for depreciation)
Salvage value of equipment
Total annual output (e.g., tons, lbs of product)
Annualized costs:
- Maintenance
- Energy
- Waste
- Depreciation
- Labor
- Other Operating Costs (explain)
Annual production cost savings
Selling price of unit |
The following discussion describes a complete financial analysis including income taxes. A complete financial analysis requires examination of each year's current-dollar cash flow, including income taxes for that year. The results of a financial analysis will provide information on:
- Payback period (discounted)
- Total life-cycle cost
- Total life-cycle benefits
- Net present value
- Benefit cost ratio
- Internal rate of return
- Return on investment
Table AA-3 defines the terms used in the following calculations.
Table AA-3. Definition of Terms Used
for Current-Dollar Cash Flow Analysis
(AA.1)
N = Equipment useful lifetime
Ci = Cost in ith year after initial investment
d = Real discount rate
B = Non-energy (productivity increase, waste reduction) benefit in the ith
year
Qji = Annual net energy saved or produce of jth type in ith year
Pj = Price of jth energy type
T = Annual income taxes paid
t = effective income tax rate (including federal, state and local taxes)
UCRF = Uniform Capital Recovery Factor (see equation (6.1))
C0 = Capital investment
N = Investment useful lifetime |
| |
In this analysis, currency loses value. The uniform capital recovery factor, which is based on equipment life and interests rate, is used to determine the base rate to be compared with the benefits. The uniform capital recovery factor (UCRF) is

It is then used to estimate the levelized energy cost (LCE) as follows:
(AA.2)

The following equation is used to estimate the income taxes in levelized costing:
T= [t/(1-t)] [UCRF x C0 - C0/N] (AA.3)
where T is annual income taxes paid; t is the effective income tax rate (including Federal, state, and local taxes); C0 is capital investment; and N is the investment's useful lifetime. Profit is T/t. The annualized production cost savings (APCS) are as follows:
(AA.4)

APCS connects energy savings to national economic benefits via productivity (Pj) improvement. This normally is a lower level of the economic benefit. Equation [6.5] shows the total life cycle cost (TLCC):
(AA.5)

TLCC can be used to compute a more sophisticated version of equation 6.2. The total life cycle benefit (TLCB) is the present value of all benefits associated with the technology. Embodied energy and all "externalities" are incorporated into Bi:

First, costs must be compared with a baseline system, and a system must be selected. The system's boundary must be drawn, and all unit processes within that boundary must be understood. Boundary considerations relative to the audience, user, developer, supplier, customers, public, and so forth, must be determined. Finally, the financial analysis boils down to a comparison of competing systems. No matter which system is used , check the completeness of the project scope, and draw from statistically proven cost experience. Make sure that you include fundamental components: materials, labor, and overhead. The net present value equals the benefits (TLCB) less the costs (TLCC), or
NPV = TLCP - TLCC
NPV is the worth of the technology to the investor. It must be positive and have a certain "hurdle" rate to be considered for investment by either the company or the TWP. Net present value is a generally applicable measure that corrects for the time value of money. It can also be used to help optimize the scale of a prospective plant or technology unit.
Another commonly used measure is the benefit-cost (BC) ratio:
BC = TLCP/TLCC
Just as the net present value must be positive, the benefit-cost ratio must be greater than 1. The latter ratio is used more often by public policy makers. Two other measures are follows: The simple pay back time T1 or 1/ROR, where ROR is the internal rate of return; and the discounted pay back period is year n for which TLCC is less than or equal to TLCB.
The internal rate of return (IRR) is the discount rate for which NPV is zero, or the rate for which the benefits and costs are equal. This rate is commonly used to compare internally funded projects.
Such a comparison assumes all revenues are reinvested at the internal rate return. Since solving this equation6 can be difficult, the rate of return (ROR) is often used as a first approximation for the internal rate of return:

where all subscript l quantities are taken as their value in the first year, C0 is the capital investment, and C1 is the non-energy operation and maintenance cost in the first year. This does not account for the value of money over time, so the approximation is better when the time scale is short or the discount rate is very low. The rate of return also does not provide information on the overall life cycle. The full discounted pay back period (see Table 6.7) is the time to break even; this measure is more accurate and easier to understand but often difficult to calculate. It is also the minimum time span for which assumptions must hold for the analysis to be valid. Its full discounted pay back period is estimated for option analysis, but it does not account for investment lifetime or size.
|
Net Present Value (NPV) = TLCB-TLCC
Benefit Cost Ratio (BC) = TLCB/TLCC
Simple PayBack Time T1 = 1/ROR
Discounted Pay-back Period = year n for which TLCC < TLCB |
A breakdown of typical capital costs is listed in Table AA-4. The rules of thumb in Table AA-5 provide some guidelines for estimating costs when data is unavailable. Table AA-6 describes typical operating and maintenance costs encountered in manufacturing facilities.
Table AA-4. Breakdown of Capital Costs
|
First cost of equipment
Installation cost
Construction indirects (percentage of installed costs)
Land (percentage of installed costs, other)
Engineering, site preparation ( percentage of installed costs)
Contingency allowance (percentage of installed facilities)
Working capital (number of days cash operating expenses)
Interest during construction (cash flow, time)
Start up expenses (number of days operating expenses)
Less any applicable investment tax credits |
Table AA-5. Factors for Estimating Capital Costs
|
Total Installed Plant Cost Breakout
(as percentages of total) |
Direct Costs |
Indirect Costs |
| Process Equipment |
25-40 |
Home Office Expense |
|
| Process Equipment Labor |
1.5-4.5 |
Plus Overhead |
11-20 |
| Process Material |
8-18 |
Field Expense |
|
| Process Material Labor |
4.12 |
Overhead |
6-14 |
Table AA-6. Typical Operating and Maintenance Costs
|
Payroll plus labor indirects
Operating supplies (percentage of payroll)
Maintenance supplies (percentage capital investment)
Electricity
Purchased fuels4
Outside services
Local taxes and insurance (percentage of total investment)
Feedstock costs
Pollution control costs
Transportation costs
Waste disposal costs
Royalties
Income taxes |
Benefits calculations often require projections to 2010, and decisions must be made as to how to project costs. Table AA-7 summarizes the features of two common methods. Future costs can be expressed in constant (base year) dollars or current (nominal) dollars.
Table AA-7. Constant and Current Dollars
| Constant Dollars |
Current Dollars |
Exclude future inflation or deflation.
Benefits and costs must be in same-year dollars.
If inflation is the same for all commodities, the pre-tax analysis is unchanged.
Real, not nominal, discount rate must be used. |
Include the effects of future inflation.
After-tax cash flow analysis requires current dollar prices.
General price indexes (e.g., Standard & Poors) may not apply to specific items.
|
Cost-Benefit Estimation Through Scaling
If some of the data is unknown, there is an alternative: Rather than calculating everything from first principles, one can often take advantage of an existing cost calculation provided that one scales accordingly. Using previous calculations is appropriate if only the size or capacity is different, the year is different, or the experience of the equipment operator is different.
- To scale by equipment size or capacity, one assumes that cost is proportional (to within an exponent) to size or capacity. Thus the new cost is described by
C(S) = C(S0) x (S/S0)f
where C(S) is the adjusted cost and the known cost of equipment at size/capacity S; and C(S0) is the known cost of equipment at size/capacity S0. The new size/capacity is S, and the size/capacity for which cost is known is S0. Finally, f is the fitted exponent, which varies for different equipment types, but it is often taken as 0.6.
- To scale by year, base year adjustment, one assumes cost is proportional to the relative producer price index (inflation rate factors) in base and scaled to year, as shown in
C = C0(i/i0)
Where C is the cost during the reference year, C0 is the known cost during the previous year, i is the wholesale or producer price index in the reference year, and i0 is the wholesale or producer price index in previous year.
- To account for dropping cost of a unit as its users become more experienced,use
C(N) = EXP{ln C0 - (f x ln N)} + K
where C(N) is the total constant dollar unit cost of Nth unit, N is the total number of units produced to date, and C0 is the cost of the first unit less the fixed portion (equal to c(1) - K). Additionally, f is an empirically fitted constant, usually between 0.2 and 0.3, and K is the fixed portion of unit cost
Technology Forecasting: Scenarios for the Future
This section provides background on how economic analysis takes into account competition, market penetration, and market growth over the next twenty years. . The underlying assumption in all these calculations is that the performance of the technology under development will result in an increased market share compared to the baseline technology. This baseline technology is referred to as the current technology; it is the technology with which the proposed technology will be competing once the proposed technology is commercialized.
There may be more than one competing technology. However, new technology projects are only expected to project into the future based on the dominant competitor. The performance and economic characteristics associated with this current technology can be difficult to estimate. This is especially true when the proposed technology commercialization date is far in the future or the technology is used in highly innovative industries. In such situations, future trends in other (as yet uncommercialized) technologies must be estimated. Three scenarios are possible, and are discussed below in order of increasing difficulty:
Situation 1. The current technology does not change significantly between now and the commercialization of the proposed technology. The characteristics of the best cost-effective available technology are used for comparison; if they are not available, the industry average of what is in place today is used. This is probably a reasonable assumption for industries characterized by long-lived capital stock, such as many of the process industries.
Situation 2. The current technology continues to follow a (currently observable) trend in performance and cost. A linear extrapolation is used unless it produces unreasonable answers. Such extrapolation is necessary for technologies used by rapidly innovating industries, such as electronics.
Situation 3. The current technology is not really current, but is a future technology. It is an alternative substitute technology now under development that will be competing with the proposed technology. Such extrapolation is necessary in the case of a banned substance, such as chlorofluorohydrocarbon (CFC) replacement technology.
The economic growth of the U.S. market for the proposed and the competing technologies must be estimated in order to estimate concrete benefits. Three growth scenarios are possible:
Scenario 1: Growing Market The demand for the technology under development (and its competitors) will increase in the future. This is the case for new consumer products such as high technology manufacturing, information technology, and the like.
One could use the average annual gross national product increase (around 2-3 percent) to scale up the current total to the year 2010, then take the market penetration percentage of that total number or use a more rigorous approach. Or assuming the technology is commercialized in 1994, one would start with the current market as shown below in Figure AA-1, then project the growth of the market that would be expected (independent of the "proposed technology") as shown as the open squares. If the current level represents the existing installed capacity at plants, this market segment may be more difficult to penetrate, because long-lived units may have been recently installed, whereas the new plants or units that need to be installed may be easier to penetrate with the new technology. Eventually, the new technology may begin to penetrate the established plants, replacing units that have worn out. Or if the new process is sufficiently economic, it may replace old units before their normal end-of-life. Therefore it is possible for the new technology to end up with a greater number of units than currently exists, without assuming 100 percent market penetration in 2010. Extra care must be taken in situations involving severe international competition or technology markets (such as information technology) which are now growing much faster than the overall economy.

Figure AA-1. Scenario of a growing market.
Scenario 2: Declining Market. In this case, demand for the technology is declining. One example is when the proposed technology is specifically for eliminating an undesirable waste, such as volatile organic compounds. Since it is undesirable, industry may already be working on entirely different technological methods for reducing the emissions of these wastes, which involve using a totally new technology. Therefore a downward trend in that market may be experienced or expected. Under these conditions, the number of such waste reduction technology units can only be assumed to be the difference between the lower level expected in 2010 and what the proposed technology can achieve, rather than the current emissions which are much higher (see Figure AA-2).

Figure AA-2. Scenario of a declining market.
Scenario 3: Product Is Banned
Extra care must be taken in situations involving a ban of a product or a chemical (i.e., chlorofluorohydrocarbon). In this case, there will be a sudden drop when the ban becomes effective, but the new product or process being developed is unlikely to be the only product competing for the replacement market. This is shown in Figure AA-3, where the proposed technology to replace the banned product and a competing new technology are both getting increasing market share, and the total replacement market may or may not be the same size as the original market. For these cases, estimate the kinds of competition from other new producers.

Figure AA-3. Scenario when a product is banned.
ENDNOTES
- Stenum, J.J. 1994. "Stoff-Energie-Umwelt", The Technical University of Graz' Institute of Chemical Engineering. personal communication,
Washington D.C., August 31.
- Additional References on economics:
Stermole, F.J. 1984. Economic Evaluation and Investment Decision Methods, Fifth Edition, Investment Evaluation Corporations.
Weber, J. E. 1982. Mathematical Analysis: Business and Economic Applications, Fourth Edition, Harper and Row, Cambridge.
Samuelson, P. A., and W.D. Nordhaus. 1985. Economics, Twelfth Edition, McGraw-Hill Book Company, New York.
Au, T., and T.P. Au. 1983. Engineering Economics for Capital Investment Analysis, Allyn and Bacon Inc., Boston.2.
- This choice of a discount rate can make or break the case for investing in reducing environmental risk. If scientists agree that a particular pollutant
would cause $100 million in damage in the year 2094, there are two economic choices: invest now in pollution abatement technology to avoid the
damage, or channel the investment elsewhere on the assumption that a century from now, the company will be richer as a result of the alternate
investment, and better able to absorb the costs of any damage to the environment. If decision makers assume that for every $1 they invest now they
can get $1.10 next year (10 percent discount rate), it would be fiscally irresponsible to spend more than $7,305 on pollution control now to avoid $100
million in cleanup costs a century from now. If, however, the discount rate is 2%, almost $14 million in pollution control expenditures are justified
today to avoid $100 million in costs in 2094. Most economists use a 10 percent discount rate, and this is also the rate that DOE analysts have historically
used for investments in conservation, but some argue that in the long run 2 percent is more reasonable based on expected future growth rates.
- Perry, R.H., and D.W. Green. 1984. Perry's Chemical Engineers' Handbook, Sixth Edition. McGraw-Hill Professional Book Group. 1846 pp.
- If multiple increments of investment occur, solving for this condition can lead to multiple positive roots.
- For costs of process equipment see the following:
- McGraw-Hill, Chemical Engineering Equipment Buyers' Guide, annual - Thomas Publishing Co., Thomas Register of American Manufacturers, annual - U.S. Bureau of Labor Statistics, Producer Price Indexes, monthly & annual - Perry, R.H., and D.W. Green. 1984. Perry's Chemical Engineers' Handbook, Sixth Edition. McGraw-Hill Professional Book Group. 1846 pp.
- For typical labor rates see the following references:
- U.S. Bureau of Labor Statistics, Employment and Earnings, monthly
- U.S. Department of Commerce, Industrial Outlook, 1994
- Perry, R.H., and D.W. Green. 1984. Perry's Chemical Engineers' Handbook, Sixth Edition. McGraw-Hill Professional Book Group. 1846 pp.
For labor indirects by industry see
- U.S. Bureau of Labor Statistics, Employer Costs for Employee Compensation
- U.S. Department of Energy, Energy Information Administration, Monthly Energy Review
- For typical costs see
- Pollution Equipment News Buyers Guide, annual
- PACE survey, annual
- Various IRS tax guides, etc.
- To find the appropriate exponent, see Perry's Chemical Engineering Handbook, "Equipment Capacity Scale-factor Cost Adjustment."
- Installation labor costs are scaled similarly but with different exponents.
- See U.S. Bureau of Labor Statistics, Wholesale Price Index.
- For reference on this formula and which constant to use, see Bost Consulting Group, Inc. 1972. Perspectives on Experience.
- See IWP Project Benefits Calculations Guidance, August 31, 1994.
- Additional references on economics:
Button, D., C. Sabo, and K. Seiden. 1990. Uncertainty in Forecasting, EPRI CU-6855 Research Project 2919-2, Palo Alto, EPRI
Judge, G.G., et al. 1982. Introduction to the Theory and Practice of Econometrics. Radius Press, New York.
Pindyck, R.S., and D.L. Rubinfeld. 1981. Econometric Models & Economic Forecasts, Second Edition, McGraw-Hill
Book Company, New York.
Return to ENERGY, ENVIRONMENTAL, AND ECONOMICS (E3) HANDBOOK |