By Hitesh Doshi, M.A. Sc., P. Eng. Nature of Roofing Decisions Roof consulting, by its very nature, requires making many decisions pertaining to the design, maintenance, repair and replacement of roofing systems. Decisions are made based on the needs of the client, requirements to resist environ¬ mental loads, and those dictated by prevalent local practices, standards, and codes. Prudent owners demand that the suc¬ cessful roofing choice meet all the functional requirements cost effectively. Initial Cost and Cost Effectiveness Designers and consultants often feel that owners view cost effectiveness the same as lowest initial cost. Proper articulation of other than initial costs can draw the owner’s attention to life cycle costs. Most business owners are famil¬ iar with capital investment decisions which form the back¬ bone of any business enterprise. These decisions require analysis which takes into account initial and future cash flow considerations. The basic methodology to deal with business investment decisions has not changed in several years and can be found in standard college courses and text¬ books on financial management and engineering economics such as Brigham (1991), Riggs (1986). Replacement deci¬ sion analysis based on future streams of costs and revenues is also commonly covered in such courses and texts. Furthermore, these types of analysis are now relatively easy to conduct using built in functions of business calculators. Roofing Decisions as Business Investment Decisions Treating roofing decisions as a capital investment deci¬ sion, similar to business investment decisions (such as buy¬ ing one type of production equipment or computer over another), allows roof consultants to apply the techniques that business owners understand. Clearly, if one roofing alternative has higher initial costs than another, the owner needs justification on how the higher initial cost alternative will save them money in the future. The future cost savings need to be compared to the initial increased cost to deter¬ mine if the alternative is worth considering. There is one major difference between roofing decisions and other business investment decisions: roofing invest¬ ments very rarely produce a revenue stream like other busi¬ ness investments. Roofing investments only produce a cost stream. The question of cost effectiveness of roofing alter¬ natives should therefore be addressed by asking the ques¬ tion: Does the future cost reduction of one alternative justi¬ fy paying more for it in the present? For example consider roofing systems A and B. Roofing system A has an initial cost 25% more than that of B. If the yearly maintenance costs and periodic repair and eventual replacement costs of A are also higher than that of B, no further analysis is required. Alternative B is clearly cost-effective over A. On the other hand, if the yearly maintenance costs and periodic repair and eventual replacement costs of A are lower than B, then at what point will A be more cost effective than B? When future costs are involved, two things are necessary to estimate their impact: a) The timing and amount of the future cost b) Discounting Timing of cost is indicated relative to the base year. The amount of future costs is generally stated in terms of base year values. Discounting acknowledges the time value of money. The time value of money is a measure of the earn¬ ing power of money compared to a base year. It is different from inflation, which is a measure of the purchasing power of money compared to a base year. Discounting recognizes that a future cash flow stream is equivalent to a lesser base year amount because of the power of interest compounding. In typical business investment decisions, the future cash flow stream is converted to its equivalent value in the base year (present) by applying the selected discount factor. The net present value is then calculated and compared for each alternative. In building economic decisions this is simi¬ lar to calculating the net present value by a method often referred to as Life Cycle Costing (LCC). Other types of decision-making criteria include: a) Payback—time required for the cumulative savings to equal the added initial investment b) Savings to investment ratio (SIR)—ratio of the dis¬ counted net present value of savings to the increased February 1997 Interface • 7 initial investment. c) Internal rate of return (IRR)—or the rate of return of the increased initial investment resulting in the future savings. Initial and Future Costs Initial Costs To facilitate economic analysis of roofing decisions as business investment decisions, it is necessary to identify the initial and future costs of the various alternatives. Initial cost of an alternative is the sum of all costs incurred at the time of implementation. Initial costs may be taken from the received bids, or estimates prepared as a part of a consul¬ tant’s report. Future Costs Future costs related to roofing are divided into annually recurring maintenance and operating costs and non-recur¬ ring operating and maintenance costs and capital costs. The annually recurring costs typically include: costs associated with wintertime net heat loss and summertime net heat gain, cost of visual inspection, and cost of general preventa¬ tive maintenance. Non-annually recurring operating and maintenance costs generally include: cost of system-specific preventative maintenance (e.g. recoating surface finish), cost of non-destructive evaluation and cost allowance for minor repairs. Non-annually recurring capital costs include items such as roof restoration, flashing restoration, and cost allowances for larger areas of repairs, including wet insula¬ tion, and the cost of replacement/recover. Study Period and Life Expectancy Generally, future costs are considered over the study period of the decision analysis. For instance, if the owner will keep the building for 40 years, all costs that occur over the 40 years will need to be considered. If a roof alternative for this owner has a life expectancy of 15 years, its replace¬ ment cost will be considered at 15 and 30 years, along with the other annually recurring and annually non-recurring costs. If the study period does not coincide with the life of the alternative, a salvage value may be assessed to the alter¬ native. There are different techniques available for calculat¬ ing salvage value. A simplified straight line depreciation can be applied without much error in analyzing roofing invest¬ ment decisions. The life expectancy of the alternative will impact the net present value, but it is not required that all alternatives have the same life expectancy in performing the analysis nor is it required that life expectancy be the same as the study period. Economic Analysis and Roofing Decisions The application of economic analysis to roofing invest¬ ments is not new. For example, Griffin et al (1995) have shown the use of life cycle costing in determining the opti¬ mum roof slope to be 2%. In another example, they use life cycle costing to show that additional investment to provide slope will provide an equivalent return on that investment of 37%, when compared to the alternative of not providing slope and an early roof failure. They also show an example where the life cycle costing analysis is done between a pro¬ tected membrane roof (PMR) and a conventional roof to show how the increased cost of PMR more than pays for itself over a life cycle study period of 20 years. Life cycle costing of a re-cover over an existing wet roof was done by Desjarlais (1995) and associates. They used the Internal Rate of Return (IRR) as a measure to compare the cost effectiveness of initial investment of the recover option, which resulted in decreased operating and mainte¬ nance costs. They have also shown the impact of different maintenance rates on the IRR. An illustration on the use of LGG to choose between patching an existing 30-year-old BUR and installing a new single ply roof is shown by Melvin (1992). Examples that apply to different building decisions similar to roofing investment decisions can also be found in ASTM standards and Marshall (1990). Steps in Conducting Roofing Economic Analysis Step 1 – Define the objective Before an economic analysis is performed to evaluate a roofing investment, the objective of the evaluation has to be defined. The objective may be to select the cost-effective roofing system; to select the cost-effective R value of insula¬ tion; to decide if it is cost effective to defer maintenance; or to decide if it is cost effective to repair an old roof or replace it. The objective will lead to the formulation of alternatives. Step 2 – Identify feasible alternatives The second step is to identify feasible alternatives for accomplishing the objectives. It is imperative to identify functionally comparable alternatives. Alternatives that do not meet the functional requirements should not be consid¬ ered. Step 3 – Identify the study period It is necessary to determine the study period over which the economic analysis will be performed. This may or may not be the same as the life expectancy of the alternatives. In general, the effect of discounting diminishes the impact of costs and revenues significantly on the outcomes beyond a 25-year study period. Furthermore, future costs may become more unpredictable as the study period is increased. For most roofing-related evaluations, the error possible by limiting the study period to 30 years is minimal. Step 4 – Compile data for each alternative For each of the alternatives, it is necessary to determine initial costs, the annually recurring costs, the non-annually recurring costs and their timing, including end-of-life replacement costs. Where the life expectancy of the alterna¬ tive is longer than the study period, an appropriate salvage value may need to be assigned to account for the alterna¬ tive’s potential to remain functional. For most roofing-relat- 8 • Interface February 1997 ed evaluations, salvage value based on straight line prora¬ tion will provide acceptable results. In most standard analy¬ ses it is assumed that the costs occur at the end of the year. The time-based cost profile of the alternative is called its cash flow stream. A cash flow stream should be developed for each alternative. If the rate of inflation applies equally to all costs, the cal¬ culations can be based on non-inflated values or constant dollar values using a discount rate that is the net of inflation rate. Energy costs and disposal costs are likely to be the only costs that may rise at a faster rate than the general inflation and may need special treatment. Marshall (1990) provides more details on the impact of inflation, taxes, depreciation, financing and study period. Step 5 – Select an appropriate discount rate A discount rate is used to discount the future cash flows to their present value. Typically, the discount rate can be thought of as the interest rate that the user would be expected to earn if they chose not to invest in the roofing investment. For businesses, the discount rate reflects the return on investment they expect to make on their invest¬ ments. For homeowners, it reflects the interest rate of their mortgage or term deposits. Proper consideration has to be made when selecting dis¬ count rates. Uncertainties in the discount rate can easily be handled by conducting the analysis for different rates and noting the variations in the outcome. This type of analysis is also called sensitivity analysis. Step 6 – Discount future cash flow streams This step ensures that the value of all future project income and expenditures reflects the effect that time and interest have on money values. It allows one to compare a stream of future costs and benefits by transforming them to the same point in time, generally the base year or the pre¬ sent—hence the term present value analysis. Future and annual time equivalencies are also possible to do and desir¬ able under some circumstances. Step 7 – Select the cost-effective alternative Once the discounted values are calculated, the economic measure of interest can be calculated—i.e, Life Cycle Cost (LCC) or the Net Present Value, Discounted Payback, Savings to Investment Ratio (SIR) or Internal Rate of Return (IRR). This step can be performed using standard formulas for discounting—Marshall (1990), ASTM stan¬ dards, or using a computer program such as building life cycle costing developed at the National Institute of Standards and Technology, Gaithersburg, MD. The choice of the alternative, based on the economic measure, can then be made; i.e., select the alternative with the lowest LCC, or with the greatest SIR greater than 1, or the greatest IRR greater than the discount rate. A sensitivi¬ ty analysis can be performed to see how the outcome changes if one of the parameters such as cost, life expectan¬ cy or discount rate changes. A decision as to the most costeffective alternative can then be made. Examples/Case Studies The following examples demonstrate some of the infor¬ mation presented earlier based on common situations encountered by a roof consultant. As is evident in these examples, the use of economic analysis creates an opportu¬ nity for roof consultants to make better cost-driven deci¬ sions. Example 1 Situation: An owner prefers a hybrid 4-ply BUR system for a new 30,000 sq. ft. roof. This is called alternative A. The cost of this roof is estimated to be $180,000 and includes upgraded membrane flashing and comes with a manufacturer’s war¬ ranty of 15 years. The manufacturer estimates that other than normal preventative maintenance and visual inspection by the owner’s representatives, there is no other mainte¬ nance required. The maintenance cost is estimated to be $l,800/year for the life of the roof, which is estimated to be 25 years. The owner has been approached by a contractor who can provide a conventional 4-ply BUR system with glass felts for $150,000. This is called alternative B. The contractor only provides a standard association warranty of 2 years. The maintenance cost of this roof is estimated to be $3,600/year for the life of the roof. Flashing repairs may be required around year 15 at a cost of $8,000. This will ensure that the roof will last 25 years. Table 1 Cost Category Alternative A Hybrid BUR 15-Yr. Warranty Alternative B 4-Ply Glass Felt BUR 2-Yr. Warranty Cost BCost A Initial Capital Cost $180,000 $150,000 ($30,000) Present Value of Operating Costs $16,339 $34,592 $18,254 Present Value of Replacement Costs Present Value of Salvage Subtotal of PV of Costs and Salvage $16,339 $34,592 $18,254 Total Life Cycle Cost $196,339 $184,593 ($11,746) February 1997 Interface • 9 The owner retains a roof consultant to determine whether alternative A is better than B over the 25 years of expected life of the systems, based on life cycle costs. The discount rate for the owner is 10%. Solution: Steps 1 to 5 have been completed in the above situation. The next step is to discount future cash flow streams. The cash flow streams for the two alternatives are shown in a cash flow diagram. The BLCC program version 4.2 was used to complete step 5 and the results obtained are shown in Table 1. Based on the results shown in the table, it is obvious that increased initial cost of alternative A of $30,000 is larger than the present value of the savings of future costs of $18,254. This cost saving of the client’s preferred alterna¬ tive A is less than its initial cost outlay. The LCC of A is more than B by $11,746, and, therefore, makes B cost effec¬ tive over A. The SIR is calculated as the ratio of the savings of $18,254 to the increased initial investment of $30,000 and gives a value of 0.61. This is less than 1, indicating once again that the alternative A is not cost effective. Simple pay¬ back can be calculated as the number of years it takes to pay back the initial investment not accounting for the effects of discounting on the savings. The yearly savings are $3,600-$l,800 = $1,800. Simple payback of the initial increased investment of $30,000 will be $30,000/$l,800 = 17 years. Example 2 Situation: An owner of a 50,000 sq. ft. facility has a new roof installed for $250,000. A roof consultant approaches the owner to suggest that an inspection and maintenance pro¬ gram should be implemented to ensure that the 20-year life of the roof is realized. It is determined that it would cost $3,000/year to implement such a program. The consultant estimates that the consequence of not maintaining the roof is a reduced life expectancy from 20 to 15 years. The owner does not mind the reduced 5 years if it makes business sense to do so—i.e., if it is cost effective. Assuming a 10% discount rate, the consultant is required to determine the answer for the owner. Solution: Steps 1, 2, 4 and 5 have been completed in the above sit¬ uation. The study period is taken as 20 years to coincide with the life expectancy of the roof with preventative main¬ tenance. The next step is to discount future cash flow streams. The cash flow streams for the two alternatives are shown in a cash flow diagram. The BLCC program version 4.2 was used to complete step 5 and the results obtained are shown in Table 2. In this particular instance, there is no change in the ini¬ tial investment. However, if the roof is maintained at a cost of $3,000/year or approximately 1.2% of the initial cost, then preventative maintenance is cost effective. In fact, for the discount rate of 10%, preventative maintenance will be cost effective as long as the costs are below approximately $4,000/year or 1.6% of the initial cost. Calculations with a higher discount rate will tend to favor alternative A. Lower Table 2 Cost Category Alternative A No Maintenance Alternative B Preventative Maintenance Cost BCost A Initial Capital Cost $250,000 $250,000 $0 Present Value of Operating Costs $0 $25,541 $25,541 Present Value of Replacement Costs $59,848 $0 ($59,848) Present Value of Salvage ($24,786) $0 $24,786 Subtotal of PV of Costs and Salvage $35,062 $25,541 ($9,521) Total Life Cycle Cost $285,062 $275,541 ($9,521) 10 • Interface February 1997 than 15 years life for alternative A will favor alternative B. The above analysis does not account for any water leak¬ age incidence and associated costs. Such incidences can only strengthen the case for alternative B. It is possible to conclude from the above scenario that as long as the main¬ tenance costs are manageable to below 2%, alternative B will be more cost-effective. Example 3 Situation A roof consultant just completed a survey of a 50,000 sq. ft. facility for an owner. The consultant has determined that the 10-year-old roof needs repair and maintenance to realize a life of 10 more years without which it is difficult to say if it can even last 5 more years. The immediate repair costs are $10,000 and thereafter the maintenance costs are $5,000 per year. At the end of their lives, the roofs will be replaced with the same type of roof costing $250,000 and requiring similar levels of annual maintenance. The owner needs cost justification from the consultant for the recom¬ mended repair work based on a discount factor of 10% and a study period of 20 years. Solution: Steps 1 to 5 have been completed in the above situation. The next step is to discount future cash flow streams. The cash flow streams for the two alternatives are shown in a cash flow diagram. The BLGC program version 4.2 was used to complete step 5 and the results obtained are shown in Table 3. This example shows that the LCG of alternative A is Cash Flow – Example 3, Alternative A No Repair or Maintenance, Early Replacement 10% No Annual Maintenance Performed Repairs Needed But Not Performed Replacement Cost – $250,000 0 1 23456789 10. 15 – -’20 Years £ Salvage Value / -$100,000 Cash Flow – Example 3, Alternative B Repair and Preventative Maintenance With Longer Life Table 3 Cost Category Alternative A No Repair or Maintenance Alternative B Repair w/Preventative Maintenance Cost BCost A Initial Capital Cost $0 $10,000 $10,000 Present Value of Operating Costs $0 $42,568 $42,568 Present Value of Replacement Costs $155,230 $96,386 ($58,845) Present Value of Salvage — ($14,865) ($22,298) ($7,433) Subtotal of PV of Costs and Salvage $140,365 $116,656 ($23,709) Total Life Cycle Cost $140,365 $126,656 ($13,709) February 1997 Interface • 11 higher than alternative B by $13,709 and therefore it is more cost effective to carry out the repair and maintenance work as required. Note that the cost of maintenance work is estimated at 2% of the total replacement cost. The net present value savings from the operating cost are $23,709 for an initial investment in repair of $10,000. This results in an SIR of $23,709/10,000, or 2.4. This SIR is greater than 1, confirming the results of the LCG. The benefits of mainte¬ nance will be decreased if the prediction regarding the life expectancy without repair of 5 years is underestimated. A sensitivity analysis can be carried out to determine the vari¬ ations. Example 4 Situation: The director of the Parks and Recreation Department of the local municipality calls on a roof consultant. The direc¬ tor needs to know if they should install a 20-year warranty shingle or a 30-year warranty shingle on a small recreation facility. The cost of the 20-year warranty shingle material is $7,1 14 and the cost of 30-year warranty shingle is $10,968. A quick calculation by the consultant shows that the non-discounted cost per year of 20-year shingle versus 30-year shin¬ gle is $356/year versus $366/year. The 20-year shingle costs $10/year (or $200 over its 20 years) less than the 30-year shingle. For this small amount the owner would rather go with the 30-year shingle. The consultant is asked to deter¬ mine if discounting at a rate of 10% and assuming a study period of 30 years makes any substantial difference. Solution: Steps 1 to 5 have been completed in the above situation. The next step is to discount future cash flow streams. The cash flow streams for the two alternatives are shown in a cash flow diagram. The BLCC program version 4.2 was used to complete step 5 and the results obtained are shown in Table 4. This example shows that the LCC of the 20-year shingle is lower by $3,000 or 37% than the LCG of 30-year shingle. Based on the assumptions made, the 20-year shingle would be more cost-effective. The lowering of the discount rate will lower the savings, and vice-versa. Even for a discount rate of 5%, the 20-year shingle will be shown to have a lower cost. There are no other uncertainties that can practi¬ cally impact the above decision. References ASTM Standards on Building Economics Brigham, E. F., Kahl, A. L., Rentz, W. F., Gapenski, L. C., Canadian Financial Management, Holt, Reinhart and Winston of Canada Limited, Toronto, 1991. (pp 275 to 316) Desjarlais, A. O., Petrie, T. W., Christian, J. E., Mclain, H. A., Childs, P. W., “A Whole Building Demonstration of Re¬ Table 4 Cost Category Alternative A 20-yr. Shingle Alternative B 30-yr. Shingle Cost BCost A Initial Capital Cost $7,114 $10,968 $3,854 Present Value of Operating Costs $0 $0 $0 Present Value of Replacement Costs $1,057 $0 ($1,057) Present Value of Salvage ($204) $0 $204 Subtotal of PV of Costs and Salvage $854 $0 ($854) Total Life Cycle Cost $7,968 $10,968 $3,000 12 • Interface February 1997 Cover Over an Existing Wet Roof,” Proceedings of the 11th Conference on Roofing Technology, NIST/NRCA, Gaithersburg, Maryland, September 21-22, 1995 (pp 115-117) Griffin, C. W., and Fricklas, R., The Manual of Low Slope Roofing, McGraw-Hill, New York, 1995. (pp 33-34, and 388) Marshall, H. E. and Ruegg, R. T., Building Economics: Theory and Practice, Van Nostrand Reinhold, New York, 1990. Melvin, E., “Plan, Predict, Prevent: How To Reinvest in Public Buildings,” The APWA Research Foundation, May 1992. (pp 56 to 61) Riggs, J. L., Rentz, W. F., Kahl, A. L., West, T. M, Engineering Economics, McGraw Hill Ryerson, Scarborough, Ontario, 1986. Register for a Building Envelope Symposium February 20-21 in Hartford, CT October 23-24 in Dallas, TX November 20-21 in San Francisco/Oakland, CA About The Author Hitesh Doshi is a Building Science Educator, Researcher and Practicing Professional Engineer. He teaches building performance, including building economics, in the Department of Architectural Science and Landscape Architecture at the Ryerson Polytechnic University in Toronto, Ontario, Canada. Prior to joining Ryerson, Doshi was with Trow Consulting Engineers Limited, a large, multi-disciplinary engineering firm. He has written several articles, including a series of ten roof columns for Plant Engineering and Maintenance Magazine, where he was the contributing editor and continues to write for peer reviewed as well as trade publications. Doshi is the chair of the Seventh Building Science and Technology Conference to be held in Alarch 1991, and serves on the Education Services and Code Committees ofRCI. 200 Cottontail Lane Somerset, NJ 08873 When it comes to your business, rely on the company that’s committed to your success! HPG Roofing Systems HPG HPG Roofing Systems … Leaders in hot air welded roofing membrane technology. Single source for all your product needs Specification writing assistance CAD design support Wind engineering tests On-site crew training Extended warranty/service Tel: (908) 271-1300 (800) 457-6634 Fax: (908) 302-4220 oofing Sys February 1997 Interface • 13
