remove this checkout from your record. FINES wi11 ——— be charged if book is returned after the date stamped be10w. #7 M86 BEIURNING MATERIALS: Place in book drop to LIBRARIES —‘__- _________—-—-—-' THE U.S. DISTRICT HEATING INDUSTRY: A CASE STUDY OF CORPORATE STRATEGY AND PUBLIC UTILITY REGULATION BY Robert Loube A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Economics 1983 _—‘ _ .~ . 35*.) I‘Y’o‘I - ABSTRACT THE U.S. DISTRICT HEATING INDUSTRY: A CASE STUDY OF CORPORATE STRATEGY AND PUBLIC UTILITY REGULATION BY Robert Loube The U.S. district heating industry is in a state of decline. This dissertation examined the reasons for the industry's decline, the viability of new district heating projects, and explored whether and in what manner state regulation can be altered to change the state of the industry. To explain the history of the industry two theories of firm behavior -- profit maximizing and strategic satisficing -- were compared. Profit maximizing theory suggests that the industry decline was caused by a decrease in the demand for steam due to the substitution of natural gas for steam. A series of demand relationships was estimated in an attempt to test this hypothesis. From these estimations it was not possible to infer that steam and gas were considered sub- stitutes by steam customers. Alternatively, it was possible to amass evidence that shows that electric utilities used the heating subsidiaries as part of a strategy to establish regional electric monopolies. Steam was sold cheaply in order to discourage cogeneration by isolated producers of steam. Wherever the utility heating services expanded, isolated producers shut down. Once the heating subsidiaries accomplished their specified task, they disappeared from the planning tabloids of the electric utilities. Cost saving technologies were never exploited. Markets were never expanded. The vision of electric utility executives bounded by the rationality of the electric power process foreclosed the successful development of the heat industry. To test the viability of heating projects a simulation model was developed. The model selected the pipe sizes and lengths, determined costs and revenue, and calculated the net present value of the project. A base case was estimated. The base case results were compared to other cases by allowing the assumed values of selected variables to change. It was found that district heating projects were viable when reasonable variable values were used. It was recommended that state regulatory commissions encourage district heating due to the positive net present value associated with the projects. Commissions could encourage district heating by establishing an incentive rate of return scheme that would tie the allowed rate of return inversely to each utility's heat rate. ACKNOWLEDGEMENTS I would like to thank Professor Harry Trebing who kindled my interest in the study of public utilities. I could always count on him for an interesting suggestion and a kind word. I would also like to thank Warren Samuels and Norm Obst for their questions about the direction and pur- pose of this study. Ken Boyer provided valuable assistance with econometric work. I have been fortunate to know fellow graduate students, Shashi Gupta and Donald Stabile, with whom I could talk, argue and laugh. Shashi listened to my ideas, forced me to clarify them and helped me make them more precise. Donald could always offer an instantaneous alternative explanation to every new explanation I struggled to obtain. Here at James Madison University, I have found two new friends, Russ Smith and Dave Schirm, who have given freely of their time to read and discuss various drafts. I will never be able to repay the debt I owe my mother, father, brother and sister. Their faith in my ability kept me going through all the dark days when I thought I would never finish. Finally, I want to thank my wife, Robin Gordon, for all the moments we have spent together. Each one has been precious to me. ii TABLE OF LIST OF TABLES . . . . . LIST OF FIGURES . . . . . CHAPTER I. II. INTRODUCTION . . Methodology . . CONTENTS Profit Maximization vs. Bounded Rationality . . The Source and Measure Feasibility Study Institutions and District Heating: The European Experience . . . Regulatory Change REFERENCE . . . DISTRICT HEATING: Introduction . . Innovation and Development . . Company Activity Technology Regulation The Mature Industry Company Activity Technology iii THE U.S. EXPERIENCE of X-Inefficiency 10 12 14 16 20 22 28 28 30 30 38 42 46 46 52 Page CHAPTER Regulation . . . . . . . . . . . . . 59 Decline and Resurrection . . . . . . . . . 60 Company Activity . . . . . . . . . . 60 Natural Gas Competition . . . . . . . 65 Demand Curve Estimation . . . . 65 Model Specification . . . . . . 70 Additional Statistical Problems. 75 Results . . . . . . . . . . . . 78 Test for Stability . . . . . . . 82 Elasticities . . . . . . . . . . 83 Summary . . . . . . . . . . . . 84 Technology . . . . . . . . . . . . . 86 Regulation . . . . . . . . . . . . . 90 Regulatory Lag . . . . . . . . . 90 Re-examination of Steam Rates . 91 Federal Regulation . . . . . . . 101 .REFERENCE . . . . . . . . . . . . . . . . 144 III. DISTRICT HEATING: THE EUROPEAN EXPERIENCE . . . . . . . . . . . . . . . . 160 Introduction . . . . . . . . . . . . . . . 160 Sweden . . . . . . . . . . . . . . . . . . 161 The Development of District Heating . 161 Impact of District Heating on Sweden . . . . . . . . . . . . . . . 167 Institutional Setting . . . . . . . . 168 iv CHAPTER Rates . . . . . . . . . . . . . . . . 172 Finances . . . . . . . . . . . . . . 173 Energy Planning . . . . . . . . . . . 174 Denmark . . . . . . . . . . . . . . . . . 175 Development . . . . . . . . . . . . . 175 Rates . . . . . . . . . . . . . . . . 177 Institutional Setting . . . . . . . . 179 Finances . . . . . . . . . . . . . . 180 Energy Planning . . . . . . . . . . . 181 United Kingdom . . . . . . . . . . . . . . 184 Development . . . . . . . . . . . . . 184 Finance . . . . . . . . . . . . . . . 186 Rates . . . . . . . . . . . . . . . . 186 Institutional Setting . . . . . . . . 186 Energy Planning . . . . . . . . . . . 188 West Germany . . . . . . . . . . . . . . . 190 France . . . . . . . . . . . . . . . . . . 192 REFERENCE . . . . . . . . . . . . . . . . 205 IV. THE VIABILITY OF DISTRICT HEATING IN THE UNITED STATES . . . . . . . . . . . 211 Pipe Size Determination . . . . . . . . . 212 The Rate of Heat Transport (H) . . . . . . 214 Pipeline Length . . . . . . . . . . . . . 215 Change in Temperature . . . . . . . . . . 216 V CHAPTER Capital Cost . . . . Electricity Cost . . Total Pipeline Cost Net Present Value Determination Revenue . . . . . . Costs . . . . . . . Results . . . . . . Transmission Distanc Interest Rate . . . e Energy Inflation Rate Busbar Cost . . . . Pipeline Cost . . . Nominal Rates . . . Attraction Rate . . Summary . . . . . . Regulatory Practices Regulatory Reform . Summary . . . . . . REFERENCE . . . . . V. CONCLUSION . . . . . The Institutional Setting REFERENCE . . . . . APPENDIX C O O O O O O O O . BIBLIOGRAPHY . . . . . . . . vi 217 221 221 222 223 224 228 229 230 231 232 233 234 235 236 242 249 253 281 286 291 295 296 318 Table 10 11 12 13 14 15 16 17 LIST OF TABLES Economy of High Back Pressure Turbines Capital Costs for Dual Purpose Plant . Total Cost for Dual Purpose Plant . . Calculation of Steam Send-out Heat Rate For First 400,000 lb. per hr. at 2,000,000 kw. system load . . . . . Alternative Fuel Use Allocation Schemes Alternative Boiler Capacity Allocation Schemes . . . . . . . . . . . . . . . Total Gas Utility Sales . . . . . . . Natural Gas Transmission Pipeline . . List of Cities . . . . . . . . . . . . Study Group Cities . . . . . . . . . . Statistics for Study Group Cities 1929-1945 0 O O O O O O O O O O O O 0 Steam Sales of Study Group Cities . . Data Sources . . . . . . . . . . . . . Variables of the Model . . . . . . . . Demand Relationship: Single Equation MOdel O O I O O I O O O O O O O O O 0 Demand Relationship: Single Equation MOdel O O O O O O O O O O O O O O O 0 Demand Relationship: Two-Stage Estimation vii Page 103 106 107 109 110 112 113 113 114 114 115 116 117 118 120 121 122 Table 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Demand Relationship: Two-Stage Estimation . . . . . . . . . . . . . . Estimation of the Number of Customers Estimation of the Number of Customers Transformed . . . . . . . . . . . . Summary of Significant* Results for the Price of Steam . . . . . . . . . . Summary of Significant* Results for the Price of Gas . . . . . . . . . . Summary of Significant* Results for the Number of Customers . . . . . . . Summary of Significant* Results for Degree Days . . . . . . . . . . . Summary of Significant* Results for Retail Sales . . . . . . . . . . . . . Development of the Swedish District Heating Industry . . . . . . . . . . Average Annual Percentage Growth Rates for the Summary Statistics of the Swedish District Heating Industry . . . . . . Statistical Data on Some Member Towns of the Swedish District Heating Association for the Operating Year 1971-72 . . . . Existing Combined Power and Heating Plants 1975-07-01 . . . . . . . . . Survey of Danish Heating Supply Systems Pipeline Cost Per Foot of Dual Pipe (1980 DOllarS)* o o a o o o o o o o 0 Original Pipeline Cost (1980 Dollars) Population Density 30,000 Per Sq. Mile Population Density 20,000 Per Sq. Mile Population Density 10,000 Per Sq. Mile viii Page 123 124 125 126 127 128 129 130 193 194 195 198 199 255 256 257 259 261 Table 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 Net Present Value - Transmission Distant Net Present Value Interest Rate . . . Net Present Value Energy Inflation Rates 0 O O O O O O O O O O O I O O O O Busbar Cost . . . . Net Present Value Net Present Value Pipeline Cost . . . Nominal Rates . . . Net Present Value Net Present Value Attraction Rates . . City Population Rank and Density: 1975 The Relationship Between Feasibility and Sponsoring Agent . . . . . . . . . . Single Purpose Electric Utility Revenue Gas Sales . . . . . . . . . . . . . . . Heat Cost . . . . . . . . . . . . . . . District Heating Profits . . . . . . . . Profits of the Electric Utility . . . . The Detroit Edison Company Availability Incentive Provision . . . . . . . . . . Changes in Customers Served: All Cities, 1950-1978 . . . . . . . . . . . Model One: Ordinary Least Squares 1946-78 . . . . . . . . . . . . . . . . Model One: Ordinary Least Squares 1947-78 Using Cochrane-Orcutt Transformed Data . . . . . . . . . . . . Model Two: Ordinary Least Squares 1946-78 0 o o o o o o o o o o o o o o 0 Model Two: Ordinary Least Squares 1947-78 Using Cochrane-Orcutt Transformed Data . . . . . . . . . . . . ix Page 263 263 264 264 265 265 266 267 268 271 271 272 274 275 276 294 296 299 302 304 Table 56 57 58 59 60 61 Model Two: Generalized Least Squares 1946-78 0 o o o o o o o o o o o o o 0 Model Two: Generalized Least Squares 1947-78 Using Cochrane-Orcutt Transformed Data . . . . . . . . . . . Model Two: Ordinary Least Squares 1946-78 0 o o o o o o o o o o o o o 0 Model Two: Ordinary Least Squares 1947-72 Using Cochrane-Orcutt Transformed Data . . . . . . . . . . . Model Two: Generalized Least Squares 1946-72 . . . . . . . . . . . . . . . Model Two: Generalized Least Squares 1947-72 Using Cochrane-Orcutt Transformed Data . . . . .'. . . . . . 306 308 310 312 314 316 LIST OF FIGURES Page Figure 1 Comparative Urban Heat Load Density Values [MW(t)/km2] . . . . . . . . . . . . l7 2 Population Density, People/sq mi . . . . . 18 3 Steam Sales per Customer . . . . . . . . . 131 4 Steam Sales . . . . . . . . . . . . . . . 132 5 Capital Invested . . . . . . . . . . . . . 133 6 Customers Served . . . . . . . . . . . . . 134 7 Natural-Gas Statistics 1930 . . . . . . . 135 8 Natural-Gas Statistics December 1954 . . . 135 9 Natural Gas Pipelines - 1950 . . . . . . . 136 10 Major Natural Gas Pipelines as of June 30, 1974 . . . . . . . . . . . . . . 137 11 Cleveland . . . . . . . . . . . . . . . . 138 12 Denver . . . . . . . . . . . . . . . . . . 139 13 Detroit . . . . . . . . . . . . . . . . . 140 14 Rochester . . . . . . . . . . . . . . . . 141 15 Seattle . . . . . . . . . . . . . . . . . 142 16 Philadelphia . . . . . . . . . . . . . . . 143 17 Utilization Time Tv in Hours/Year . . . . 201 18 Specific Length 1 M/GWh/year . . . . . . 202 19 Efficiency n in % . . . . . . . . . . . . 203 xi Figure 20 21 22 23 24 25 26 Concentration of SO in the Air in some Swedish Towns. Insulated Pipe Insulated Pipe Insulated Pipe Insulated Pipe Insulated Pipe Febguary 1971 . . . . in Concrete Enclosure . in Steel Conduit . . . in Poured Concrete . in Plastic Conduit . . - Construction Completed Electric Utility Technology . . . . . xii Page 204 277 277 278 278 279 280 CHAPTER I INTRODUCTION The district heating process provides heat to a number of buildings from a single source. A central boiler produces steam. The steam itself can circulate to the buildings through buried pipes, or the steam can heat water in a heat exchanger, and the heated water can then circulate through the pipe system.' The boiler that provides steam for heating can also provide the steam required for generating electricity. In this case, steam passes through a turbine or engine before it is used for heating purposes. This process is called "cogeneration." Plants which produce both outputs are called "cogenerators" or "heat and power stations." When only one output is produced, the plant is called a "single purpose station." The district heating industry was founded in 1877. In the United States, it has passed through various stages of growth and decline. It is now stuck at a low level of activity. The number of companies providing the service dropped from 211 in 1962 to 95 in 1975,1 and sales of the leading firms fell from 73 million pounds of steam in 1972 to 61 million pounds in 1978.2 At present, the industry provides less than one percent of the space heating needs of the country;3 or equivalently, it meets the space heating needs of about 2.5 million people.4 This stagnation stands in stark contrast to the indus- try's potential here and to its experience in Europe. Two studies document the industry's potential in the United States. Karkheck et a1. estimated that if power plant heat were provided to district heating systems and its price set equal to the effective energy price of natural gas, then a population of 32 million individuals could be served profitably. They estimated if the price were equal to the effective energy price of imported oil, 73 million people could be served profitably.5 McDonald et a1. estimated that the conversion of 337 power plants to dual purpose facilities could supply 2 quads of heat (15 percent of U.S. residential and commercial space heating demand) to district heating systems. The power plants studied were located within 10 miles of the service center; the population of the service area was greater than 50,000; and population densities were greater than 1,000 persons per square mile. It was estimated that district heating would be competitive with gas in such service areas.6 The district heating industries of several Scandanavian and Eastern European countries illustrate the industry's potential. Denmark has the highest thermal energy capacity per capita, 2000 MW/million inhabitants. Approximately forty percent of Danish households are connected to a district heating system.7 In the Soviet Union, seventy percent of the urban heat demand is provided by district heating systems.8 Rationales for the existence of the gap between the U.S. industry's reality and its potential fall into two categories. One rationale asserts that profit maximizing firms, seeing their profits erode, left the industry. Those companies that remained did so because of a perceived political constraint against a shut-down of operations. The profit erosion was caused by consumer product substitu- tion when natural gas became available in the interstate market during the early fifties. The drop in consumption in the seventies was caused by consumer reaction to sharp rises in the price of steam. Finally the potential for expansion is non-existent, having been fabricated by researchers who made overly-optimistic assumptions about consumer response to new service and about future cost. ' The other rationale states that the history of the district heating industry has been an innocent pawn used and then discarded by the electric utilities in their drive for regional monopoly control. The electrics did not take advantage of opportunities to increase the profits of district heating subsidiaries, because these opportunities did not fit into their companies' self—vision. Instead, the electrics used district heating services as a loss leader to attract self-generators and potential self-generators of electricity. In so doing the demand for utility electricity increased and the load on the utility became more diversified. The increase load forced and enabled the electrics to invest in larger plants. The new plants that either embodied new technology or exhibited economies of scale lowered generation costs, allowing utilities to cover losses incurred by the district heating subsidiary and to attract more customers via lower prices. As soon as the electric utility established its regional monopoly it stopped promoting the district heating service. Once the service was no longer needed to fulfill an active part in the company's long-run strategy, the subsidiary fossilized. It did not take advantage of new markets. It did not actively seek rate increases needed to maintain profits. It did not adopt new cost-saving technology. It simply provided the same service to the same service area. This idea--that business executives form a vision of reality and then acted as if that vision were real even if the vision and reality differ--is encompassed by the theory of bounded rationality. This theory starts from the proposition that internal constraints on the decision maker are as important in the prediction of his or her behavior as external constraints.9 Internal constraints fall into three categories. First, there is the problem of uncertainty about the external constraints. The uncertainty forces the decision maker to follow behavioral rules of thumb that will not necessarily lead to a maximization result. Second, the decision maker has only incomplete information about alternatives. It is not possible to follow a maximizing rule ala Stiglerlo because the decision maker does not know the marginal benefit schedule associated with the unknown alternatives. Third, the complexity of the problem is so great that the decision maker can not determine the best course of action.11 External constraints are the set of constraints on which most economists choose to focus their attention. They are demand curves, cost curves,income and legal environment. Herbert Simon's complaint against maximizing theory is that it ignores the internal constraints and uses only the external ones. He believes it is necessary to use the theory of bounded rationality because "the capacity of the human mind for formulating and solving complex problems is very small compared with the size of problems whose solution is required for objectively rational behavior in the real 12 world"; or more succinctly--decision makers must "satis- fice because they don't have the wits to maximize."13 Instead of maximizing, the theory of bounded rationality holds that business executives follow one of two behavioral modes. First the decision maker could simplify the problem until it reaches a manageable form and then maximize the simplified problem.l4 Second, the decision maker could form an aspiration level that is a satisfactory outcome, then search alternatives until the alternative that ful- fills the aspiration level appears, and choose that alternative as the satisficing alternative. The decision to set the aspiration is an iterative process in which the ease and difficulty of finding the satisficing alternative influences the direction of the aspiration level. That is, if the satisficing alternative is easy to find, the aspira- tion level will probably rise; and if the satisficing alternative is impossible to find, the aspiration level will fall.15 Of course it is possible to mix the two behavioral modes. First, the decision maker can simplify the problem. Second, she can go through the process of setting aspiration levels and finding satisficing alternatives. To reinterpret the behavior of the electric utility executives in light of the theory of bounded rationality, it can be claimed that the executives simplified the problem of running their businesses by evaluating the entire company on the basis of the relationship of each part to the goal of building an electric power and light system.16 How and why this particular set of blinders became imbedded within the minds of the executives is beyond the scope of this paper. For this paper it is important only that the blinders exist and that their consequences are important. The most important consequence is that because of the choices made, there exists a large x-inefficiency. If the executives had perceived their companies to be energy trans- formation companies, then either prices of heat and electricity would be lower, or profits higher, or both. The gap between the reality and the potential of the dis— trict heating industry is an alternative measure of the x-inefficiency. The x-inefficiency was ignored in the United States for a long time because technological change, economies of scale, and cheap fuels had been reducing the relative prices of electricity and heat. Now that those relative prices are rising, it is time to change the regulatory environment to provide firms with a set of incentives that if followed would eliminate the x-inefficiency. The rest of this chapter is divided into six sections. The first section discusses the methodological problems involved in choosing between theories. The second compares the two theories. The third describes the sources of the possible x-inefficiency and its measures. The fourth details the uniqueness of this study and its applicability to major American cities. The fifth examines European practices and institutional forms. The sixth puts forth a recommendation designed to eliminate the x-inefficiency. Methodology Economists use a variety of criteria to evaluate theories. These criteria include simplicity,17 elegance, internal consistency, realism on assumptions, number of assumptions, explanatory power, and predictive power. The last, predictive power, came into the forefront with Milton Friedman's Essays in Positive Economics,18 and today still commands the greatest allegiance as the ultimate litmus test of a good theory as opposed to the other criteria.19 This criterion must be used with great care. Too often, economists commit the logical fallacy of affirming the consequent. For example, in the Lipsey-Steiner prin- ciples textbook, the following argument is made: One, utility theory predicts that demand curves are negatively sloped. Two, we find negatively sloped demand curves. Three, therefore utility theory is the correct description of human behavior.20 This argument is invalid. The evidence proves only that for a theory of human behavior to be acceptable it must be consistent with negatively sloped demand curves.21 To use the predicative criterion correctly, it is necessary to attempt to falsify the hypothesis. For example, a statement A implies B. We find that B is not true. Therefore we can safely assume that A is not true either. This argument is called modus tollens, and is a correct logical deduction.22 Of the two theories discussed above, the theory of profit maximizing is easier to test because it provides the researcher with definitive predictions. Even when the theory is broken into short-run and long-run maximizing, it is still testable. For example, while a firm that maximizes in the short run might have a different pricing strategy when compared to the firm that maximizes in the long run, neither firm would adopt a given production technique if a cheaper technique was available. Onthe other hand, it is not possible to falsify the theory of bounded rationality in general. A person attempt- ing to prove the superiority of this theory when faced by seemingly falsifying evidence can claim that the evidence compiled did not contain a true picture of the decision maker's original simplifying assumptions or search technique. Researchers claiming the superiority of the bounded ration- ality theory do so because the theory is more realistic in its behaviorial assumptions, it is more consistent with the evidence, and because it is possible to falsify and/or limit the applicability of profit maximization theories.23 10 'Profit Maximization vs. Bounded Rationality The theory of the profit maximizing firm predicts that a firm will always use the cheapest technique available to produce the desired output,24 that a firm will take advan— tage of new cost saving technologies when they become available,25 and that for a known demand and cost curves output will be adjusted to maximize profit.2 The history of the district heating industry contra- dicts those three predictions. First, it was always known that cogneration of electricity and steam was cheaper than producing either separately. However some firms never used cogneration facilities and others have discontinued its practice.27 Second, two techniques, the use of hot water as a distribution medium and the use of trash as a fuel have been demonstrated as superior techniques in Europe for a long period of time. Both techniques went through an innovation 28 Since the end of and development stage in the 1930's. World War II most new systems built in Europe use hot water transmission and distribution.29 In the United States, there are no companies using hot water. While it is true that the use of hot water would entail expensive retrofit costs for old systems, hybrid systems (mixtures of steam and water) could have been built.30 11 The number of private systems in the United States receiving steam from trash burners is less than five. There are several public systems that burn trash. Compared to the 243 European plants, the United States' statistic is quite low.31 Third, opportunities to expand district heating service areas existed during the fifties and sixties in localities undergoing major urban renewal projects. In only one instance, Hartford, Connecticut, was new service provided. Existing firms did not even estimate potential sales in 33 If a firm does not estimate revenue and these areas. costs for potential markets, then it cannot know that the profit maximizing behavior is to not offer the service. Profit maximizing theory also claims that firms will leave an industry if profits drop below a normal level. Profits can fall if either revenue drops or costs rise or both. The period since World War II reveals a period of low profits and existing firms. A hypothesis that explains this phenomenon is that profits fell due to revenue loss caused by shifting to natural gas and away from steam. By implication steam customers must have considered gas to be a substitute for steam, and shifted out of steam as the price of gas fell. In an effort to substantiate this hypothesis a large number of steam demand functions were estimated. These estimations provide little to no support for this hypothesis.34 32 12 On the other hand, evidence consistent with the hypothesis that electric utilities acted within the frame— work of the theory of bounded rationality is available. First, profits were sacrificed in an attempt to attract customers.35 Second, a large number of independent boilers were shut down and some independent steam companies were forced out of business.36 Third, the assumption that business leaders were building electric systems can be confirmed by their statements and actions.37 Fourth, failure to adopt new technologies or seek new markets shows that the companies did not examine all possible avenues to increased profits.38 To juxtapose the evidence and theory in the above fashion does not prove the correctness of the theory of bounded rationality. It merely shows that the theory is adequate to the task of explaining the history of the industry. When combined with the falsification of profit maximization theory, the evidence lends credence to the alterative theory.39 The Source and Measure of X-Inefficiency Electric power can be generated in either a single purpose or dual purpose (also known as combined heat and power, CHP, or cogenerating) power plant. The single pur- pose plant burns fuel in a boiler to generate steam. The 13 steam passes through a turbine and then into a condenser. The steam spins the turbine and it is this motion that generates electricity. The fuel efficiency of this system depends on the temperature difference of the steam when it enters and leaves the turbine. In a coal burning plant under ideal conditions steam can enter at 1000° F. If these conditions are met then the plant will operate at 40 percent fuel efficiency. Two percent of the energy will be lost to mechanical inefficiencies. Ten percent will go up the stack, and 48 percent of the energy will be dissipated into the atmosphere.40 The dual purpose power plant attempts to capture the 48 percent that goes into the atmosphere and transform it into a saleable commodity. In order to do so the outlet temperature of the steam must be raised to at least 250° F. This change sacrifices electricity generation which drops to 30 percent of the energy consumed. However, useful heat, 58 percent of the energy input, can be captured for sale. The remaining 12 percent of the energy is lost due to the mechanical inefficiency and stack losses. The fuel effi- ciency of the dual purpose plant is 88 percent (30 elect, plus 58 heat).41 The transformation of the single purpose plant into a dual purpose does not depend on fuel efficiency alone, but it is the fuel efficiency that creates the cost saving that 14 allows for the transformation. To make this decision it is necessary to compare the cost of alternative energy supply systems. It will be shown that, under certain conditions, the dual purpose power plant with its accompanying heat supply system is the least cost energy supply system. The cost difference between the dual purpose system and existing systems is the measure of the x-inefficiency. Feasibility_Study The feasibility of serving a hypothetical residential community via district heating was examined. Revenues were limited to be below the cost of alternative heat delivery systems. District heat costs were estimated under a variety of different conditions. The unique features of this study were its use of pre-insulated pipe in the distribution net- work and the low heat density pattern of the service area. Pre-insulated pipe has become the dominant form of pipe construction in Europe. This type of pipe was first used in the early 1960's. By 1975 it represented 50 percent of new pipe construction.42 In only one other study evaluating United States conditions, at Piqua, Ohio, was pre-insulated pipe used in the distribution network.43 In Piqua, conditions advantageous to successful district heating exist. First, the power plant that will provide heat to the system is located 1% miles from the service area.44 Second, a 15 hospital, industrial customers and several schools are located in the service area.45 These customers generally have high demands and flat load curves.46 These character- istics allow the system to take advantage of economies of scale in piping without encountering peak load problems.47 Because of these unique conditions it is difficult to generalize from the Piqua experience. The present study eliminates these unique features. By doing so, it will be able to come to some general conclusions related to the use of prefabricated pipe. The heat densities of the service areas examined were 13, 16, and 19 megawatts per square kilometer (mw/kmz). These densities are below the density, 20 mw/kmz, usually considered necessary for profitable district heating.48 Figure One related heat densities to urban living patterns. Note that the densities in this study would be no higher than the level: residential area with two-family houses. The heat densities of 13, 16 and 19 mw/km2 are equiva- lent to population densities of 10,000, 20,000 and 30,000 inhabitants per square mile, respectively. Population densities of major American cities are shown in Figure Two. The cities that appear in Figure Two were considered possible candidates for new district heating systems by the Karkheck study.49 The densities are city-wide averages. Most of the cities fall within the 10,000 to 30,000 inhabitants per square mile range. However, district heating has never been 16 proposed as the sole heat supply source for any of these cities except for New York. One study estimated that district heating could profitably serve 33 percent of Los Angeles, 45 percent of Baton Rouge, and 83 percent of 50 Jersey City. The service areas recommended included only the most dense areas of the above cities. Institutions and District Heating: The European Experience The section addresses the question, does the institu- tional framework of a society affect the percent of any nation's heating needs met by a district heating service? The answer, according to McIntrye and Thorton, and Lucas, is yes.51 McIntrye and Thorton compare centralized decision- making economies (in particular the U.S.S.R.) to market economies. They argue that centralized decision-makers have the opportunity and the incentive to reduce trans- actions cost inherent in providing district heating;52 and that the benefits from the reduction of environmental pollution associated with district heating will have a greater impact on centralized decision-makers than on decentralized ones.53 The transactions cost identified by McIntrye and Thorton includes the "need to persuade potential customers 17 MW/km2 ______ .71_ MOST DENSE PART OF ’7—400 MINNEAPOLIS AREA > 200 AROUND IDS BUILDING _j_100 REGIONS BEST SUITED FOR DISTRICT 1 DOWNTOWN MINNEAPOLIS > HEATING _1__50 (220 MW/kmz) DOWNTOWN ST. PAUL —J——25 RESIDENTIAL AREA WITH-—————9’ _JL_ TWO-FAMILY HOUSES —~——12.5 SINGLE-FAMILY RESIDENT (1/3 acre lot) > Figure 1 Comparative Urban Heat Load Density Values [MW(t)/km2154 PIICIITA“ N non: RESIN“ II AMFIIIYS 18 f'T—rerTII F V t IIITI] '00 manna to» 10 :- CHICAGO unluuu\‘. / w I. I VIII NI. YORK 1 I“: cum: ”"00““ no >- L0! mom: 1 I‘umou cm MING"? , “38" CITY .0 - «mat " 0 IN! {unvmrou :3 » utth:0~.//§u'°”“‘ “wan“: to» 5mmmuumu .0 ’{ unamnuo Inmuwmu .L‘.,LL111111 m 300” Figure 2 Population Density, People/sq miSS 19 to purchase heat from a district-heating network and to coordinate decisions with the electric utility if a heat 56 The central decision and power station is involved." maker can mandate customer hook-up to the system. This policy will reduce the average cost to each customer because it spreads the burden Of the fixed cost to a larger number. A private producer cannot be assured customer acceptance of a product. The private producer must incur the costs of persuading his customers. Plus the private system will be saddled with either negative profits or high rates or both in its formative years if it cannot attract a large number of customers.57 The dual purpose power plant must be integrated with both the heat delivery system and the electric grid. If one organization is responsible for the integration, losses due to coordination inefficiencies can be minimized.58 District heating reduces air and thermal pollution because less fuel is burned, and the fuel is burned under conditions where emissions are controlled. The benefits Of the reduced pollution will more likely be of concern of the central decision maker than to a private utility because total benefit is large while the benefit to any individual is small.59 Lucas examines district heating in several Western European countries. He concludes that "the degree Of local government participation in electric supply is closely 20 correlated with CHP (combined heat and power)."60 When the local communities do not control electric supply then these utilities use strategies of "dynamic conservatism" to thwart the development of district heating.61 Such strategies include excluding CHP projects from national electric grids, offering to buy power at rates below the cost Of utility alternatives, and selling gas at high rates.62 The details of these strategies and a description of other institutional factors are provided in the chapter on the European . 63 experience. Regulatory Change A primary objective of public utility regulation is to promote efficiency. The existence of x-inefficiencies defy this regulatory standard. State Regulatory Commissions have taken steps to eliminate some x-inefficiency. A regulatory scheme similar to the Michigan plan would provide the incentive to eliminate the x-inefficiency associated with single power plants. The Michigan plan includes a variable allowed rate of return that is triggered by a plant availability factor. If plant availability is above a certain target the allowed rate rises and if plant availability is below another tar- 64 get the allowed rate falls. To transform this scheme so that it is relevant to the district heating case, all that 21 needs to be done is to change the target from plant avail- ability to fuel cycle efficiency. Firms that convert single-purpose plants into dual-purpose plants and sell the heat will increase their fuel cycle efficiency and thus be allowed to earn higher profits. A constraint that electric rates can be no higher than the rates would have been if the electric customers were served by single-purpose plants must be imposed. Otherwise utilities could set up ineffi- ciency district heating systems, subsidize them with electric service revenues and increase company profits. The elimination of the x-inefficiency generates a net benefit stream that can be shared by the company and its customers. Profits can rise. The prices Of heat and electricity can fall. REFERENCE CHAPTER I 1"Public and Private Sellers of Heat in the United States and Canada," District Heating, April 1962, pp. 155- 57; "District Energy Suppliers in the United States and Canada," District Heating, January 1977, pp. 6-8. 2See p. 62 below. 3Richard L. McGraw, "Keynote Address," Proceedings of Sixty-ninth Conference of the International District Heating Association (Pittsburgh: International District Heating Association, 1978), p. 20. 4J. Karkheck et al., "Prospects for District Heating in the United States," Science, 11 March 1977, p. 950. SIbid., p. 952. 6Craig L. McDonald et al., An Analysis of Secondary Impacts of District Heating from Existing Power Plants (Richland, Washington: Battelle Pacific Northwest Labora- tories, [1977]). p. 4. 7Martin A. Broders, Potential for District Heating: An Historical Overview (Oak Ridge, Tennessee: Oak Ridge National Laboratory [1981]), p. 29 81bid. 9Herbert Simon, "Theories Of Bounded Rationality," in Decision and Organization, ed. C. Radner and R. Radner (Amsterdam: North-Holland Publishing Company, 1972), p. 162. 10George Stigler, "The Economics of Information," Journal of Political Economy 69(June 1961), p. 213. 22 23 11 Simon, "Theories of Bounded Rationality," pp. 162- 163. 12Herbert Simon, Models of Man (New York: John Wiley and Sons, 1957), p. 198. l3Herbert Simon, Administrative Behavior (New York: MacMillan, 1947), p. xxvii. l4Simon, "Theories of Bounded Rationality," p. 170. 15Ibid., p. 168. 16Thomas P. Hughes, "The Electrification Of America: The System Builders," Technology and Culture 20(January 1979), pp. 124-161; Jack Riley, Carolina Power and Light Company: 1908-1958 (Raliegh: Carolina Power and Light Company, 1958); Raymond Miller, Kilowatts at Work (Detroit: Wayne State University Press, 1957). 17In response to the accusation that his theory was too complicated, Simon said "But Occam's Razor has a double edge. Succinctness of statement is not the only measure of a theory's simplicity. Occam understood his rule as recom— mending theories that make no more assumptions than necessary to account for the phenomena (Essentia non scent multiplicanta procter necessitatem). A theory of profit or utility maximization can be stated more briefly than a satisficing theory of the sort I shall discuss later. But the former makes much stronger assumptions than the later about the human cognitive system. Hence, in the case before us, the two edges of the razor cut in Opposite directions." Simon, "Rational Decision Making in Business Organizations," American Economic Review 69(September 1979), p. 495. Also when Reb Reuben Ben Naftala was asked about the importance of Occam's Razor, he scratched his beard and responded: A researcher using Occam's Razor to make decisions regarding a solution to a complex problem will probably leave a stubble; while the Lord, blessed be he, Often hides the truth in the subtle stubble. 18Milton Friedman, Essays in Positive Economics (Chicago: University of Chicago Press, 1953), pp. 8-9. l9Martin Blaug, The Methodology of Economics or How Economists Explain (Cambridge: Cambridge University Press, 1980), p. 128. 24 20Richard Lipsey and Peter Steiner, Economics, 6th ed. (New York: Harper and Row, Publishers, 1981), pp. 160- 163. 21Charles W. Kegley and Jacquelyn Ann Kegley, Intro- duction to Logic (Columbus: Charles E. Merrill Publishing Company, 1978), p. 409. 221bid., p. 408. 23Simon, "Rational Decision Making in Business Organ- izations," American Economic Review 69(September 1979), pp. 496-498; Herbert Simon, "Rationality as Process and Product of Thought," American Economic Review 68(May 1978), p. 9. 24George Stigler, "The Xistence of X-efficiency," Ameri- can Economic Review 66(March 1976), pp. 213-216. 25The relationship between market structure and tech- nological innovation and adoption has long been debated; see F. M. Scherer, Industrial Market Structure and Economic Performance, 2nd ed. (Chicago: Rand McNally College Pub- lishing Company, 1980), pp. 407-438. 26James M. Henderson and Richard E. Quandt, Micro- economic Theory, 3rd ed. (New York: McGraw-Hill Book Company, 1980), p. 97. 27See p. 52 below. 28Paul Geiringer, High Temperature WateriHeating (New York: John Wiley and Sons, Inc., 1963), pp. 7-8; D. J. DeRenzo, ed., European Technology for Obtaining Energy from Solid Waste (Park Ridge, N.J.: Noyes Data Corporation, 1978), P. 224. 29See p. 161 below. 30Thomas Root, Senior Engineer, Detroit Edison Company, interviewed at his office, April 1979. 31DeRenzo, European Technology for Obtaining Energy from Solid Waste, p. 224. 32See p. 62 below. 25 33George Urbancik, Senior Engineer, Baltimore Gas and Electric Company, interviewed at his office, February 1980. 34See p. 84 below. 35See p. 37 below. 36See p. 49 below. 37See p. 53 below. 38Thomas Root, Senior Engineer, Detroit Edison Company, interviewed at his Office, April 1979. 39The method of explanation used in this paper has been called contextual validation; see Charles Wilber with Robert Harrison, "The Methodological Basis of Institutional Economics: Pattern Model, Storytelling and Holism," Journal of Economic Issues 12(March 1978), pp. 61-90. 40Philip Hill, Power Generation (Cambridge: MIT Press, 1977). p. 85. 4lIbid., p. 87. 42Walter L. Mikkelsen, "Development of District Heating in Denmark," in Total Energy: Proceedings of the First International Total Energy Congress, ed. Eric Jeffs (Copen- hagen, 1976), p. 762. 43City of Piqua, District Heating and Cooling for Communities Through Power Plant Retrofit and Distribution Network, prepared for the U.S. Department of Energy Con— tract NO. EM-78-C-02—4976 (1979), P. 336. 441bid., p. 163. 45Ibid., p. 164. 46Heating and Ventilating Research Association, District Heating: A Survey of Current Practice in Eurgpe and America (London: National Coal Board, 1967), p. 16. 26 471. Oliker, "Economic Feasibility of District Heating from Coal-fired Power Plants," American Power Conference 43(1981). p. 1065. 48Broders, Potential for District Heating, p. 10. 49Karkheck et al., "Prospects for District Heating," p. 36. 50 Ibid., p. 41. 51Robert McIntyre and James Thorton, "Urban Design and Energy Utilization: A Comparative Analysis of Soviet Practice," Journal of Comparative Economics 2(December 1978), pp. 334-354; N. J. D. Lucas, "CHP and the Fuel Indus- tries," in Combined Heat and Power, ed. W. Orchard and A. Sherratt (London: George Godwin Limited, 1980), pp. 59- 78. 52McIntrye and Thornton, "Urban Design and Energy Utilization," p. 336. 53Ibia., p. 341. 54Broders, Potential for District Heating, p. 11. 55 p. 95. Karkheck et al., "Prospects for District Heating," 56McIntrye and Thornton, "Urban Design and Energy Utilization," pp. 339-340. 57Ibid. 531bid., p. 342. 591bid., p. 344. 60Lucas, "CHP and the Fuel Industries, p. 60. 61The term, dynamic conservatism, was used by Mans Lonnroth to describe the attempts of the Swedish private power producers to thwart the development of district heat- ing via cogeneration; see Mans Lonnroth, Peter Stern and 27 61 cont'Thomas Johansson, Energy in Transition (StOCk- holm: Secretariat for Future Studies, 1977), p. 104. 62Lucas, "CHP and the Fuel Industries," pp. 62-67. 63See p. 168-172 below. 64Harry Trebing, "Motivations and Barriers to Superior Performance Under Public Utility Regulation," in Produc- tivity Measurement in Regulated Industries, ed. Thomas Cowing and Rodney Stevenson (New York: Academic Press, 1981), p. 390. CHAPTER II DISTRICT HEATING: THE U.S. EXPERIENCE Introduction The history Of the district heating industry can be divided into three periods: innovation and development, the mature industry, and decline and possible resurrection. Each period is marked by a similar group of problems and accomplishments. During the first period the industry's technological feasibility was proven. Economically the link between the district heating and the electric utilities was firmly established. The electric utilities used their district heating subsidiaries as pawns in their strategy to establish dominance in particular geographic service areas. The mature period is marked by a large growth in sales and customers. Institutionally the industry remains linked to the electric utilities. The electric utilities ignored possible cost savings inherent in cogeneration as they strove to increase efficiency in electric generation rather than overall energy efficiency. 28 29 The third period, the decline and possible resurrec- tion, is marked by a loss of customers and the bankruptcy of many companies. The decline might have been caused by the availability Of natural gas. This investigator doubts that hypothesis. Evidence is provided that decline occurred while the price of natural gas was still greater than the price of steam. An alternative explanation for decline is that the electric utilities simply ignored their small subsidiaries. The utilities failed to bring rate cases to maintain the cash flow of the subsidiary. They did not investigate possible expansion related to urban renewal projects. As the urban centers of the northeast and midwest declined the electric utilities allowed the district heating systems to shrink and disappear. The resurrection Of the industry in the late seventies was the work of the federal government. In its desire to save energy on a national level, the federal government provided the research funds to investigate the energy saving potential of the industry. These studies highlight the energy savings and possible profits ignored by the electric utilities. However, the federal involvement almost disappeared in fiscal 1982; and it is not clear how long the remaining programs will last. 30 Innovation and Development CompanyiActivity In 1877, following experiments in which he heated his own home and his neighbors from a boiler in his basement, Birdsill Holly founded the first district heating system, Holly Steam Combination Company, in Lockport, NY. The company originally served fourteen customers. Holly's plant and equipment consisted of one boiler (7 feet in diameter and 10 feet high) and 2350 feet Of iron pipe. The pipes were buried in wooden boxes with wood shavings used as insulation. The steam was distributed at a pressure of 30 PSIG. Later, the company laid additional lines to connect several factories to the system. The steam pressure in the new lines was 80 PSIG.l Imitators of the Holly systems appeared immediately. Within the next ten years, district heating companies served at least 19 additional cities.2 The imitators fell into three categories: first, companies that sold steam or hot water primarily for the profit that could be earned in the space heat and industrial heat business; second, univer- sities that tried to reduce the cost of heating a group of buildings; third, companies that combined the sale of heat with the sale of electricity in an attempt to increase the profits Of the electric business. 31 The most important imitator of the Holly system was the New York Steam Company, which Wallace Andrews founded in 1878. He secured permission from the state to operate in 1879, the same year in which he obtained permission to use the Holly patents.3 Wallace Andrews fits the stereotype of an American entrepreneur of the period, innovative and risk taking. He was on the first board of directors of the Standard Oil Company.4' He developed the first coal slurry pipeline, securing a patent for the idea in 1891.5 He was president of the Standard Gas Light Company, which he sold to the Consolidated Gas Company.6 Andrews financed the fledgling New York Steam Company over its first three years by selling Standard Oil Stock at the rate of $1000 a day.7 New York Steam laid its first pipe in 1881. Simul— taneously, a competing company, the American Steam Company, entered the picture. The rivalry grew intense with each company striving to be the first to provide steam service. However, the American Steam Company disappeared on the morning it tested its mains. A leak developed in the mains allowing the steam to mix with the insulating material. This material, 3 million tons of lampblack, blew up, creat- ing one million Al Jolsons in black face. The New York Steam Company, on the other hand, used mineral wool as an insulating material. Mineral wool proved to be acceptable . . 8 as Insulation. 32 New York Steam began service on March 3, 1882.9 By 1886 the company had 350 customers, with 5 miles of main.10 The company developed two separate service areas: first, a downtown area including Wall Street and most of the finan- cial district; second, a midtown district originally serving a residential market, but later, in the 20's and 30's, when the area developed into Office buildings, serving Rockefeller Center and the Empire State Building. The company sold steam to residential customers as a premium fue1--c1ean, fire proof and automatic. Advertising copy included an endorsement by John D. Rockefeller: "I have had my house heated for several seasons by steam sup- plied by your company, and am satisfied with the service 11 given." Early customers included H. O. Armour, William Rockefeller and the Metropolitan Club.12 Besides selling steam for heating purposes, the New York company also sold steam for industrial uses. Its early customers included United States Illuminating Company, and Consolidated Gas. United States Illuminating was a com- petitor and later a subsidiary of New York Edison. Consoli- dated Gas used the steam in the production of town gas.13 It is a curious footnote in history that these two customers later merge to become Consolidated Edison, and that Consoli- dated Edison now owns New York Steam. Other companies Or electric utility subsidiaries were formed to provide heat in Boston, Rochester, Eugene, and 33 Washington, D.C. The National Superheated Water Corpora- tion, sponsored by Theodore Vail, served Boston from 1887 until it closed down in 1905. It was the first company to use water as a heat transmission vehicle, the technique of water heat transmission having been patented by E. Pratt in 1878. The water was heated to temperatures of 334°F. A plunger pump forced the water through the pipe system. At the consumer end, the water was decompressed, releasing heat. A suction pump drew the water back to the central station.14 Rochester Gas and Electric entered the heating business in 1889. Its initial system heated a residential district. In 1898 a second plant was built next to the Eastman Kodak plant and the Bausch and Lamp Optical Company. This heating plant, by 1920, served 84 factories with 664 x 1061bs. of steam annually.15 In Eugene, Oregon, in 1910, a group Of sawmill Opera- tors formed the Central Heating Company, which used sawdust as its primary fuel. The company was tied to a system run by the Fruit Growers Association. The interconnect was beneficial to both entities because the canning season and heating season peaked at different times of the year.16 In Washington, D.C., two systems were established during this period. In 1905, Pepco contracted with the Navy to supply steam to the Navy Yard.17 In 1910, the federal government built a system to service federal 34 buildings on Capitol Hill including the U.S. Capital, the Library Congress, the Botanical Gardens and House and Senate Office buildings.18 Universities also entered the district heating industry at this time, producing steam for their own consumption, not for sale. In many instances, they produced steam in conjunction with electricity. Princeton produced electricity as early as 1880. Steam was exhausted from diesel engines to the heating system. By 1903 most of the academic buildings were connected to the steam system.19 The University of Michigan began heating with steam in 1894, enlarging its system in 1915.20 Other electrical utilities constituted the third group interested in district heating during this period. These companies provided heat as part Of a corporate strategy to eliminate self-producers of electricity within service areas, and to develop a full line of services necessitated by utility competition for customers and service area. Detroit Edison was an example of a company facing both problems. In 1903, the Murphy Power Company, not affiliated with Detroit Edison, was established to sell steam in down- town Detroit. It also had the ability to produce electric- ity. Murphy's problem was that the diurnal peak for steam demand occurred in the morning while the electricity demand peaked in the late afternoon or early evening. Thus, in the morning, it had excess electricity, and in the evening it had excess steam. The solution to the morning problem 35 was to sell electricity to Detroit Edison. This contract gave Detroit Edison control over Murphy Power Company. Murphy had no option to sell to anyone else because Detroit Edison refused to wheel Murphy's electricity to other electric companies.21 Detroit Edison entered into direct competition with Murphy by organizing a steam company, Central Heating. This company began Operations in 1903 with 12 customers along 3000 feet of mains. Detroit Edison used the heating company to attract as customers of its electric business, self-producers and/or potential self-producers of electri- city. At that time many large Office buildings and commer- cial establishments had boilers for heating that could be converted to provide steam for the generation Of electricity. By Offering steam to those consumers, Detroit Edison reduced the probability of building owners' converting to self- generators. The steam service allowed building owners to reduce costs and to increase the space available for revenue-producing activities.22 While profit data are not available from this period, conjecture based on company announcements leads one to conclude that steam was sold below cost. Detroit Edison used below cost pricing to drive the Murphy Power Company out of business (at which time Detroit Edison purchased the assets). Detroit Edison clearly felt that additional 36 profits earned in the electric business would more than offset steam losses.23 In Philadelphia, while the facts are not as clear, the trail of events was essentially the same. As Of 1895 there were twenty electric companies serving Philadelphia. In that year, Pennsylvania Heat, Light and Power Company, which used Siemans-Halske patents, initiated a steam heating system.24 Edison Electric immediately responded by founding the Philadelphia Steam Heating and Power Company. Until then Edison Electric had supplied steam to only one building adjacent to its plant and to its'company Offices.25 Later, Philadelphia Electric sold steam and electricity as an integrated package to downtown stores. A different pattern developed in St. Louis. There, Union Electric did not face competition from any other electric companies. Still, it wanted to eliminate self- generators. To do so, starting in 1905, it leased and Operated boilers in its service area. Each boiler served adjacent buildings. Pipes were laid to connect the boilers into a system. By 1909 it operated 23 plants serving 58 buildings. The load grew over time so that the company placed into operation a large central boiler in 1917.26 In 1922, at the end of this era, Boston Edison started a system. Its announced purpose was to gather in electric customers. At first it leased nine boilers from downtown department stores. These boilers were connected by a pipe 37 system. New customers were added along the pipe system. Boston Edison also purchased the Boston and Main RR steam electric system, in which steam was used to heat railroad cars. Electricity was produced as a by-product of the steam. The railroad remained a major steam customer and its electricity demand was now served by the larger Boston Edison electric facilities.27 The number of steam utilities was estimated to be 150 in 1910 by the NDHA.28 However, an industry specialist estimated that, in 1905, there existed 250 steam plants and 75 water plants.29 This estimate may have included multi- plant firms. Another industry analyst estimated that between 300 and 400 companies existed in 1915.30 The profit picture is even less clear. In 1918, following sharp increases in coal prices, the largest 31 Most of company, New York Steam, filed for bankruptcy. the other companies were tied to electric companies. In those cases there are acknowledgements that steam was deliberately sold below cost, so that profit figures, even if they existed, would be meaningless.32 Only one fact is definite: Companies were started and enlarged. Thus there must have been individuals who believed that district heat was a potentially viable industry. 38 Technology Technological problems for district heating producers during this period centered on three issues: metering, distribution, and the type of transmission medium to use. Developing an accurate meter was essential. Charging customers for steam without one involved complicated formulas that sometimes contained perverse incentives. Distribution problems involved the need to effectively seal joints and to prevent steam flow blockage in pipes. The transmission medium problem involved choosing either steam or water as the preferred heat carried. Birdsill Holly patented a steam flow meter for use in his original system. However, this meter proved to be inaccurate.33 Instead of using a meter, district heating companies estimated energy use through a variety of tech- niques. The company would then charge a flat rate for the building based on the estimated energy use. The estimation techniques used a combination of factors such as cubic feet of building space, square feet of exposed surface, square feet Of windows, number of doors, and square feet of radiator surface.34 Perverse incentives entered this system in two ways. First, while small radiators lead to lower bills, a smaller radiator will draw more steam than a large radiator for heating the same space.35 Second, if the rate were set for an entire season and not according to energy use, then a 39 building owner had no incentive to conserve steam by placing controls on radiators or his heat exchanges. As long as the district heating sent steam through the line the building used it. If some occupants became hot, windows were Opened.36 In 1904 the American District Heating Company patented a condensate meter. This meter measured the consensed water as it left the customer's heat exchanger. This device proved to be accurate. The cost Of the new meter was $10 per year. Savings ran as high as a twenty-five percent reduction in steam use.37 Distribution problems centered on two areas, the first of which was sealing pipe connection. New York Steam started with cast iron flanges that were sealed with gas- kets, then bolted together. The change was successful.38 The second problem in distribution centered on con- densate within a steam line. If the condensate was not removed from the pipe, the water build up would eventually block the distribution system. Early systems worked on a gravity flow principle. All pipes ran downhill from customers to the plant. The steam under pressure would flow uphill, while the water would run back to the plant. Obviously, unless highly gratuitous circumstance existed this approach led to major engineering and excavating difficulties that significantly increased the cost of the distribution system. 40 The introduction of the steam trap solved this prob- lem. The trap mechanism allows water to leave the pipe system without allowing any steam to escape. It consists of a portal placed at the base of a pipe. A lever con- trolling the portal is attached to a flotation device. As the water in the pipe rises, the float rises, Opening the portal and allowing the water to drain out. The water level drops, causing the flotation device to drop, closing the portal. In principle, the portal closes before all Of the water drains so that the remaining water blocks the release Of steam.39 Since heat can be transmitted via either water or steam each company had the choice of medium. Each fluid has its own unique properties that under a given set of circumstances would make it the preferred transmission fluid. At the beginning of the twentieth century the cir- cumstances favoring steam were prevalent. The factors that favor steam are listed below. Later, the factors favoring water will be discussed. First, if the condensate is not returned to the boiler plant the energy of the condensate is lost to the heat system, but this loss is much smaller for steam systems than for water systems due to the fact that the energy per pound Of condensate is less than the energy per pound Of water leaving a customer's heat exchanger. This factor led many early heating companies to build steam systems without 41 return pipes, saving capital.40 This strategy would be Optimal if it coincided with two other factors: (a) the availability of a heat sink into which the condensate could be dumped. In most instances, the city sewer system was used. And (b) either the price Of fuel must be low relative to the price of capital, or due to capital market imperfec- tions which restricted the ability of heating companies to raise funds, the price Of fuel must be low relative to capital to the heat companies. Second, station equipment is lower for steam heat. Steam circulates through the pipes under its own pressure. Water systems need pumps to force circulation.41 Third, steam easily rises in tall buildings due to its inherent pressure. To raise water, pressure must be added.42 Fourth, steam can be used in a variety of industrial processes. Hot water must be converted back to steam for these processes. Many of the processes require heat ranges of greater than 250°F. Hot water distribution at these temperatures had proven to be extremely difficult. It was not until the late 20's that high hot water temperature systems were successful.43 Fifth, customers could control steam flow within their buildings. At the turn of the century, customer control of the water flow was expensive both to install and maintain.44 42 Regulation This era saw the transformation of public utility regulation in the United States. Changes occurred in the substance Of regulation; in the level of government responsible for regulation; and in the type of institution responsible for implementing the regulations. Prior to this period, the substance of regulation was limited to the granting of a franchise. The franchise allowed the company to do business for a specific period of time, and to construct needed facilities along or under pub- lic thoroughfares, and sometimes granted the right of eminent domain. The new regulatory format provided the state with the right to circumscribe the business activities of the utilities on an ongoing basis. The state can set prices, determine profits and supervise the sale of corpor- ate securities. The authority tO regulate public utilities moved from the muncipalities to the state governments. At the munici- pal level, elected mayors and city councils exercised regulatory authority. On the state level, the authority was delegated to an independent commission, a new institu- tional form, whose only task was to regulate utilities. Between 1907 and 1913, 29 states established utility commissions.45 Advocates of the commission system argued that these commissions, through the use of scientific expertise, would 43 take public utilities out Of the political arena and thereby lead to good government. However, when the advocates are scrutinized a little more closely, the picture grows murky.46 The utilities, ever wary of government control, became major advocates of state regulation. By following this policy, utility executives felt that they could obtain the right kind of regulation before the wrong kind was thrust upon them. This position was rational, given that the executives saw themselves caught between the scylla of municipal franchise and the charybdis of municipal owner- ship.47 Utility executives perceived two dangers in the munici— pal franchise format. First, the system was inherently corrupt. For example, in Chicago, the city councilors established dummy utility corporations. These corporations were sold to the established utilities. If the utility did not buy the dummy corporation, its franchise which was granted for only two years, might not be renewed.48 In Philadelphia, the city councilmen granted the Pennsylvania Electric Light Company (in which the councilmen owned stock) the right to own conduit under the city streets. Edison Electric Light Company of Philadelphia was not given this right. It had to rent the right-Of-way from Pennsyl- vania Electric.49 44 Second, the corruption which was exposed became the catalyst for public ownership. Reform mayors such as Tom Johnson of Cleveland, Samual (Golden Rule) James of Toledo, and Hazen S. (Potato Patch) Pingree Of Detroit pressed hard for municipal ownership. Tom Johnson believed in: municipal ownership of all public service monopo- lies for the same reason that I believe in the municipal ownership of waterworks, of parks, of schools. I believe in the municipal ownership of these monopolies because if you do not own them they will in time own you. They will rule your politics, corrupt your institutions and finally destroy your liberties. Chicago in 1887 and Detroit in 1889 established city corporations to generate electricity for street lighting purposes.51 These municipal corporations deprived the utilities of their largest customers. In Cleveland, Tom Johnson tried to start a municipal lighting company. Cleveland Electric Illuminating Company pressured(and bribed) the city council to vote against the bond authority Johnson needed to finance a municipal company. Next, Johnson proposed a special election on the bond issue. Cleveland Electric Illuminating Company's lawyers obtained a court injunction forbidding the election.52 The city of Cleveland's municipal company was finally established when Cleveland annexed South Brooklyn. This suburb already owned a plant, and its plants formed the foundation for the city's system. While Cleveland Illumin- ating fought the annexation, there was little it could do 45 after the annexation passed. Court action was impossible because the courts were ruling that municipalities had the right to own and Operate electricity systems.53 The impact of the new regulatory format on steam heating companies varied from state to state. In New York, the New York Steam Company came under the regulatory authority Of the commission in 1913. The next year the Public Service Commission required the company to rebuild some of its mains.54 In 1915, the Commission forced the company to install meters for customers. In 1918, the Commission requested that the company set promotional rates for high load customers, and in 1918, the Commission granted the company the nation's first fuel adjustment clause.55 In 1918, the Public Service Commission of Indiana set standards of service for hot water systems and specified a fair rate setting procedure. The standards included: first, that the company must supply hot water from October 1 through April 30, whenever the outside temperature is below 60°F; second, that the temperature Of the supply of water must be at least 85°, 154°, and 184°F when the outside temperature is below 60°, 30°, 0°F respectively;56 third, that customers pay the company in seven installments during the heating seasons; and fourth, that the customers must weather strip his doors and windows.57 Rates were set in proportion to heat demand. The commission provided each company with a specific formula to 46 use in estimating heat demand. The variables in this formula were cubic feet of the building, square feet of the windows, square feet of exposed walls, and square feet Of doors.58 The more typical situation existed in Michigan. In 1909, the Michigan Railroad Commission was given the following set of powers to regulate utilities: 1. It approved utility securities Offerings. 2. It required public filing of rates. 3. Upon appeal from a city government, it could set maximum rates. During this early period of regulation, the commis- sion's only action was to require companies to publish rates. This requirement led to a reduction in price discrimination, which in turn seemed to mollify public demand for regulation. The Mature Industry Company Activity Events outside the industry, the commercial building boom in the late twenties, the depression and World War II, provided the incentives for change during this period. The district heating industry altered its course with each change in the economic environment. The industry expanded to meet the growing demand of the late twenties. It pros- pered early in the depression due to the lagged impact Of 47 the building boom and cost reductions. It stagnated late in the depression. Finally it supplied large increases in output during World War II. The commercial building boom extended for the years 1926 to 1929. In each year the value of real commercial construction was greater than $3 billion in 1958 dollars. This $3 billion mark was not reached again until 1955.60 Simultaneously, sales of district heating companies increased rapidly. Twenty-two companies, that consistently reported sales from 1925 to 1929, sustained an 8% average annual growth rate for those years. The performance Of individual companies varied. Companies that began the period with high sales volume such as Detroit, Rochester, and Milwaukee featured growth rates Of 9.4, 8.6, and 7.0 percent annually respectively. Companies that began the period with low sales volumes expanded more rapidly in percentage terms. Pittsburgh sustained a 17.2 percent annual growth rate, while Boston achieved a 25 percent annual growth rate.61 In 1925, Rochester Gas and Electric started its down- town commercial system. That year it sold 56 million pounds of steam to 35 commercial customers. By 1929, it was selling 365 million pounds of steam to 154 commercial customers.62 In 1926, Boston Edison revised its business strategy. Prior to that year the sole purpose of the district heating 48 subsidiary was to attract electric customers. Starting in 1926, Boston Edison attempted to make its district heating subsidiary profitable in its own right. It promoted the sale Of steam based on three advantages steam heat would provide customers. These advantages were: first, a reduction in capital expenses; second, the elimination of jobs associated with individual building heating systems; and third, the elimination of coal handling problems.63 An analysis of the New York Steam Company in 1926 showed a viable company. It had lowered the price of steam from an average $1.11 per 1000 pounds in 1922 to 95 cents per 1000 pounds in 1926.64 Increased boiler efficiency, more steam sold per pound of coal burned, was the prime cause of this reduction. The overall energy efficiency of the company was 56 percent. This efficiency was based on a 75 percent boiler efficiency, an 81 percent distribution efficiency and a 92 percent customer heating exchange efficiency.65 The rate of return for 1926 determined as the sum of profits after tax plus interest divided by capital investment was 9.3 percent.66 The early years of the depression did not slow the growth of the industry. In fact to some observers it seemed that the industry was depression proof. From 1929 to 1933 an analysis of eleven cities (the cities were New York, Detroit, Milwaukee, Boston, Rochester, Dayton, Philadelphia, Pittsburgh, St. Louis, Lansing, and Baltimore. These cities 49 were chosen because they were the only cities to complete the statistic surveys Of NDHA in every year under Observa- tion. A comparison of the eleven cities to all reporting cities is given in Figure 3. This comparison shows that the eleven cities on average have customers who have high steam demands relative to all reporting cities showed: one, an increase in steam sales of 23 percent or 5.4 percent at average annual rate; two, an increase in capital invested of 20.6 percent or 4.8 percent at an average annual rate; three, an increase in maximum hourly capacity Of 74 percent 67 I These Increases or 19.3 percent an average annual rate. were achieved in the face Of an economy whose real gross national product declined by 30.5 percent.68 On an individual company basis the increases were equally remarkable. The New York Steam Company doubled its sales from 1927 to 1932.69 Baltimore Gas and Electric purchased Terminal Heating and Freezing in 1927. It increas- ed steam sales in Baltimore by 300 percent from 1928 to 1931. In its service area one hundred sixty-five indepen- dent steam generating plants shut down. No new plants were built.70 In St. Louis Union Electric's customer load rose from 164 in 1928 to 304 in 1932. It laid 10% miles of pipe in 1931.71 From 1929 to 1933 profit indicators rose. For the 11 cities previously cited, average coal cost fell by 15.4 percent from 14.98 cents per 106 BTUs to 12.65 cents per 50 106 BTUs while average revenue fell by only 11.0 percent from 87 cents per M lbs. to 77 cents per M lbs. Given a 13,000 BTU average per pound of coal, and that in 1929-- 12 pounds of steam were sold per pound Of coal burned and in l933--14 pounds of steam were sold per pound of coal burned, then the cost Of coal per 1000 lbs. of steam fell from 16 cents in 1929 to 12 cents in 1933. Steam sales rose from 1690240 M lbs. to 20864552 M lbs. Thus revenue minus coal expenses rose from $12 million to $13.5 million or by 12.5 percent.72 Labor costs were also falling. A survey Of six cities taken by the NDHA showed that total hourly labor costs dropped from $20,000 to $16,500 from 1930 to 1933. This decline was due to decreases in both the number of employees and average hourly wages.73 The indicators featured only the relationship between average revenue and average variable costs. They showed that cost on an average annual rate (6.6 percent for labor and 4.2 percent for coal) were declining faster than average revenue (3.0 percent).74 Data detailing depreciation and interest costs were not available. An additional indicator of profit, actual or potential, would be the market's willingness to purchase a company's debt. In March 1932, the New York Steam Company Offered an $8.7 million bond issue. It was oversubscribed.75 51 The effects Of the depression began to hit the study group cities in 1934. Steam sales leveled out and remained constant through 1939 (the drop in 1938 was due to unusually warm weather).76 Companies responded to the lack of growth in sales by slowing the growth in capital invested. For two years, 1938 and 1939, the value of the capital invested actually declined.77 Average revenue was constant from 1934 to 1937 then declined in 1938 and again in 1939. Coal costs per BTU increased but due to increased boiler efficiencies coal cost per pound of steam sold remained constant.78 The total hourly wage bill was constant as wage increases were offset by employment reductions.78 This review of price and cost trends suggests that profits probably were stable from 1934 to 1937 then fell dramatically in 1938 due to both the decline in sales and average revenue. In 1939, sales returned to their previous levels but average revenue declined so that profits prob- ably remained low.79 This dismal picture of the industry in the late thirties was broken by the impact of World War II. Sales jumped from 21.8 x 109 lbs. in 1939 to 30.4 x 109 lbs. in 1945. Contributing to the sales increase were increases in both the number of customers connected to steam systems and the average amount of steam purchased by each customer.80 52 The profit picture also improved. Large increases in coal cost per BTU were offset by increases in average reve- nue and plant efficiencies leaving revenue after coal cost Cons“:ant at .67 cents per M lb. for the years 1941 to 1945. Mllltiplying this net revenue figure by the large increases in Sales must have pushed profits up.81 Technology The choice of plant type--cogeneration versus steam- Orlley boilers--became the focus of technological decision- making in this period. The trend was away from cogeneration tc>Vvard steam-only boilers. In 1925, fifty-one percent of the companies sold exhaust steam. (Cogeneration plants p3'=‘<>crr commercial reasons rather than for production savings. "VNTGB closed many a profitable electric contract," he said, "t:¥f1at could not have been obtained unless we had been able tlc>' furnish steam service at the same time."86 Support for cogeneration came from individuals closely tLied to the district heating industry. In two reports, fj~3=7£st in 1926 and second in 1948, the research committee of t . . . . . . flee: National District Heating AssOCIation documented the a . (i‘°”islntages of cogeneration. These reports concluded as an electric utility grows a limited capacity in turbo-generators, Operating non-condensing [therefore supplying steam to a district heating system] and housed in strategic locations [will result in] mutual economic advantage to the electric and heating utilities. The research committee reports focused on two key lsssties. First the committee demonstrated the ability of a cogeneration facility to produce more revenue per fuel input 54 than a single purpose facility.88 Second, the committee showed that this initial advantage could be maintained when the cogeneration facility was integrated into the steam system and electric grid.89 In the 1926 report, the analysis started from the con- ditions of steam as it enters the turbine. The steam pressure was 650 PSIG and had a temperature of 700°F.90 At these conditions, each pound of steam contains 1165 BTU/lbs. or 1,000 lbs. contains 1,165,000 BTUs. A cogeneration facility that exhausts steam at a pressure of 85 PSIA can transform the 1,000 pounds Of steam in 40 Kwh of electricity. The 1,000 pounds of exhaust steam contains 952 BTU/1b. Approximately 76,000 BTUs, 7.6 percent of the energy input, was lost in heat radiation Of turbine inefficiencies. Total revenue, at 8 mills/Kwh and $1.00 per 1,000 of steam equaled $1.32 per 1,000 lbs. of steam input.91 A single purpose plant, Operating under the same input conditions and exhausting steam into a condenser at a 29" vacuum pressure, would have generated 103 Kwh electricity, the energy equivalent of 352,000 BTUs. Conversion losses will be 21,000 BTUst Heat loss to the condenser was 892,000 BTUs, representing 68% of the heat in the input steam. Total revenue, at 8 mills/Kwh, equaled .83 cents per 1,000 lbs. of steam input.92 The advantage achieved by the cogeneration plant can only be realized if the steam demand is large enough to 55 insure base load use of the plant. To examine this problem, the 1926 report set up a hypothetical steam system. The yearly energy demand of the system was 4.25 million lbs. of steam with a peak load of 1.9 million lbs. per hour.93 The 1926 report compared the cost of supplying steam from a cogeneration facility with a steam-only plant/ electricity-only plant configuration to this hypothetical steam system. The cogeneration facility not only produced steam but also had 30,000 Kwh electricity capacity. The turbines Operated at an annual load factor of 48.6 percent and a 64 percent load factor during the heating system. Steam passing through the turbines provided 75 percent of the heat energy demand and 39 percent of the peak capacity. The remainder of the demand was carried by low pressure boilers. These boilers were Operated at an annual load factor Of 10 percent.94 The cost comparison made the following assumptions: First: electricity supplied from single purpose plants cost 8.90 mills per Kwh (Line 34, Table 1). Second, steam supplied from single purpose plants could be profitably sold for $1.00 m lbs. (Line 32, Table l). The electricity generated at cogeneration facilities should be charges only for those costs needed to transform a single purpose steam plant into a cogeneration facility. For example, the cost of electricity included the difference between the cost Of 56 a high pressure boiler and a low pressure boiler, where the boiler is needed for cogeneration and the low pressure boiler could fill the steam demand needs (see Tables 2 and 3). Cost calculations based in these assumptions show that electricity can be produced at a cost Of 3.62 mills per Kwh with a heat rate of 4,530 BTU/Kwh. (There are slight dif- ferences between Tables 1 and 2 due to rounding errors.) Total savings due to cogeneration were $675,000 annually (Line 36, Table 1). If this amount was used to reduce the price of steam then that price would decline by 15.88 cents per m lbs. (Line 38, Table 1). Alternatively, the amount could be divided between steam customers, electricity customers and stockholders with each group receiving its prorationed amount. The 1948 research committee report examined the existing Consolidated Edison system. This system included two cogen- eration facilities. The cogeneration plants contain 31 percent of the steam system's capacity while providing 80.2 percent of its energy needs.95 The 1948 report emphasizes the relationship of the cogeneration facility to other electric generating plants. The electric system Operates on the basis of economic loading or incremental heat rates. That is, plants with low heat rates carry the base load. As demand increases, additional plants are brought on line in order of ever 57 increasing heat rates.96 Consolidated Edison's major cogeneration facility, Waterside Station, had a heat rate, operating as a single purpose electricity generator, of 11,946 BTU/Kwh.97 When the Waterside Facility Operates as a cogeneration facility, its output can be altered in any of three ways: first, electricity output can be held constant and steam output increased; second, both steam and electricity output can be increased; third, electricity output can decline while steam output increases (see Table 4). If the system electric demand is such that the incre- mental plant has a heat rate greater than the Waterside plant (if less than Waterside, then Waterside should not be supplying electricity using this method of operation), then the steam output will be responsible for changes in the Operating rates of other electric plants. Thus the cost of steam is determined incrementally as the sum of the incremental heat needed to generate electri- city at other facilities, given that the electricity generation at the cogeneration facility changes from electricity only output level plus the additional heat input needed at the cogeneration facility to produce steam.98 Using the above steam costing method, the cost of steam from the Waterside plant ranges from 647 to 1,117 BTUs/lb.99 Given that the best live steam boilers use 1,512 BTUs to produce a pound of steam, the report shows that on an S8 incremental heat using basis, cogeneration is superior to live steam production.100 These two NDHA reports conflict with the conventional wisdom as expressed by Alex Dow, and implemented by many companies. Reconciliation of these conflicts can occur only on a non-economic plan. Executives of electric utilities, protected from competition by regulators, competed for honors and prestige along alternative (non- profit maximizing) routes. For instance,"in the opinion of knowledgeable observers, such rivalry for technological advance existed between AEP and Philadelphia Electric, whose Presidents, Philip Sporn and R. G. Rincliffe, both were intent on advancing plant thermal efficiencies by increasing operating pressures and temperatures."101 This rivalry led to a focus on large condensing plants while closing off alternative visions of plant configuration. Professional prestige prevailed over economic rationale. This particular conclusion can only be the speculation of the author. However, as John Stuart Mill observed: It would be a great misconception of the actual course of human affairs, to suppose that com- petition exercises in fact this unlimited sway. I am not speaking of monopolies, either natural or artificial, or any interference of authority with liberty of production or exchange...I speak of cases in which there is nothing to restrain competing...yet in which the result is not determined by competition, but by custom or usage. 59 Electrical utilities have monopolies either natural or artificial. Given these monopolies, the probability must increase that action follows custom and usage rather than cost minimization. Regulation A 1933 NDHA survey questioned NDHA's membership on the existence and extent of regulation. Forty-two company replies were published. Of these forty-two companies, only three were privately owned and not regulated. Two companies were municipally owned. Twenty-one were regulated by state commissions only. Eleven weresubject to municipal regula- tion only, and five were regulated by both state and municipal governments.103 All of the regulated companies had to file rates and submit annual reports. Most of these companies were also required to file rules and regulations pertaining to customer and/or utility obligations.104 However, the existence of regulation does not guarantee regulatory supervision. The review process of commissions is contingent upon the utility filing a rate case. During this period, very few utilities filed such cases. For example, from the time Sioux City, Iowa, system started, in 1918, until 1959, the only rate changes that occurred 105 were changes triggered by a fuel adjustment clause. The New York Steam Company did not file for a rate increase 60 from 1928 to December 1946 (the increase was granted in 6 September 1949);10 and Rochester Gas and Electric did not file for an increase in its steam rates from 1933 until 1951.107 Decline and Resurrection Company Activity The period after World War II began optimistically. In the new District Heating Handbook the authors state that "the future looks promising and it appears that the industry has entered into a period of healthy expansion, sound operation and financial 'stability'."108 In July 1952, an editorial in District Heating announced that "heating companies are profitable" and that "competition is not a problem."109 However, the industry atmosphere soon turned sour. The number of customers served by the eleven study group cities peaked in 1954. By 1978, 2271 fewer customers were being served than in 1946, representing a drop of 23 percent.110 For the entire industry, 116 of 211 surveyed companies folded between 1962 and 1975.111 A survey of the remaining companies showed that their profits were often low or negative.112 The alleged culprit responsible for the demise of the district heating is natural gas. The use of this fuel 61 irlczreased at an average annual rate of 9.9 percent in the 113 The rapid expansion fjxrst decade following World War II. gram. The remaining three sites are at different stages Off completion. The University of Minnesota system heating Twas been completed. The electrical units have not been put in place yet. The Clark University system has announced that it would be completed in 1982. The Trenton system has yet to sell its construction bonds. It does have its heat customers lined up and a take or pay contract with the local electric utility to purchase cogenerated electricity.123 The second program featured the retrofit of existing power plants, transforming the plant into a cogeneration facility. The plant would then be linked to the expansion of an existing district heating system. In most instances the planned expansion would more than double the size of the existing system.124 Eight test-sites were chosen. Only one, at St. Paul, Minnesota, has moved beyond the planning stage. In St. Paul, a new non-profit institution, the District Heating Corporation of St. Paul, was established. This corporation purchased the old district heating system from the local utility. It has purchased additional boilers; and has customers signed up to take the additional heat. The new pipeline system will use hot water as the heat carrier.125 Whether or not the other projects will be completed is problematic. The federal government has pulled out of the 65 Errkogram. The funding level for the community system program irl fiscal year 1982 budget is zero.126 Natural Gas Competition The impact of natural gas on the demand for steam was estimated for twenty-one different cities. Each city was considered a separate market. For fourteen cities, the estimation includes data from the years 1946 to 1978. For the other seven cities data could be collected for only a subperiod of the above interval. Two model specifications were used. These models will be compared below. Statisti- cal problems encountered will be analyzed. A summary of the results will be provided. The purpose of this exercise is to estimate the size of the price elasticity of steam and the cross elasticity of steam with respect to the price of gas. Demand Curve Estimation. Estimations of demand curves that test for interfuel substitutions fall into two categories. The first group is derived from the theory of consumer demand and second for the theory of the firm. When consumer theory is the basis for the estimation, it is assumed that a consumer purchases fuel in order to use a given stock of equipment. The demand for fuel becomes a function of the demand for equipment and the equipment utilization rate. The demand for equipment is a function of equipment prices, income, prices of the particular fuel 66 aJnci its substitutes, along with other variables (usually dennographic characteristics and/or housing stock data). 'Phe utilization rate is assumed to be a function of the price of the particular fuel, income, and variables related to use (for example degree days, or percent of homes that are all electric).127 The demand curve for the fuel becomes 01 = f[Pil le PEI Y! X] (2’1) where Qi = quantity of the particular fuel Pi = price of the particular fuel PS = prices of substitute fuels PE = prices of equipment Y = income X = all other variables When the equation is estimated equipment prices are usually ignored. Thus the estimated equation is Qi = g [PiIPSIYIX] (2‘2) These models have been criticized because they did not take into consideration the process through which the equip- 128 A solution to ment holdings adjusted to price changes. this problem, suggested by Nerlove129 and applied by Houthakker and Taylor,130 is to include a specific adjust- ment process. In particular the desired quantity is a function of the set of variables so that Qi* = h[PiIPsIYIX] (2-3) 67 wileere Qi* is the desired quantity. The desired quantity caxnnot be measured. The decision maker approaches the <1esired amount through changing existing purchases. However the actual change will not be as great as the desired change because of the time needed to make the transformation. This process can be depicted by the following equation: Qit - Qit-l = A(Qit ' Qit-l) (2’4) 0 < A i 1 where Qit - Q actual change in the quantity it-l Qit - Qit-l = desire change in the quantity 1 adjustment coefficient By combining equation (2-3) and (2-4) an equation of the form Q- = JIPirPspQ 1t X] (2-5) it-l’Y’ is arrived. This process will provide an estimation of the adjustment.131 An alternative solution is to develop a new dependent variable, called the quantity of new demand. This variable was suggested by Balestra and used by Balestra,132 MacAvoy and Pindyck,133 and Berndt and Watkins.134 The new demand is the sum of the incremental demand plus the replacement demand. The incremental is defined as .AQi = Qit - Qit-l' To find the replacement, demand fuel use is assumed to be a function of the existing equipment stock. If the utilization rate is constant then 68 Qit = A E (2-6) Wflnere A is the utilization rate and E is the existing stock. 'If the existing stock of equipment depreciation at a con— stant rate (r) then a given year (t) an amount of demand will exist that could have been transferred to another fuel by the comsumer, the replacement demand, which is XrE or t-l 135 r Qit-l' The new demand variable, int' is NQit = AQi + rQit-l (2-7) This variable is placed into a demand equation such as: NQit = f(Pi'PS'Y’X) ‘ (2-8) The estimations, using equation (2-8), have been made for residential natural gas demand. The additional informa- tion needed to perform this task, depreciation rates, are not always available. The alternative method of demand curve estimation is derived from the theory of firm. Recent practice starts with a translog cost function.136 This function is trans- formed into a set of input cost share equations.\ Each cost share equation is estimated. The coefficient values estimated can be manipulated to determine the own price elasticity and the cross elasticity of demand for the inputs.137 This technique was recently used by Halvorsen138 to estimate the own price elasticities and cross elasticities of demand for electricity, oil, coal and gas for each 69 <>ff twenty two-digit industries for the year 1971. He aéssumes the existence of a production function for each tnno-digit industry in every state. Next he assumes that the production function is weakly separable between energy inputs and all other inputs. Here separability means that the rate of technical substitution between any pair of energy inputs is not affected by the quantities of non- 139 This assumption allows Halvorsen energy inputs used. to estimate an energy cost function in terms of an energy input and the prices of the different forms of energy ("... energy cost function, W = J(Z, PE' PO, PG' PC)")140 where W is total energy cost, Z is an energy input, and PE' PO' PG, PC’ are the prices of all input. This method reduced the total data needs and circumvents the tricky problem of defining the price of capital. However there are at least two problems with this short cut. First coal needs coal handling equipment and number six fuel needs to be kept warm if it is to be used. The implication of these production relationships is that the rate of technical substitution among energy inputs depends directly on the amount of capital employed. Second, energy forms have multidimensional chemical properties. Along one of these dimensions, the amount of BTUs contained per unit of account, it is possible to aggregate energy across the various forms. The purchase of energy is not made solely on the basis of BTU content. Other factors, such as sulfur’ 7O conter1t2, dust content, viscousity, and volatility are imporjlaant determinants of energy use. The implication of energy '3 multiple dimensionality is that a variable Z called energy}? input is not definable. Diodel Specification. The consumption of district heatLjrng services usually takes place in office buildings, largyee apartments, schools, hospitals and government build- in955.. Estimations of fuel demand for this group, the (xnmnmeercial class, have always used equations derived from the 1:heory of consumer demand.141 The equation forms eSthnnate the market demand due to the inclusion of the nmmIDEBr of customers as an independent variable. Residential fUE¥L demand, also derived from the theory of consumer denuarld, is generally estimated on an individual or per caFfiitza basis. The rationale for this difference is that wh1143 residential estimates are attempts to understand the b‘aha‘lior of a typical consumer, there is no typical consumer in the commercial class.142 In particular two models, similar in form to equation 2-2' Vvill be estimated. The first equation is (.25t = bo + blPst + bZPgt + b3NCt + b4DDt + bSYRt (2-9) Wher‘a Qs = the quantity of steam sold P9 = price of natural gas NC = number of customers Pst= price of steam These 71 DD degree days YR a time trend variables were chosen because: 3L. price of natural gas: managers of district heating firms considered gas as their customers best alternative. degree days and number of customers: the statistical reports of the International Districting Heating Association often refer to these variables when they provide reasons for changes in sales. Time trend: this variable is a proxy for changes in the business conditions within the utility service area. Both prices were deflated by the GNP deflator. This Process insured that the demand curve estimates would reflect Changes in relative prices rather than changes in nominal priCeS. If the prices had not been deflated then the eStLiDilation technique would have correlated increases in the PriJZees of steam over time with increases in the quantity 0“- Steam purchases over time. The result would be a posi- the coefficient for the price of steam in all regressions. In this particular case, the GNP deflator was used to deflate the prices because neither the consumer price index run: the wholesale price index contain prices for commodities 72 sold to commercial customers. All other prices used in this study will be adjusted using the same technique. IFurther the number of customers does not necessarily have: 'to be an exogenous variable. It could also be the concitiit through which the price of gas effects the demand for’ ssteam. That is while a change in the price of gas might not. eeffect the demand for steam a change in the expected Price of gas will change the number of customers that any steam utility serves. To test this hypothesis a two stage estimation tech- nique was used. In the first stage the number of customers was estimated as a function of the expected prices of steam and gas and a piecewise time trend with a mode in 1955. The proxy used for the expected price was the price lagged one year . The piecewise time trend was added in this form to test the hypothesis that business activity in the utility service area declined from the mid-fifties to the present. It Cannot explain why the service areas were not expanded to follow the shifting trends in business and population. Alternatively, 1955 marks the approximate completion of the interstate natural gas pipeline system. If the important Vanti-able related to natural gas is its availability and not its jpmice then the time trend could also be responding to gas"availability. The availability argument makes sense When gas is compared to coal or oil because a user of gas has lower storage, capital, and labor cost, and less 73 pollution problems than users of other fossil fuels. How— ever, ‘these advantages do not exist when gas is compared to steanl- In fact users' labor, capital, and insurance costs are gaeanerally assumed to be lower for steam than for gas.143 the equation estimated was = _ * Iflct bO + bng + b2PS + b3YRt + b4(YRt YR )Dl t—l t-l (2-10) whelrea NC = the number of customers P = the expected price of gas P = the expected price of steam YR = the time trend YR* = 55 D. = 1 YR. YR* 1 1 D. = 0 YR. YR* 1 1 In the second stage of the predicted value of the num- ber- fo customers was inserted into the demand equation in Place of the actual amount providing A Qst = b0 + blpst + b2Pgt + b3NCt + b4DDt + bSYRt (2-11) Where A NC = the predicted number of customers, and all other variables retain the identification provided with equation (2-9). 74 The model specification embodied in equations (2-9) and ( 2-11) can be criticized along at least two lines. First , the yearly time trend is an imprecise proxy for increase in real income. Further by using the same proxy for every city, the estimation technique suggests that changes in income follow the same pattern in all cities. Second, the price of steam is not necessarily exogenous. This assumption relies on the fact that the price was set bY regulation prior to the purchase decision. However, the existence of declining block rates connects the price to an endogenous variable, and with ‘it the possibility of incon- sis tent estimates.144 Both of these criticisms have been taken into account in the second model. Here the proxy for income was retail sales of each individual city. The retail sales data were Obtained from the Census of Business. Census data were available for the years 1948, 1954, 1958, 1963, 1967, 1972, and 1977. Predicted values were inserted as data points for nonrCensus years. To eliminate the problem of inconsistent estimators caused by the price structure a two stage estimation tech- nique was employed. In the first stage the price of steam is Estimated as a function of the price of the fuel input to each utility and the rate of interest faced by the iminstry . 75 PSt = bO + blet + b2rt (2-12) where "U ll price of steam Pf = price of fuel H II Moody's AA bond rate for public utilities minus the percentage change in the GNP deflator. In. tzhe second stage the quantity of steam was estimated as a ifllnction of the price of gas, the predicted price of stseainh the number of customers, degree days and real retail sales. A QSt — bO + blPSt + bZPgt + b3NCt + b4DDt + bSRSt (2-13) where A PS = the predicted price of steam R8 = retail sales divided by the GNP deflator All other variables retain the identification provided with equation (2-9). .Additional Statistical Problems. Three additional problems in regression analysis were encountered in the estlimitation. 'TheSe were the possibility of autocorrelation of the disturbances, of contemporaneous correlation of the dist:‘urbances across equations, and of a misspecification of themodel due to changes in the legal environment. Each 76 problem has particular causes and consequences for the estinieition results. frhe presence of autocorrelation of disturbances implies that; 'the error terms for a given observation is related to the earror term of the preceding observation. This phenome- rmul :is common in time series analysis because what happened last year usually effects what happens this year. The consequence of autocorrelation is that the variance of the Parameters will be biased leading to the statistical acczesptance (given positive autocorrelation) of parameter estimates that should be rejected. The test for the existence of autocorrelation is ilrlprecise. A statistic is calculated from the residuals 0f the regressions. This statistic is compared to a set Of Standardized statistics. The standardized statistics determine a three part region: occurrence, uncertain occurrence and non-occurrence of autocorrelation. The regression statistic, known as the Durbin-Watson statistic, £9141 into the occurrence region for most of the model one regressions, and into the uncertain region for most of the mod£3]_ two regressions.145 While all of the regressions wer‘e statistically transformed via the Cochrane—Orcutt mettuodl46 in an effort to correct for autocorrelation one is Ilot sure if the correction of the model two regressions generated parameter estimates that are more efficient than the estimates generated by the untransformed data.l47 77 C2<3ntemporaneous correlation of the disturbances occurs when the residual in one city for a given year is correlated with the residual for another city for the same year. For instance, if it is unseemingly cold in Cleveland it will probably be unseemingly cold in Akron, Detroit and Toledo also - Thus, the residuals for the individual markets that seem unrelated are actually related.148 To correct for this correlation one can use generalized least squares technique on the seemingly unrelated equations. The tech- nique transforms the parameter estimates by using informa- tion contained in the variance—covariance matrix. The res‘lllting transformation will provide more efficient eStirnates than the ordinary least square estimates.149 The misspecification due a change in the legal environment was associated with a change in the air pollu- tion laws in December 1970.150 Following the passage of this law some heating companies switched fuel inputs from coal to gas. Thus, gas not only effects the demand for district heating, but also its supply. If this is true then it is no longer possible to esti“late the demand for steam via the two-equation model preSented here. The process designed to eliminate the Simultaneous equation bias adds multicollinearity to the equation due to the fact that the predicted estimators of the price of steam will be correlated with the price of 78 natural gas. To avoid this dilemma, it might become necessary to develop a multi-equation model of energy supply. On the other hand district heating customers and dis- trict heating companies purchase gas in different markets and pay significantly different prices for the fuel. Price changes in the commercial and industrial rates occur at different frequency and acceleration.151 If the latter is true , then the price of fuel in equation (2-12) is not correlated to the price of gas in equation (2-13) . If the change in the legal environment had a signifi- cant impact on the variable relationships, then the demand CUrVe for steam would be different after the new law. HoWever, the exact date at which the law was enforced dif- fered from state to state and industry to industry. This Study divides the time period into two with 1973-78 being the period in which the law was enforced. A Chow test was performed on each model two demand curve. If the test statistic is significant then the demand curve has been affected by the legal change.152 Results. The discussion of the results will highlight two features of the estimations. First, these will be a comparison of the expected result for each coefficient estimator to the frequency of its occurrence. Second, differences in the frequency of occurrences between models 79 Will— loe noted. Model one includes estimations of equations (2‘9 ) and (2-11) . Model two includes estimations of equa- tiorl (2-13). Appendix A contains the complete results of the regression estimations. The expected sign of the price of steam is negative. If. ‘the regressions are aggregated then the expected result °C=<2urs in 71 of the 224 estimations. The sign is positive 355_ times leaving it insignificant in 118 estimations.153 The comparison of the models reveals three differences. F:i—lecst, the proportion of negative significant signs is hi-<_3her in model one (48 percent to 23 percent). Second, the E>3E7<3portion of positive significant signs is higher in model ‘3‘A7<3 (19 percent to 10 percent). Third, the transformation C>1E the data to correct for autocorrelation had very little liltllpact on model one while it made a substantial change in In<2>zrrelation had only a minor impact on the results.l5‘7 The expected sign of the coefficient for degree days was positive. Degree days are a measure of coldness, the hjtgher the variable the colder it is; and when it is cold o"Slutside more steam is consumed. The regression results ceonfirm this expectation. Of the 224 regression the sign of the coefficient was positive and significant in 129 ca.ses while being negative and significant in only 3 cases. These results occur across both models and all estimation Configurations . 15 8 81 The expected sign for the number of customers is uncertain. In general one would expect that more customers means more demand. However, a drop in the number of customers, could mean that many small customers have been replaced by a few large customers. In the latter instance it is possible for demand to increase depending on the relative size of the large customers. The estimation result are more in harmony with the first hypothesis (87 instances) than with the second hypothesis (28 instances). The large ntlItlber of insignificant cases could be caused by the off- Se‘hting influences of both trends. Model two estimations i1'lczluded a higher proportion of significant results than Inc>Ciel one. Within model two the generalized least square he chnique included fewer significant cases than the ordinary least squares technique.159 In model one, results are inconsistent in respect to the hypothesis that the natural gas price affected steam demand via the number of customers. In more instances (12) the sign of the gas price coefficient was significant and heg‘ative than it was significant and positive (9) .160 Also in the equations estimating the number of cus- tOthers, the time trend followed the pattern of positively Significant until 1955 and the negatively significant after- we1rds 15 of 42 regressions. In only three regresssions were the coefficients significant and follow an alternative Pattern. In all other regressions at least one of the 82 vari.61]31es was not significant. These results are consistent Wittl the hypothesis that the mid-fifties marks a decline in serv ice area business activity.161 In model one, a time trend was used as a proxy for reegl income. The sign of its coefficient was positive and SiJanificant in 48 of 84 estimations. This result is con- ‘Sistent with the hypothesis that the time trend is an iicceptable proxy for a growing real income.162 However, while income was growing for the nation fairly <:onsistently over the period, this does not imply that income increased in the service area of every utility. In (an.attempt to make the income proxy specific to the partic- Illar city retail sales by city was substituted for year in «every regression in model two. The new proxy also has limitations. For instance, if banking and government service activities increased substantially to offset a drop in retail outlets, then income of the population could increase while retail sales decreased. The regression results for this parameter did not CflJearly define a trend. Out of 140 regression the coeffi— <fident was positive and significant 47 times while being negative and significant 38 times. These ambiguous results were probably due to the imprecise nature of the proxy.163 Test for Stability. Over an extended period of time the relationship between the variables could change. An 83 estiJIIation procedure that ignores this possibility could regiLsster incorrect inferences. To check for this possi- bility the period was divided into two parts: 1946-1972 5““3 1973-1978. The split of 1972/73 was chosen because, bY’ assumption, 1973 marked the year in which all companies cOmplied with the air pollution amendments. The importance of the legal change was that five <20mpanies responded to it by switching to a greater depend- ence on natural gas as an input fuel. For those companies, demand curve estimates could be inconsistent over the entire jperiod 1946-1978 while being Consistent for the subperiod 1946-1972. A possible test for this problem would be to check for significant changes in the demand curves. If the demand curves were the same then it is possible to infer that the curves are consistent. A Chow test was used to make this test. Of the ten comparisons made only in one instance was a demand curve for one period significantly different frxmnia demand curve for the other period.164 Thus there is 3hittle evidence to support the hypothesis that the use of Ifiitural gas as a fuel had a significant impact on the esti- mation process . Elasticities. The price elasticity of demand was Calculated for all non-perverse (negative sign for its own price) significant coefficients. Of the 71 relevant 84 coefhfficients only in four cases did the estimates imply that the éiemand was elastic. For the other 67 cases the range 0’13 elasticities was from -.05 to -.83. The results indicate the“: for most cities an increase in price would have gener- ated an increase in revenue. Summary. This examination of the regression patterns, 538 presented so far, does not contain answers to two crucial <1uestions: first, why are there so many negative signifi- cant coefficients for the price of gas; and second, why are ‘there so many insignificant coefficients for the price of steam? ‘ To shed light on these questions it is necessary to look a little closer at the data. The important concern is the relative price of steam to gas. At first glance it seems that the price of gas fell below the price of steam by 1950.165 The rational response should have been to switch from steam to gas. However there is little evidence to support that conclusion. (Rinsumers kept on buying more steam even though its price reilative to gas continually rose. The explanation for this EKTtion lies in the energy equivalent price of gas and steam. When both fuels are converted to energy equivalent prices theaprice of gas was below the price of steam for only five Cities until 1971.166 Those five cities do not follow the pattern of rising quantity consumed until the early 70's. 85 iIn addition, conversion costs, potential capital loss on Cik>solete equipment, higher labor and insurance cost aSSC><2iated with gas boilers would conspire against the trallsformation of energy supply systems even if differ- enilial energy cost had been favorable for gas.168 Thus even if consumers in their decision making process consider Steam.and gas substitutes there was no reason for the sub- Stitution to take place in historical time. Whatever Correlation took place between gas prices and the quantity (bf steam purchased (in this case a negative correlation) xvas probably an historical accident rather than a record of a causal relationship. The question of the lack of significance for the coef- ficient of the price of steam can also be addressed from an examination of the data. For most cities, the price of steam was relatively constant in the forties and fifties, dropped slightly in the sixties and rose sharply in the Seventies. Steam consumption increased steadily until the (Early seventies and then dropped off. Given these patterns a {elausible conjecture for the regression results could be thirt in period prior to 1973 real income (which appears (filly in proxy form in these equations and thus possibly Imisspecified) increases lead to the increases in consump- tion, and in the period after 1973 increases in steam prices caused the decreases in consumption. However due to the collinearity of all energy prices in the latter 86 peri4c>d.(steam price increase occurred through fuel adjust- ment; clauses activity rather the rate case changes) the regression technique was unable to determine which fuel Prixze increase was responsible for the decrease in steam COTlsumption. It is this problem of multicollinearity in time crucial period when prices changed, that was responsible fOr the large number of insignificant results.169 Technology Innovation centered on the use of hot water as a heat transmission fluid and the use of trash as an alternative low cost fuel. European utilities have adopted these two techniques in mass. Most European district heating systems built since 1945 use hot water to transmit heat. In western Europe there are at least 243 combustion units presently 170 recovering heat from waste. These units can devour 3250 metric tons per hour. 40 percent of this capacity is in West Germany. Denmark has the highest per capita capacity. 48.2 metric tons per hour per million persons (about 1 lb. Per hour per person) .171 In the United States, the two large systems built and OEHErating since 1945, both use steam.172 The Trenton dis- ‘trict heating system scheduled to start in the near future, Will be the first U.S. system to use hot water.173 As of 1978, only twenty plants used trash as fuel.174 A hot water distribution system is preferred to steam distribution system for at least eight reasons: 87 1. For a given supply of heat per hour to a distri- butfiLcjn network, more electricity can be generated. This advantage is the result of lower working temperatures in the water system. The use of lower temperatures allows St£eam.to do more work in the turbine prior to its extrac- tiJDn for heating purposes and thus, to generate more electricity.]'75 2. For a given supply of heat to the final consumers, less heat needs to be supplied to the distribution network. {This advantage is the result of the inherent properties of steam that cause heat loss in the distribution network.176 First, as the steam is sent out part of it will con- dense. The condensation must be removed at steam traps *which are built into the line at regular intervals. All of the heat in the condensate is lost to a system that does not return condensate. In systems that do return condensate, some of the latent heat of the steam is lost. Second, after the customer uses the steam the conden- Sate will exist as liquid under pressure greater than 1.6 PSIG and at temperatures above 212°F. The condensate must bEB lowered to atmospheric pressure before returning to the bcDiler. In the process of reducing the pressure there is a flash loss to system. This loss has two impacts: a Significant amount of the content of the condensate is vented into the atmosphere, and the heat content of the remaining condensate is reduced. 88 For example, assume steam is sent out at 400°F con- taining 1201 BTU/lb. The customer uses the latent heat of the steam 826 BTU/1b. and condensate contains the sensible heat at 375 BTU/1b. However, the condensate is still at a pressure of 247 PSIA. It must be reduced to atmospheric pressure. During the reduction the heat content of the condensate is lowered to 180 BTU/1b.; plus some of the condensate, between 5 and 20 percent is vented. The con- densate is returned to boiler. Feed water (at 60°F, 28 BTU/ 1b.) must be added to the condensate. The mixture must be heated original send out steam conditions. The arithe- matic (assuming 5 percent loss to the atmosphere) of this heating process is: 1028 = 865 + .5[375—28] + .95[375-180] latent heat sensible heat sensible heat added to feed added to water condensate Thus a steam distribution system cannot be more than 80% (8 6 5/1028) efficiency. On the other hand a hot water system re“turns all of the heat not used by the customer to the he a ting plant . 3. Maintenance expenses are lower because there are 11c) steam traps or pressure-reducing valves that need regular lrISpection and repair.177 4. For a given service area, the total length of the supply piping is shorter, because pipe length is not only a function of service area size but also pipe expansion needs. 89 Pipes expansion needs are a direct function of heat. There— foxrea , the higher steam temperatures require that steam distribution systems have longer pipes.]’78 5. Hot water has a greater storage capacity. For example, hot water systems usually send out water at 250°F. Ihaél1t. storage per cubic foot at this temperature is 12,064 BTU/cu.ft. (BTU/lb. x 1b./cu.ft.: 208 x 58). Steam systems send out temperatures of 400°F. Heat storage of steam at t31i.£3 temperature is 645 BTU/cu.ft. BTU/lb. x lb./cu.ft.: 12()21. x 538). This property provides water systems with greater flexibility in meeting peak demands.]'79 6. Hot water transmission costs are cheaper than steam transmission costs. To transmit hot water greater distances reQnires additional pumps and power. To transmit steam gleeiiiter distances requires higher outlet pressures. The lljtfiilner pressures reduce plant electricity generation, and zir1<2=reases pipe and pressure reduction value costs. The sum of the additional steam costs is greater than the sum of the additional hot water costs.180 7. Hot water distribution losses due to pipe convec- tion are less than steam pipe convection losses. Convection losses are a direct function of the difference between pipe temperature and ground temperature. Given that steam pipes are hotter than hot water pipes, it follows that steam losses are greater.181 90 8. Hot water systems can use alternative pipe materials. Compared to the standard steel and cast iron pipes the alternative pipes have a higher material cost but lower installation cost. Thus total pipe costs for alternative small diameter material pipes are below the When large diameter pipes are 182 costs of equivalent piping. needed the hot water systems can use the steel pipes. Trash burning also has a number of cost saving advan- tages. First as a fuel it is free. The heat content of a ton of trash is approximately 10 million BTUs, valued at 55 dollars per ton when the price of crude oil is 34 per bar Zc‘el.l83 Second burning trash reduces acreage need for 1<’=111d—fills;184 and third reduces transportation costs aSSOciated with waste management.185 Regulation Three trends in the regulatory arena can be identified during this period. First, a perverse regulatory lag, set in . caused by the unwillingness of companies to initiate rate cases. Second, in those rate cases that did occur, prices and cost allocation schemes were re-examined under the scrutiny of economic theory. Third, federal regulatory involvement in the industry increased as the government Set air pollution standards and fuel use requirements. Regulatory Lag. Normally, regulatory lag is caused by the regulating commission. Two characteristics of the 91 reglliLation are responsible for the lag. First, the commis- sicxrl needs time to evaluate and authorize the change in ratzeass. Second, the commission uses historical rather than foxreekcast test year data. If the future is significantly different from the past, then the authorized rate change Inigyljrt be higher or lower than necessary depending on the Change in future year costs.186 On the other hand, companies can cause regulatory lag when they fail to initiate rate cases. The reason for 1311.53 practice is due in part to the heating companies' Status as small appendages stuck onto the electric utilities. These utilities must appear before commissions to obtain rate increases for the electric business. They would prefer ru>1:_ to open their books again for the steam cases, nor to bear the burden of another rate case. The result of this Ipxreadctice is that many steam companies appear to be money 1‘C’Ssers, when in fact that might not be true. Further, when at 1:ate case is brought the increase sought is often dramatic. 3:11- one case the increase sought was 200% plus a fuel adjust- me ht clause . 187 Re-examination of Steam Rates. In two recent Consoli- deted Edison steam cases, the New York commission used its Vision of economic theory to evaluate the company's rate Change requests. That vision stresses the need to provide the consumer with proper price signals, signals that present 92 to t:]ae consumer the cost of his or her decision to society. Wheat). the consumer faces the proper prices, his decision to purchase good A or good B will lead to an efficient alloca- tion of resources. The proper price would be one that equals the marginal cost of production.188 The ability to determine a unique marginal cost is a prerequisite for adapting this strategy. The conunission addressed three problems involved in determining the Correct marginal cost: first, it noted that it had to choo se between short run and long run cost calculations.189 Second, faced with the simultaneous production of steam Ik>e the optimal guide, an additional assumption that all future price changes effecting alternative energy supply Systems must not alter the relative costs of these systems HRISS'E: be made. However, in an era when energy prices have risen very quickly and there is a likelihood of continuing €H1€3Itrgy price increases, a belief in a constant relative PI?5—<:e energy to capital is not plausible. When this rela- tj¥\rea price changes, then the relative costs of energy supply Systems change. This reasoning suggests that long run costs toc3a); cannot be an optimal guide to the future. The next problem the commission faced was how to ElJLIIbcate common cost of fuel and boiler capacity between titles steam and electric service. The costs are common rather tirlan joint because the "same equipment may be used to make E311‘oducts A and B, and when producing more A uses capacity ‘3}1at could otherwise be used to supply."194 In light of the common cost it is important to note ‘that marginal cost is transformed into marginal opportunity costs.195 94 In a concurring opinion to the 1975 decision, Alfred Karlrl provided these alternative definitions of marginal (xassizs: l. the addition to cost involved in increasing the production of one while holding the production of the other constant. 2. Value of the incremental quantities of the one sacrificed in order to increase the pro- duction of the other. 3. the incremental cost of producing the one sacrificed by alternative, single-purpose technology as might be necessary because its production is reduced in order to pro- duce additional output of the other.196 However, in the body of the decision, the Commission ignored these definitions. Prior to 1975, the fuel cost Eitl1tributed to the steam service was the additional fuel r16ieded to produce steam above the fuel needed to generate the electricity at the plant that sent out the steam.lg7 tJrlder this procedure (Table 5, scheme one) the fuel charge VvEis less than the energy in one pound of steam (800 BTU (zllarge for approximately 950 BTU latent heat in the steam). “Fhis charge would be the marginal cost of the steam if and (Duly if the operations of all other plants in the system ‘Were not altered due to the steam output, which is not always the case. 95 Further, it defines steam as the marginal output of p1“5113t. By doing so, the entire fuel savings due to cogen- eration is passed through to the steam service. If on the otzljjer hand electricity was considered the marginal output, trieaxn the electricity fuel charge would be 11,300 BTU per KW [1.JZ .800 at the cogeneration facility minus 1500 for live si::I:‘n by the electric system due to the steam production. url‘dis cost is measured by the heat rate of the peaker unit 1ill-«Eat must supply the electricity no longer generated by the QQgeneration facility. In 1978, Consolidated Edison again asked for a rate jLIlcrease.202 As part of the rate case, it proposed to undo ‘llne 1975 charge and return to prior 1975 fuel allocation SScheme. Three reasons were given for its reversal. First, ‘the cogeneration facilities were now generating electricity On a basis closer to their original rather than their latter status. Second, due to the risk of blackouts, the 97 faczjtlities should always be considered electricity plants filtsst. Third, the steam system lost customers following tiles 1975 decision. If this drain on the system becomes a gllssher then the system might be irreparably harmed and all Savings from cogeneration would be lost.203 The commission agreed with the company and ordered izfre reversal.204 Steam again became the so-called "marginal crcrtput". Yet neither decision was based on a marginal system analysis. Thus neither decision fulfills the commis- sziron's stated task of developing the proper price signal. In its 1975 decision the commission also reviewed the Iaai1te base allocation scheme. Here the problem was what .PITCDportion of the steam boiler investment should be in the Stleaenn service rate base versus the electric service rate baissee. IPrior to 1975, the steam service's share of the investment was determined by subtracting the capacity of live steam boilers from steam demand at the summer electric peaiakLthen dividing this difference by the capacity of the CCmtgeneration facilities' boilers. A summer peak was used bee-<:ause it was the peak demand period for the Company's bcJilers even though it was not the peak demand period of {tile steam system.205 The company proposed to change the scheme because it <3id not reflect actual company practice. The live steam boilers were not used to their full capacity. Thus the amount of cogeneration capacity used by the steam service 98 was underestimated by the above scheme. The company pro- posed to alter the scheme so that actual live steam boiler output, not boiler capacity, would be subtracted from steam demand. This difference would be divided by the cogenera- tion boiler capacity to obtain the proportion of the invest- ment to be included in the steam service rate base.206 The commission accepted the company's proposal. The change increased the size of the steam service's rate base and therefore simultaneously its revenue requirement.207 Again it is necessary to ask what was the commission trying to do and did it accomplish that task? The commis- sion's stated goal was to include in the steam service rate base an amount that would reflect the capacity derated from the electric service. Alternatively, the rate base could reflect how much additional capacity must the electric ser- vice have on hand due to the provision of steam for sale. To this end the company and the commission agreed that the proper peak period was the summer electric peak. Second, the choice of actual live steam output over live steam capacity also reflects the stated goal.208 However, when it divided the difference between demand and live steam output by a boiler capacity, it became essential to correctly define the nature of the boiler's capacity. This problem arises because the capacity of the boiler changes with changes in the definition of the output. The examples shown in Table 6 illustrate this point. If the 99 output of the boiler is defined as pounds of steam per hour then the steam service is responsible for 25 percent of the boiler capacity. If output is defined as the heat content of the steam then the steam service is responsible for 20.4 percent of the boiler capacity. However, if the output is the ability of the plant to generate electricity, then the steam service would be responsible for only 10.7 percent of the boiler capacity. The reason for the differences is that electricity generation converted a lower percentage of the input into a saleable output than steam production. There- fore when the steam is extracted for the turbine there is not a proportional reduction in electricity output. For the commission to fulfill its goal of estimating capacity derated (decline in capacity) due to steam genera— tion, it should have chosen the last scheme described above. Instead it chose the first scheme. It is not known whether the choice was made because of its administrative ease or due to ignorance of the production process. No matter what the cause, the outcome was to include a higher share of the joint investment in the steam rate base than was justified by the commission rationale. Following the 1978 decision, if the commission had implemented its stated rationales for rate setting then the winter rate would have included a fuel charge for 800 BTUs of fuel; and the summer rate would have included a fuel charge of 1,100 BTUs of fuel, and a capital charged that 100 was based on the inclusion of 10.7 percent of the cogenera- tion facilities in the steam rate base. Instead, the winter rate by accident included the same fuel charge of 800 BTUs of fuel; and the summer rate included a fuel charge of 800 BTUs of fuel and a capital charge that was based on the inclusion of 25.0 percent of the cogeneration facilities in the steam rate base. Finally the commission addressed the time of day peak/ off peak pricing problem. The company contended that demand charges would reduce the system peak and therefore reduce system capacity requirements.209 It proposed a demand charge based on the customers maximum hourly usage, indepen- dent of the relationship between usage and either the system's summer or winter peak. A company survey of six customers found that this charge would reduce the system peak by between 8 and 17 percent.210 The commission noted that the proposed charge would reduce the system peak only if the customer's peak was coincident with the system peak. The system peak usually occurred between 7 am and 9 am in the morning. However, New York City law requires that apartment buildings be heated by 6 am. Thus the apartment house demand peak was probably earlier than the system peak and that the owners of the apartment could not alter their demand in reaction to the demand charge. Because of the legal constraint on 101 apartment owners, the commission ordered the demand charges be included in the rates of commercial buildings only.211 This review of the New York Commission's action high- lights the problems of using economic theory as a guide to steam pricing. The theory doesn't provide a unique solution to the price problem. Decisions must be made about the determination of the so-called "marginal output", the meaning of equiproportional share, the definition of capacity, the use of system versus plant analysis and the reasonableness of alternative plant cost. Federal Regulation. The federal regulatory involve- ment began with the 1970 amendments to the Clean Air Act. Those amendments mandated that pollution standards must be set and enforced upon all stationary sources of pollutants. Regulations implementing those standards were enacted by the states under the supervision of the Environmental Pro- tection Agency.212 Faced with the new regulations, coal-burning district heating utilities had to choose between investments in air pollution control equipment or switching to low sulphur oil as a fuel source. A survey of twenty companies taken in 1969 revealed that eleven companies burned coal exclusively and that five others relied heavily on coal. By 1973, of the eleven coal-burning companies, five were now burning oil exclusively while two burned both coal and oil. Of the 102 five companies burning both coal and oil in 1970, three burned only oil by 1973.213 An analysis of the fuel expenses of the ten companies that moved towards oil showed that, at the time of the trans- formations from coal to oil, the companies on average accept- ed 30¢ per million BTU increase in fuel expenses. By 1978, this differential had increased to 65¢ per million BTU. Cost estimates for air pollution equipment are not available, so cost comparisons have not been made. However it is clear that decisions to burn oil have dramatically increased company fuel costs. 10. 11. 12. 13. 14. 15. 103 Table 1 Economy of High Back Pressure Turbines Steam pressure at throttle, lbs. gauge . . . . . . 650 400 Steam temperature at throttle, deg. Fahr . . . . . . 700° 610° Saturation temp. of steam deg. Fahr . 498° 448° Degrees superheat, deg. Fahr . . . . 202° 162° Initial heat in steam at throttle BTU . . . . . . . 1345 1309 Heat drop, adiabatic expansion to 85 lb. abs.BTU . . . 195 147 BTU per 1b. steam converted to net— work at 70% Ran- kine effeciency . 136 103 BTU per lb. taken . from steam in tur- bine (including generator loss) . 145 110 BTU in exhaust steam at 85 LB. abs. pressure . . . . 1200 1199 Steam temperature at exhaust deg. Fahr. 343° 341° Saturation tempera- ture at exhaust deg. Fahr . . . . 316° 316° Degrees superheat and quality of exhaust from turbine at 85 lb. abs. . . . . Water rate of tur- bine lb. per Kw.hr. 25.1 33.1 Steam flow lb. per hr. to 2-15,000 Kw. units full load . Steam flow lb. per hr. to 2-5000 Kw. units full load . 27° 25° 753000 993000 150 366° 366° 00 1196 53 37 40 1156 316° 316° 924000 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 104 Table 1 Continued: Total yearly steam available for turbine during 9 months of Opera- tion, million lbs. 3200 Average hourly steam flow to turbine during 9 months of operation (6650 hr.) lb. per hr. . . . 482000 Average load on tur- bine (6650 hr.) Kw. 19200 Load factor of tur- bines when running --per cent . . . 64 Load factor of tur— bines based on entire year--per cent . . . . . . 48 Kw.hr. sold during 9 months (million kw.hr.) . . . . . 128 Yearly heat consump- tion of turbine including 5% for radiation, leakage, losses, etc. (1000 million BTU) . . 489 Coal to be charged to power generation with 85% efficiency boiler and economizer, 10,000 BTU coal, tons per year 28700 Yearly coal cost charged to power generation based on coal at $4.00 per ton . . $ 114,800 BTU per kw.hr. including boiler losses and 5% loss turbine roon 4530 Coal cost charged to power generation, mills per kw.hr. 0.9 Steam sold, million lbs. per annum 4250 Peak send-out on heat- ing system thousand lb. hr. 1940 3500 526000 15900 53 40 106 404 23800 95,200 4530 4250 1940 3400 510000 5500 55 41 36. 143 8400 33,200 4530 .93 4250 1940 Table 1 Continued: 29. Per cent steam sold passing through turbine (16 + 27) 30. Per cent steam sold direct from boiler 100 -- (29) . . 31. Per cent steam peak passing through turbine (14 % 28) 32. Income from steam sold at $1.00 per M . . . . . 33. Gross income from electric output based on primary charge of $21.00 per Kw. year and 0.4c per kw.hr. 34. Gross income in mills per kw.hr. . . 35. Estimated additional investment . . 36. Net income from electric output, after deducting 138% on additional investment; 1.22 mills per kw.hr. for operating charge; total . $ 37. Net income in mills per kw.hr. 38. Net electric income per thousand 1b. steam sold, cents 39. Possible selling price of steam to yield same return as live steam heating system, with steam at $1.00 per thousand 1b., cents . . . . . Source: 75.0 25.0 38.8 $4,250,000 $1,139,000 8.90 $2,280,000 675,000 84.12 82.5 80.0 17.5 20.0 51.2 47.6 $4,250,000 $4,250,000 $1,054,000 $ 356,800 9.96 9.72 $2,000,000 $ 700,000 $ 617,000 $ 217,000 5.82 5.92 14.51 5.11 85.49 94.89 Orr, "Report of the Research Committee," p. 95. 106 Table 2 Capital Costs for Dual Purpose Plant Investment charges, 13%% of $76.00 = $10.25 per kw. capacity, or mills per kw. hr. . . . . Coal charges including standby losses, mills per kw. hr. . . . . . . . . . . . . . . . . . Operation charge for boiler room, including maintenance, based on 40¢ per ton, mills per kw. hr. . . . . . . . . . . . . . . . . . Operation and maintenance turbine room, ($14,000 per year) mills per kw. hr. . . . . . Total cost of power generation in heating plant including all charges, mills per kw. hr. . . . . . . . . . . . . . . . . . . Source: Orr, "Report of the Research Committee," .10 .11 107 Table 3 Total Cost for Dual Purpose Plant 1. Cost of 7-15000 sq. ft. boilers for 650 No. G including economizers and super- heaters (not erected) . . . . . . . . . . $ 840,000 2. Cost of 7-15000 sq. ft. for 200 No. G (not erected) . . . . . . . . . . . . . . 510,000 3. Additional boiler cost for same size of boilers (1)-(2) . . . . . . . . . . . . . $ 330,000 4. Additional boiler room and boiler building cost due to 12% larger boilers when power is generated, based on actual practice, per 1000 lb. steam capacity . . . . . . . 480,000 5. Additional cost due to heavier super- structure in high pressure plant, boiler feed pumps and other extras . . . . . . . 100,000 6. Cost of H. P. header and turbine pipe . . . 50,000 7. Total additional cost of boiler room to be charged to power generation everything included (3+4+5+6) . . . . . . $ 960,000 8. Cost of 2-15000 K.W. Turbines . . . . . . . $ 760,000 9. Cost of electric equipment for 30,000 kw . 150,000 10. Cost of turbine room building 50x70x40 with no basement . . . . . . . . . . . . 60,000 11. Freight and erection of turbines . . . . . 50,000 108 Table 3 Continued: 12. Total turbine room cost to be charged to power generation . . . . . . . . . . . . $1,020,000 Total cost (7 + 12) . . . . . . . $1,980,000 Engineering--miscellaneous . . . 300,000 Total investment . . . . . . . . $2,280,000 Total investment per kw. . . . . $ 76.00 Source: Orr, "Report of the Research Committee," p. 96. 109 .mma .m .COHDMHUOmmm maeummm poflupmflo Hmcowumz vs» mo mocmumwcou Hmscca susflclmuuwce may mo mmcflommooum =.mwuuHEEou noumwmwm may mo uuommm= "condom .38 mnm.a can mvm.a um mmumu HmucmEmHocfl mo mmmum>< n .38 5mm.a can mvm.a um mmumn Hmucmamnocfl mo mmmum>¢m AHH.H maml mam nun . . . . . . . .un umm Dem .mumu pawn usouccmm smmum >vw+ mmm mmm+ III . . . .Hn Mom Dam woe .mmcmno Doc usmcfilumoc Emummm o mmm+ oom+ III . . . .Hn Hmm Dem woa .mmcmno usmcfllummn mpflmumpmz th+ II: mam: III . . . . . . . . . . . .Hn mom Dam moa .mcflmumumx mcflpsHoxm mmcmso usmcfilumwn Emummm owuuomam omm.ma In- momm.ma an: . . . . . .uz .3x Hum Dem .m .mflm scum .moflmnmums mcflpsaoxm mumu Damn HmucmeHUCH Empmmm oeupomam oov.m mmn.m ooo.o oov.m . . . . . . . .Hn Mom Dem on .uzQCHlummn mpflmnmumz “NI.- 0 HNII I'I o o o o o o o o o o o o o o o o o chHpmpm “mguo “NI. 0 HN+ I." o o o o o o o o o o o o o o o chHumpm mU-flmumflvmz .3x coca .COHumumcmm CH mmcmso OOOsN ooo- OOO§N COOSN o o o o o o o o o o o o o o o O o o o o o o HmpopH. mnm.a mvm.a ham.a mvm.H . . . . . . . . . . . . . . . . . mGOHumum umcpo mmv mmv mhv mmv . . . . . . . . . . . . . . . . coeumum moflmumumz .3x oooa .coaumumcmm oeugomam oov oov cow 0 . . . . . . . . . .Hn mom .QH oooa .u:oupcmm Emwum a U m e H .mflm CH caonm mucflom emoH amumsm .3x ooo.ooo.~ an .un umm .na ooo.oos umuam mom mnmm ummm noonocmm Emmum mo coaumasoamo v GHQMB 110 Table 5 Alternative Fuel Use Allocation Schemes Fuel Use Data 1. Average system heat rate: 12,000 BTU/kw 2. Baseload heat rate: 10,000 BTU/kw 3. Peakload heat rate: 16,500 BTU/kw 4. Live Steam heat rate: 1,500 BTU/kw 5. Cogeneration conditions: 12,800 BTUs generate 1 kw. and 1 lb. of steam 24,000 BTUs generage 1 kw. and 15 lbs. of steam Alternative Schemes 1. Steam is the marginal output/individual plant analysis (individual plant is the average company plant): 12,800 BTU - 12,000 BTU = 800 BTU fuel charge per 1b. of steam 1 kw. ‘1 1b. of steam) (1 kw') 2. Pro rata allocation using live steam and base load plant to determine allocating rates/Consolidated Edison's 1975 proposal. fuel use: Base load = 10,000 BTU; Live steam = 1,500 BTU; Cogeneration = 12,800 BTU 10,000 BTU 10,000 + 1,500 1,500 10,000 + 1,500 = 13% = 87%; (13%) x (12,800) = 1664 BTU fuel charge per 1b. of steam 3. Pro rata allocation using live steam and company average heat rate to determine allocating ratio/Commission decision 1975. fuel use: Company average = 12,000 BTU; Live steam = 1,500 BTU; Cogeneration = 12,800 BTU 12,000 BTU 12,000 + 1,500 _ . 1,500 B _ ‘ 89%' 12,000 + 1,500 ‘ 11% (11%) x (12,800 BTU) = 1408 BTU fuel charge per 1b. of steam 111 Table 5 Continued System Margin using an average plant to cogenerage and replacing electricity lost at average plant by oper- ating a peaker plant. fuel use: average plant generates 2 kw. or 1 kw. and 15 lbs. of steam using 24,000 BTUs, peak load plant uses 16,500 BTUs to generate l kw. 16,500 BTUs 15 lbs. of steam = 1100 BTU fuel charge per 1b, of steam 112 Table 6 Alternative Boiler Capacity Allocation Schemes Typical Steam Conditions for an Extracting Turbine: Enthalpy (BTU/1b) Boiler outlet 1463 Turbine exhaust 990 Condenser outlet 69 Extracted steam 1192 Boiler Capacity: 800,000 lbs/hr Steam Extracted: 200,000 lbs/hr Analysis of Work Done: 1394 BTU/1b: energy added 473 BTU/1b: energy transformed into electricity Boiler : 1463 - 69 Turbine : 1463 - 990 Condenser: 990 - 69 = 921 BTU/lb: dissipated into heat sink Electricity Efficiency Analysis: 473 BTU/1b _ Turbine Work = 33.9% 1394 BTU/1b - Boiler Work 3413 BTU/kw .339 10,068 BTU/kw heat rate Analysis of Boiler Capacity Used by Steam Service: 1. output is the number of pounds of steam: 200,000 1b/hr steam extracted 800,000 lb/hr boiler output = 25°°% 2. output is the energy in steam: 1192 BTU/1b x 200,000 lb/hr = 20.4% 1463 BTU/1b x 800,000 1b7hr 3. output is the ability to generate electricity: (1192 BTU/lb - 990 BTU/lb) x 200,000 lb/hr = 10 7% (1463 BTU/1b - 990 BTU/1b) X 800,000 lb/hr ° 113 Table 7 Total Gas Utility Sales Year Millions of Therms Percent Change (Avg. Annual Rate) 1935 12,923 1945 25,867 7.2 1955 66,586 9.9 1965 119,803 6.0 1975 148,629 2.2 Source: American Gas Association, Gas Facts (Arlington, Virginia: American Gas Association, 1976), p. 15. Table 8 Natural Gas Transmission Pipeline Year Length Percent Change (Avg. Annual Rate) 1945 72,280 1955 142,490 7.0 1965 210,780 4.0 1975 262,600 2.2 Source: American Gas Association, Gas Facts, (Arlington, Virginia: American Gas Association, 1976), p. 23. 114 Table 9 List of Cities 1. Cambridge, Mass. 14. Baltimore, Md. 2. Concord, N.H. 15. Milwaukee, Wis. 3. Piqua, Ohio 16. Cleveland, Ohio 4. Cheyenne, Wyo. 17. St. Louis, Mo. 5. Philadelphia, Pa. 18. Dayton, Ohio 6. New York, N.Y. 19. Pittsburg, Pa. 7. Toledo, Ohio 20. Denver, Colo. 8. Akron, Ohio 21. Seattle, Wash. 9. San Diego, Calif. 22. Harrisburg, Pa. 10. Detroit, Mich. 23. Lansing, Mich. 11. Boston, Mass. 24. Atlanta, Ga. 12. Indianapolis, Ind. 25. Grand Rapids, Mich. 13. Rochester, N.J. 26. Spokane, Wash. Table 10 Study Group Cities New York Detroit Milwaukee Boston Rochester Dayton Philadelphia Pittsburg St. Louis Lansing Baltimore HOKDCDQO‘WhUJNH o o o o o o o o o o o Hra 115 .hmaumsa .ma .Amvma .coaumaoOmmm mcaummm uoauumao Homoeumz uzmnsnmuuflmv GOflDMHUOmmm acapmmm poeupmfln fianceumz may mo mocmuwmcou Hmsccm nuxflmlwunflne mnu mo mmcflpwmooum =.omuuHEEoo mofiumflpmum may mo puommm: "condom no >.o~ >m.ma mm mva.mm mam.ooa mmm.m omm.mmv.om mvma no m.om mm.ma mm mam.Hm mmw.am vmm.m moa.qom.om vvma no m.ma H5.¢H am oom.am omm.mm moo.m mma.vom.m~ mama no m.ma mm.mH mm Hen.om omo.mm oom.m moo.mmm.v~ mama no m.ma mm.ma om Han.om mma.hm ham.h ooH.hmm.vN avma mm m.mH vm.va mm omm.om mmm.vm mom.h omh.mmo.v~ ovma mm m.ma mm.vH mm www.ma wha.om omh.n oav.vam.am mmma mo o.mH vm.va mm mmm.ha mmm.mm omh.w mmm.mmm.o~ mmma no m.vH om.wa om mmm.>a mom.qm Hm>.> 5mm.mah.am hmma no m.¢a mm.vH om mmw.ha mma.mm vmm.h omv.HHm.H~ mmma mo H.wa om.ma om Hoa.ma moa.mm omv.> hmm.omh.am mmma no h.va mm.va om mom.>a mmm.am omm.n mam.amm.am wmma mm m.mH va.va hm mnm.ma mmm.mm mmm.h Nmm.wmm.om mmma om mmo.ma vmm.mm mmm.h mmm.vvm.om NmmH an o.mH nv.ma hm Hmm.m omm.vn aaw.h ovm.omw.a mmma impcmo. 12ml “moo maa\mna imaa 23 umou Hmsm UmcHDm Amucmov .mmo Awe Anna 2V Hmou mace: Hmumfl Hmou mom mscm>mm mausom cmumm>cH om>nmw mmamm .>mm .um>< mscm>mm oaom Emmum mmmum>< .xmz Hmuflmmu mHmEogmso Emmum Hmmw mvaImNmH mmauao macho mcsum now moaumaumum HH OHQMB 116 Table 12 Steam Sales of Study Group Cities Percent Change Year Steam Sales (mm lbs) (Avg. Annual Rate) 1935 21,720 --- 1945 30,425 3.4 1955 38,154 2.2 1965 54,976 3.7 1975 66,197 1.8 117 Table 13 Data Sources Title Used in Item the Source Source 1. Quantity of Steam Total steam sales IDHA annual Sold proceedings 2. Price of Steam Average gross revenue IDHA annual proceedings 3. Degree Days Actual degree days IDHA annual proceedings 4. Number of Customers Number of customers IDHA annual served proceedings 5. Price of Gas Gas utility revenue American divided by gas Gas utility sales for Association commercial class by state 6. GNP Deflator GNP deflator Survey of Current Business 7. Pipeline Length Total length supply IDHA annual (steam) piping proceedings 8. Capital Investment Capital investment IDHA annual proceedings 9. Maximum Hourly Maximum hourly IDHA annual Capacity send-out capacity proceedings IDHA: International District Heating Association 118 Table 14 Variables of the Model The price of steam is the average gross revenue received by steam heating companies. The price is divided by the GNP deflator to transform it into a relative price. The price of gas was calculated by dividing gas utility revenues by gas utility sales for commercial class customers. These statistics are only available by state. The state-wide price was adopted as the price for every city in that state. The price is divided by the GNP deflator to transform it into a relative price. This price was chosen over prices available from the Bureau of Labor Statistics' consumer price index for two reasons. First, the consumer price index does not survey many of the cities in the data set. Second, the price used by the consumer price index is the price to single—family residential dwellings. Steam companies, usually, do not service that type of residential market. The number of customers served is recorded on December 31 of the given year. The number of degree days, annually, is calculated by first subtracting for each calendar day, the difference between 65 degrees and the average daily temperature. 119 Table 14 Continued Second, these differences are summed to arrive at the annual figure. Only calendar days with an average temperature of below 65 degrees are included in the calculation. 120 m N m m Aon on Iv. A om om w A om Wom muHsmmH mm o m o o mama mmummo .m o NH o m new» .v .H n H H mumsoumso mo Hmnesz .m H N o m mmo mo mOHum .m o m , H oH Emmum mo moHum .H om.w m A mm m w mm om.w m A mm m H mm mmHQMHHw> ucwpcmmmUcH pcmHOHmmmou m>HuHmom ucmHOHmmmoo m>Hummmz owfiuommcwuuao "mmHQMHHm> cHom Emmum mo prucmsq "mmHQMHum> usmocmmmw Hm "mmHUHo mo Hones: Hobos :oHumsvm mHmch "mHszOHDMHmm ccmfima mH OHQMB 121 h N m m Ace on v A om om Aw Aom wom muHDmmH mm m OH O o mama mmumma .m H NH o m ummw .H m H o H mnmEOHmso mo umbesz .m o o m I a mum no mOHHm .m o m H m ammum mo moHum .H om w m A mm m.w mm om w m A mm m w mm mmHQmHHm> pampcmmmUcH ucmHonmmoo m>HuHmom namHonnmoo m>Hummmz musomooum unsoHOImcmunoou MH> UmEHommcmuu "mmHanHm> UHOm Emmum mo mpHucmsq "mHQMHHm> ucmpcmmwp Hm "mmHuHo mo Hones: Hobos :oHumsqm mHmch “chmGOHUMHmm panama mH OHQMB 122 v m m. m A05 on.w A om om.w A om .wom muHSmmH mm H m H H mumaoumso mo HmnEdz bmHOHpmum .m H m o m “mow .e H m o o mama common .m H o H m mmw mo moHHm .N o m H m Emmnm mo moHHm .H cm W m A mm m.w mm om w m A mm m w mm mmHQmHHm> pcmocwmmocH ucmHonmmou m>HuHmom ucmHonmmoo w>prmmz omenommcmuucz "mmHQMHHm> oHom Emmnm mo muHucmzq “mHanHm> usmocommp Hm umeuHo mo Hones: COHumEHumm mmmumnoBB "QHSmCOHHmHmm panama 5H mHQMB 123 A05 o>.w A om om.w A om .wom muHsmmH mm o a o o mumEoume mo Hmnadz pmuoHUmum .m m m o 0 Ham» .v m m o o mhmo mmummo .m o o m m mmw mo moHHm .m o N H m Emmum «0 moHum .H cm W m A mm m.w mm om w m A mm m.w mm ucmHonmoou m>HuHmom acmHonmmoo m>Hummmz muscmooum upsoHOImcmunoou MH> UmEH0mmcmuu "mmHQMHHm> UHom Emwum mo mpHucmsq umHQmHum> ucmwcmmmc mH "mmHuHo mo gonads coHmeHumm monumnoze "mHnmcoHumHmm panama mH OHQMB 124 H N v «H AOF on.w A om om.w A om .wom muHsmmu mm o H H 0H mmmH mscHE Hmmw .v o OH H m Hmww .m o v m o mmw mo mOHHm pmmmmq .N o m H o Emmum mo moHHm Ummmmq .H om w m A mm m.w mm cm W m A mm m.w mm mmHQMHHm> ucmocmmmocH ucmHonwmou m>HuHmom uGMHOHmmmou m>Hummmz omEuommcmnucs "mmHQMHHm> mumfioumso mo Hones: “GHQMHHM> “cmpcmmmp Hm ”mmHuHo mo Hones: mumEOHmDO mo Hmnfidz mnu mo COHumEHumm mH OHQMB 125 AOh on.w A om om.w .Aom Woo prSmmH Nm H H N h mmmH mscHE Ham» .v o m o N Hmmw .m N m H N mmu mo moHHm pommMH .N o m e n Emmum we moHum pmmmmq .H om w m A mm m.w mm om w m A mm m H mm mmHQMHum> unmocmmmccH ucmHonmmou m>HpHmom ucmHOmemou m>Hpmmmz muspoooum HusoHOImcmusoou MH> Umfiuommcmup "mmHQMHum> mnmaoumso mo Hones: "mHQMHum> ucmocmmmp HN ”mmHuHo mo gonads omEHommcmua mquOHmso mo umnadz mnu mo GOHHMEHumm ON mHQMB \ 126 Table 21 Summary of Significant* Results for the Price of Steam l 2 3 4 5 6 7 8 9 Positive 4 4 11 11 12 15 12 15 35 Negative 21 18 12 12 23 9 ll 21 71 Insignifi— cant 17 20 33 33 35 46 47 34 118 Total 42 42 56 56 70 70 70 70 224 *To be counted as significant an estimate had to be signifi- cant at the 10% confidence level Column one: Column two: Column three: Column four: Column five: Column six: Column seven: Column eight: Column nine: results from model one, estimations results from model one, estimations results squares results squares results results results in 1972 results Total, from model two, estimations, 14 from model two, estimations from model two, from model two, from model two, from model two, sum of 1, 2, 5, original data transformed data ordinary least cities only generalized least untransformed data transformed data time period truncated all years 6 or 1, 2, 7, 8 Summary of Significant* Results for the Price of 127 Table 22 Gas l 2 3 4 5 6 7 8 9 Positive 3 0 5 5 7 4 9 2 14 Negative 17 16 32 32 41 34 30 45 108 Insignifi- cant 22 26 19 19 22 32 31 23 102 Total 42 42 56 56 70 70 70 70 224 *To be counted as significant an estimate had to be signifi- cant at the 10% confidence level Column Column Column Column Column Column Column Column Column one: two: three: four: five: six: seven : eight: nine: results from model estimates results from model results from model squares, 14 cities results from model squares results from model estimates results from model results from model cated in 1972 results from model Total; sum of 1, 2 I one; one; two; only two; two; two; two; two; 5, 6 original data transformed data ordinary least generalized least original data transformed data time period trun- all years or 1, 2, 7, 8 128 Table 23 Summary of Significant* Results for the Number of Customers Positive 14 10 29 23 30 33 32 31 87 Negative 4 1 12 8 9 14 11 12 28 Insignifi- cant 24 31 15 35 31 23 27 27 109 Total 42 42 56 56 70 70 70 70 224 *To be counted as significant an estimate had to be signifi- cant at the 10% confidence level Column one: results from model one; original data estimates Column two: results from model one; transformed data Column three: results from model two; ordinary least squares, 14 cities only Column four: results from model two; generalized least squares Column five: results from model two; original data estimates Column six: results from model two; transformed data Column seven: results from model two; time period trun- cated in 1972 Column eight: results from model two; all years Column nine: Total; sum of 1, 2, 5, 6 or 1, 2, 7, 8 Summary 129 Table 24 of Significant* Results for Degree Days 1 2 3 4 5 6 7 8 9 Positive 18 22 33 35 32 57 42 47 129 Negative 0 O 0 3 2 l 1 2 3 Insignifi- cant 24 20 23 18 36 12 27 21 92 Total 42 42 56 56 70 70 70 70 224 *To be counted as significant an estimate had to be signifi- cant at the 10% confidence level Column Column Column Column Column Column Column Column Column one: two: three: four: five: six: seven: eight: nine: results from model estimates results from model results from model squares, 14 cities results from model squares results from model estimates results from model results from model cated in 1972 one; one; two; only two; two; two; two; results from model two; Total; sum of 1, 2, 5. 6 original data transformed data ordinary least generalized least original data transformed data time period trun- all years or 1, 2, 7, 8 130 Table 25 Summary of Significant* Results for Retail Sales 1 2 3 4 5 6 7 Positive 18 19 17 30 20 27 47 Negative 20 15 23 15 23 15 38 Insignificant 18 22 30 25 27 28 55 Total 56 56 70 7O 70 70 140 *To be counted as significant an estimate had to be signifi- cant at the 10% confidence level Column Column Column Column Column Column Column one : two: three: four: five: six: seven: results from model squares, 14 cities results from model squares results from model estimates results from model results from model cated in 1972 results from model Total; sum of 5, 6 two; only two; two; two; two; two; or 3, ordinary least generalized least original data transformed data time period trun- all years 4 m lbs 30 I‘d U1 20 15 11 cities All reporting cities 29 31 33 35 37 39 41 43 45 YR Figure 3 Steam Sales per Customer - 11 cities - all reporting cities 132 Total Sales 30 ~nmllb 28 26__ 24 _ 22.— 20.. 18.. 16 l l l l l l l 1 J 29 32 33 35 37 39 41 43 45 YR Figure 4 Steam Sales 11 cities 133 Capital 100 95 90__ 85 80 75.. 1 L 1 1 1 l 1 1 J 29 31 33 35 37 39 41 43 45 YR Figure 5 Capital Invested 11 Cities 134 Customers Served 9500 9000 _ 8500 _ 8000 _ 7500 _ 7000 L l J l l l l I l 29 31 33 35 37 39 41 43 45 YR Figure 6 Customers Served 11 Cities 135 “GUN! 15 MAJOR NATURAL-GAS PIPE LINES tom. [mo Sums” Incas-arose: Figure 7 Natural-Gas Statistics 1930214 MAJOR NATURAL‘GAS PIPE LINES DECEMBER I954 215 \ W (Q Figure 8 room! Natural-Gas Statistics December 1954 “[3 c4 SONIC" If: ' noouuuo nus .. - .v u --— novusro am I, .1 SMCC' (bait. 136 .do a (79.1 .24 01.00.... . l I u .- r0‘J' ‘ u. .. s. Q . . . .. 6 .. . ommH I mmsHHmmHm mmw Hausumz m musmHm .0 . . WM» H ? Aw‘ .. Ion-fl .. .34 71).,“ x . \.\ .n.|lhl.ll\\a . | .. w ’ ”I ‘5‘ a ... o\m\ ............. “I. U ‘0‘.-".\( '5" ‘ . ._ . .. . .u5 .. .-.-.(... L. .2... . tl.u.....ru.. N I t. . - . . s... u 3 . . s .,. O I \o\ go. I.b.,.!1 fl . f . .W .. Q II.., J‘, I . . . . . ..I v 0 I .u- 137 OH wusmHm .14 :. 58.5.1.1... . ,.....:......11..2..... L .13 3 l...- I .3 :32...“ an: .., 5.: :3 22%.: E: 55:... ._<=...:_M._.l . .4. . a. .\ .. .2.a3..; 3...... . . .2». .\\.u\.\ \ \.» a a I .Tolfl. L. m . ..; iii-cilsliea A a.wuh.3.¢flfi ./ s . . \W a 34.0. a). .. N .1 3. . . r . ...\ \\ 43.. . . _ . I .. . .. . .3...— I . .MAQH ‘5 . , ’ . . , . . ,, , 3:1: Q N . i. in; . . . . .,.....,_ game”: 3.. l!!! I :1: in I‘ll I...- 833 f6... x.\ \ nI)... .\ \ it... .3 .32: 6.... usage «4...... assess. unseat". 138 ocmHm>mHO HH muamHm «SH 3.3 2.3 33 c2: 33 83 3.3 / \II‘I / a :K A :25 nH / , mm.— m... \/ / \I \ 8mm V ’4” II \ / /V \ . . (1,... k/ 83 1.. 8.3 833.12. . wN \ 3.3.8“ new we outalfimcm \. mam no ~23!ch— BHBZcmo NH muzmHm mmmm, mumH cuaH mme ocaH mmmH enaH ceaH KL} \\ III/L me \ /I\11\II/\\IA \II \I w VK/\ /4 V\ mom\ /\I/ mom (1 //\||\ I/fl > .I/ mam \ /I\IU // V\ \ \ \ ems \ 5:: x. \\ \ \L 9H mH ed Wu :6 m.» mama 140 OOON ooom ooov ooom ocom coop ooom NBHBZfiDO uHonumo MH musmwm m5 Om. mo 00 mm Om 0v n |_ H m _ _ m we I 1 o.H I mum ~ mom .. mH 1 mmm I o N we . fl '1 mom dmom mmHmm Emmum n we o.m I owumznom mum mo moHHm u B~BZBHHZ9~BZ m om NON Am OHm OOO moo H mHh OmmH OOO OHH mmuwumm> mmsmmm .e om m.eHH Hm I HHH bee H mam HbmH coo em bemeeO enammm .v om n I hum mm om OOOH OOO NH moan m om m I NON OH h.n OOOH OHO OH mmcmua H em a Ha I owe mm m.mH emmH Hmm em mHHbupmbom H om v.m I OON HN HH OhmH OOH m :EmnHmcom mmswmm mH O om v.OH O men HHN H.mHH mmmH OOH NH mumnmnpcom HmmHosz NN OH oom,mmN H.va bmmH H om no Am I vow ONO Ohm mmmH mH v om mm OO HNm ONO N.mHm mmmH OOO mNn EHonxooum mm5mmm mg V cm m.NH I mmm MHm m.va mmmH oom mm mcHom mg O om n.N I OOO mm O.mm OOOH OOO m mcsgcmHHom m cm O.H I OHO NH quH ObmH who Om m©>oxm v om w.m I OOO NH m.HH OhmH OOO NN cmxH>ccmm Ex 32 numcmq H030m £32 32 mmud mCHmz THSmmmHm UTHHmmsm cmoH a: comma cH Hmsm xcsua xomm moumcm owuomccou qumum mucmanmccH momHm "omscHucoo ON mHnma 197 .OmImm .mm =.cmcm3m cH OCHpmmm HUHHumHQ= .HHDS HHmz u mohflom 32 NON “M\NNOH Am muouommscmfi Hmpco 32 OOH + An pwccmHm 32 OOH x N .O OGHGOHmmHEEoo 32 NON + .O Hmouo so 32 NON Am "OmscHucou ON mHnme 198 .O .m .ummm mo mm: Hmmoum mo COHummsm d I ummm Ucm Hm3om UHHuomHm mo GOHHUSOOHN UmcHQEoo SHHZ mmcH>mm mmumcm .chHumz pmuHcD ~mousom OOO H 0.0Hm HH .m.HOO O OOH N H.OHO m m.mOO H Hmuoe OOO H H.NOO O.mHm OOm N O.mOm O.mmH ounmuo emm H H.ebH m.mm ewe m.MH e.m~ oflxb> OOO H 0.0mO H O.NmO OOO H m.ObO 0.0mm mmumumm> OON H H.ONN H h.OmO OOO N O.mOO 0.00N mHmwmmD OHO H N.HON m.mvH OOO m O.vm 0.0H mumnmnocsm OOO H O.mvm H 0.00H OOO N m.ONN O.mm EHocxooum ONO H N.ONO N.Omv. OOO H H.NOm 0.0NN mchoquoz OHN H H.ON> H O.hmO OOO m H.NOm 0.00H oEHmz emb H m.~mm m.mee eem m e.mbm e.mm meHeoHeHH Ohm H H.NO m.Nm ONH N O.mH m.O UmumHme OOO H m.OOO H m.OmO OOO N.Om O.Nm mnonmpow OOO N m.va O.NNN OHN v 0.00H 0.0N mpuom Hmmmxmun £30 32 Hmmm\mun £30 32 upon m>\OOOH Hm mb\thH ON OHQMB HoIhOIthH OHGMHN OCHummm paw Hmzom UmcHQEou OCHumem 199 Table 30 Survey of Danish Heating Supply Systems Country Town Capacity Length Capital Date therms/h of cost of mains (1963) B inst'l'n miles sterling year Denmark Brunderslev 590 8.1 204,000 1921 " Esbjerg 7,280 49.4 1,017,000 1927 " Varde 160 2.1 34,000 1927 " Randers 5,600 28.1 933,000 1931 " Slagelse 680 6.0 236,000 1936 " Herning 2,670 64.0 1,163,000 1950 " Grindsted 180 1.6 45,000 1950 " Kildong 1,740 17.5 608,000 1951 " Rodding 280 3.2 54,000 1951 " Silkeborg 1,780 14.7 550,000 1953 " Sunby-Hvorup 620 n.k. 200,000 1953 " Kristrup 260 " 157,000 1953 " Svendborg 290 3.35 192,000 1953 " Vordingborg n.k. 0.3 14,000 1953 " Jiborg 1,600 14.7 551,000 1954 " Aalborg 8,000 50.5 898,000 1954 " Frederica 760 8.7 158,000 1955 " Aarhus 400 6.95 119,000 1955 " Graasten 360 4.66 128,000 1955 " Ranum 120 2.2 36,000 1955 " Hovbjerg 880 12.1 390,000 1956 " Faaborg 420 7.15 210,000 1956 " Logstor 380 6.2 116,000 1956 " Brande 240 3.4 72,000 1956 " Vording Borg. 28 0.2 12,000 1956 " Odder 600 14.7 219,000 1957 " Kjellerup 360 5.6 131,000 1957 " Hammel 260 4.0 96,000 1957 " Norresundby 800 6.2 341,000 1958 " Aalestrup 220 4.35 94,000 1958 " Struer 450 6.6 178,000 1959 " Bjerringbro 420 8.4 202,000 1959 " Hammerlom 260 7.1 117,000 1959 " Uraa 272 6.8 94,000 1959 " Lokken 200 5.5 68,000 1959 " Vodskov 180 10.1 99,000 1959 " Vamdrup 140 1.9 65,000 1959 200 Table 30 Continued: Country Town Capacity Length Capital Date therms/h of cost of mains (1963) B inst'l'n miles sterling year Denmark Naesby 680 10.6 350,000 1960 " Hadsten 260 4.85 115,000 1960 " Assens 510 7.1 159,000 1960 " Dronnlaglund 200 4.7 100,000 1960 " Padborg 120 4.5 111,000 1960 " Ejby 136 3.5 60,000 1960 " Jelling 168 3.85 59,000 1960 " Sondenagreda n.k. 1.0 41,000 1960 " Nykohingf n.k. 1.83 101,000 1960 " Dalum 960 20.2 680,000 1961 " Hjorring 580 12.1 305,000 1961 " Vejen 260 5.85 157,000 1961 " Frederikshavn 416 4.85 169,000 1961 " Hedensted 240 5.0 101,000 1961 " Sore 168 2.2 85,000 1961 " Glamsbjerg 96 n.k. 81,000 1961 " Vejlby-Risskon 280 6.9 61,000 1961 " Nibe 220 5.7 121,000 1962 " Hillerod 368 3.35 140,000 1962 " Gentofte 100 0.95 31,000 n.k. " Rabk Mowe 48 0.56 10,000 n.k. Heating and Ventilating Research Association, District Heating: A Survey of Current Practice in Europe and America, p. 95. Source: 201 2.3 w>om< =WAI-- I-‘ Hmm>\musom CH > a mEHB :OHHMNHHHHD OO OH ousmHm OOHN OOON OOO— OOO. OOH.— OOO. I OOON . - OOO. . OOO. I OO: - OOO. -OOO. 'IIIIII- 0.01 I] oofl 30.5.5 .I-I .IIIIJ 2m.n>m .34 mo or/ OH 0" ON ON 202 Hbm>\b3o\z H numcmH OHOHoomm HO OH musmHm 9mm... H... AW I...” .1...” H." be all .,u n: «.m— r o~ mN Iwhm>m .54 “5.: < 203 man a hocmHonwm NO OH musmHm 3 c mm m 3 com r. mm o i. m .2. I ”.384 -m. 8 M3 - ONH. 1.3 “OH... 26.63 4. 1 zOHOHO 444 KG {a / Y .9 ON ,ON On 204 WHO recommended mean monthly lImit i vasteras 100 000 i Inh.‘ 957. . .. . I Linkoplng 80000 I " district heating 7O " ' famounting to: 60 .. oooooooooo ooooooooooooooooo i Uppsala 90000: " ooooooooooooo ............ '''''''''''''''''''''''''''''' Nor r kopmgiOO 000: " J 60 " ooooooooooooooooooooooooooooooooooooo 00000000 'Ot Y.‘ urns, ooooooooooooooooooooooooo . - ooooooooooooooooooooooooooooooooooooooooooooooooooooooo L ‘ "m .......... :7 ......................... 1:31:15 ' EEREREEEEEEEEEEEEESS aééktttt 'ii?2 .1 . .11 dd. 91.10 I I a. 35000; is O m 9 district heating ....... ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo ooooooooooooooooooooooooooooooooo ' Sund SVO 116000011 L . L 0 i 2 3 4 5 6 7 pphm 502 (5 pp hm: 143/ug/m3) Figure 20 Concentration of $02 in the Air in some Swedish Towns. February 1971.83 REFERENCE CHAPTER III lAllyn Strickland, Government Regulation and Business (Boston: Houghton Mifflin Company, 1980), p. 299. 2National District Heating Association, District Heating Handbook, 3rd ed. (Pittsburgh: National District Heating Association, 1951), pp. 25-27. 3Geiringer, High Temperature Water Heating, pp. 6-8. 4Lucas, "CHP and the Fuel Industries," pp. 67-71. 5United Nations, Economic Commission for Europe, Combined Production of Electric Power and Heat in Moscow (Seminar on the Combined Production of Electric Power and Heat), November 1978, p. 1. 6Neil Muir, "District Heating in Sweden," in Combined Heat and Electric Power Generation for U.S. Public Utilities: Problems and Possibilities, ed. Roland Glenn and Richard Tourin (New York: New York State Energy Research and Development Authority, 1977), p. E-3. 7United Nations, Economic Commission for Europe, District Heating in Sweden (Seminar on Combined Production of Electric Power and Heat, November 1978), p. 3. 81bid., p. 2. 9Ibid., p. 1. 10Sweden, the Swedish Institute, Fact Sheets on Sweden, May 1979, p. 2. 1 . . . International Energy Agency, Energnyolic1es and Programs of IEA Countries 1979 Review (Paris: Organization for Economic Co-coperation and Development, 1980), p. 192. 205 206 12United Nations, District Heating in Sweden, pp. 4-5. 13"Report of the Statistics Committee," Proceedings of the Seventieth Annual Conference of the International District Heating Assocation (Pittsburgh, International District Heating Association, 1979), pp. 1-26. 14United Nations, District Heating in Sweden, p. 6. 15"Report of the Statistics Committee," Proceedings of the Seventieth Annual Conference of the International District Heating Association, p. 1-26. 16United Nations, District Heating in Sweden, p. 6. 17"Report of the Statistics Committee, Proceedings of the Seventieth Annual Conference of the International Dis- trict Heating Association, pp. 1-26. 18Muir, "District Heating in Sweden," p. 55. 19United Nations, Economic Commission for Europe, Obstacles to Increased Combined Heat Power Production (Seminar on the Combined Production of Electric Power and Heat, November 1978), p. 3. 20Muir, "District Heating in Sweden," p. 57; Kjell Larsson, District Heating: Swedish Experience of an Energy Efficient Concept (Stockholm: Swedish Board of Trade, 1978). P. 5. 21Lonnroth, Energy in Transition, pp. 96-97. 221bid., p. 100 231bid., p. 101. 24Ibid., p. 102. 251bid., pp. 102-103. 26Muir, District Heating in Sweden, p. 28. 207 27Ibid., p. 29. 28Kjell Larsson, District Heating: Swedish Experience of an Energy Efficient Concept, p. 32. 291bid. 30International Energy Agency, Energy Policies and Programs of IEA Countries 1979 Review, p. 192. 311bid., p. 191. 32Larsson, District Heating: Swedish Experience of an Energy Efficient Concept, p. 31. 33Denmark, Ministry of Commerce, Energngeport 79, p. 91; W. Mikkelsen and F1. Hammer-Sorensen, "National or Localized Plans for Combined Heat and Power Developments," Papers of the Second International Total Energy Congress (San Francisco: Miller Freeman Publications, 1979), p. 2. 34Heating and Ventilating Research Association, District Heating: A Survey of Current Practice in Europe and America, p. 98. 35Denmark, Energy Report 79, p. 90; Mikkelsen and Hammer—Sorensen, "National or Localized Plans for Combined Heat and Power Developments," pp. 3-4. 36Denmark, Danske Elvaerkers Forening, Dansk elfor- syning 1978, p. 3. 37Ibid., pp. 10-11. 38Hans Neuffer, "Potential of District Heating in the Federal Republic of Germany," Proceedings of the Sixty- ninth Annual Conference of the International District Heat- ing Association (Pittsburgh: International District Heating Association, 1978), p. 578. 39Denmark, Ministry of Commerce, Act on Heat Supply, Act No. 258 of 8th June 1979, p. 8. 4°Ibid., p. 9. 208 41Ibid. 42Ibid., p. 8. 43Lennart Larson, "Development Plans for Low-Tempera- ture District Heating in Odense," in Combined Heat and Power, ed. W. Orchard and A. Sherratt (London: George Godwin Limited, 1980), p. 117. 44Ibid., p. 127. 451bid. 46Ibid., p. 123. 471bid., p. 125. 48 ‘ Denmark, Energy Report 79, pp. 91-92. . 49Denmark, Ministry of Commerce, Danish Energy Policy 1976, p. 1. 50Denmark, Ministry of Commerce, First Report from the Heat Plan Committee, p. 8. 51Thomas Kronborg, "Organization and Enforcement of Heat Planning," Papers of the Second International Total 'Energngngress (San Francisco: Miller Freemand Publica- tion, 1979). pp. 27-39. 52Denmark, First Report from the Heat Plan Committee, p. 56-62. 53Denmark, Act on Heat Supply, p. 2. 54Ibid. 551bid. 56Ibid., p. 6. 57Ibid. 209 58Ibid. 59Great Britain, Standing Group on Long Term Co- operation, Subgroup on Energy Conservation, Expert Group on District Heating, Existing Combined Heat and Power District Heating Systems, p. 1-3. 60Norman Jenkins, "District Heating in the U.K.," Energy International, July 1978, p. 37. 61Ibid., p. 38. 62Norman Jenkins, "District Heating Conference Presents Progress and Problems," Energy International, August 1979, p. 18. 63Ibid., p. 20. 64Great Britain, Secretary of State for Energy, The Combined Heat and Power Group, Combined Heat and Electric Power Generation in the United Kingdom Energy Pappr Number 35, p. 6. - 65M. E. Price, "Hereford Combined Heat and Power Station," Papers of the Second International Energy Con- ference (San Francisco: Miller Freeman Publications, 1979), p. 30. 66Ibid., p. 61. 67Great Britain, Department of Energy, The Structure of theIElectricity Supply Industry in England and Wales, pp. 6-7. 68 p. 29. Price, "Hereford Combined Heat and Power Station," 69Lucas, "CHP and the Fuel Industries," pp. 62-64. 70Ibid., p. 65. 7lIbid. 210 72Great Britain, Energy Paper Number 35, p. vi. 73Ibid., pp. 8-10. 74Ibid., p. 15. 75Hans Neuffer, "Potential of District Heating in the Federal Republic of Germany," p. 576; R. Diamant and D. Kut, District Heating and Cooling for Energy Conservation (Lon- don: The Architectural Press, 1981), pp. 445-446. 76National District Heating Association, District Heatinngandbook, 3rd ed., p. 27. 77Hans Neuffer, "Potential of District Heating in the Federal Republic of Germany," p. 597. 78Steag, Annual Report 78 (Essen, Germany: n.p., 1979), p. 22. 79United Nations, Economic Commission for Europe, Contribution of Urban Waste to the Combined Production of Heat and Electricity (Seminar on the Combined Production of Electric Power and Heat, November 1978), p. 1. 8oLucas, "CHP and Fuel Industries," p. 69. 81United Nations, District Heating in Sweden, p. 8. 821219- 83 Ibid., p. 9. 84Kjell Larsson, "District Heating: Swedish Experience of an Energy Efficient Concept," p. 6. CHAPTER IV THE VIABILITY OF DISTRICT HEATING IN THE UNITED STATES This chapter answers two questions: first, can a viable district heating system be built in the United States, and second, will such a system be built here? To answer the first question, a hypothetical system is simulated. This system is operated under a variety of test conditions. Comparative results are examined using the net present value of the project as the criterion of evaluation. To answer the second question, the present utility regulation framework is compared to an alternative framework. It will be argued that the present framework impedes the growth of district heating while the alternative would promote growth. The simulation model works in two stages. The first stage determines optimal pipe diameter sizes, given cost, technical and demand information. The second stage deter— mines the net present value of the project using investment cost based on the optimal pipeline network determined in 211 212 stage one plus demand and cost information, a stylized construction program, and a specified project life. Pipe Size Determination To determine the optimal pipe size of each section of a distribution network, the total cost of the network must be minimized. Total costs are the sum of capital costs and . 1 pumping costs. Capital costs are Ck=aDL (l) where a = the cost of one foot of pipeline of a given diameter. (It is a function of the diameter) times a capital recovery factor. D =pipe diameter L =pipeline length The annual pumping cost is calculated in the following manner . First, pumping power is Wp=_frle3 (2) 4 where f = the friction factor for the pipe material r = density of water v = the velocity of circulation of working fluid Second, the rate of heat transport is H: D2vrcAT (3) where c = the specific heat of water 213 AT = the difference between the supply and return temperature of the working fluid Solving equation three for v and substituting in equation two _ 16f1 -5 _H WP - 7273 D [m] ‘4’ H r C _ 16f1 _ . . Let ~§—§—§-- 2, then energy cost of pumping is H r c Cp = usz-5 H 3 (5) Air where u=the annual capacity factor b=cost of electricity used for pumping Total cost becomes t k p E C =C +C =aDL+uszu5 [:H] 3 (6) To minimize total cost, the derivative of cost with respect to the diameter is set equal to zero. All the other variables are assumed constant. The rationale for that assumption is explained below. Therefore the cost minimizing diameter is " 1/6 1/2 _ Suzb _H D-IaLi [mi 1/6 Let[%::b = S. Then the fluid velocity can be obtained by substituting (7) into (3). 4 -2 -1 -l v = —s c r Given v, D the pressure drop along any pipe line is Ap = -§LD-lrv2 (8) 214 However, the pressure drop is constrained by the availability of pumping equipment. In this program the largest allowable pressure drop was set at 75 PSI. Faced with this constraint, the minimum cost pipe was determined in an iterative manner that searched through that cheapest pipe until it found the pipe diameter that was compatible with the constraint. The next five sections (heat demand, pipeline length, change in temperature, capital cost, and electricity cost) will provide the assumptions made in specifying the other variables that appeared in equation 5. The Rate of Heat Transport (H) The rate of heat transport is dependent on the heat demand in the different sections of the service area. Heat demand is a function of the outside temperature, design temperature, inside design temperature, building structure, and the demand for hot water for direct consumption (sanitation, etc.). In particular, heat demand is the heat needed to raise the indoor temperature to 62°F when the outside temperature is at the design temperature. For this study the design temperature was set at 21°F. It was assumed that internal sources are capable of raising the temperature the final 10 degrees to the indoor design temperature of 72°F.2 The 215 design temperature was set relative to the typical winter climate. This study used the New York City weather pattern as the typical climate. In that climate there are normally only 133 hours in which the temperature drops below 21°F. (One would hope these hours occur at night with everyone under blankets.) At outside temperatures above 21°F, heat demand was assumed to vary proportionately to the ratio of the difference between 62° and the outside temperature to the difference between 62° and 21°F. The housing stock was assumed to be two story and four story apartment buildings. Each two story building con- tained 18 apartments. Its peak heat demand was 259,000 BTU/ hour, and its annual heat load was 1080 x 106 BTU. The four story building heat demand and load was calculated by doubling the two story building estimates.4 Hot water demand for consumption purposes was set at 58 gallons per person. The water temperature was raised by 80°F. These parameters translated into a building (for the two story apartment house) peak demand of 58,000 BTU/ hour and an annual load of 50.8 x 106 BTU.5 Pipeline Length Population density and the housing pattern determine pipeline length. Three population densities, (10,000, 20,000, and 30,000 people per square mile) were chosen for 216 study. The apartment houses were set in a rectangular grid. Increases (decreases) in density were achieved by moving the buildings closer together (farther apart). As the location of the buildings was moved, pipe lengths were changed accordingly. Translated into thermal loads the above-mentioned heat demand became 13.5, 27, and 40.5 megawatts (mw) per square mile respectively. The Pine Study, from which the standard building heat demands were taken, used density patterns of up to 15,000 persons per square mile or equivalently 20 mw per square. The Pine Study stands alone as the only study to show that district heating is feasible at those den- sities.6 Other studies state that below 52 mw per square mile district heating is not feasible.7 However, many of these studies used obsolete pipe construction techniques. New techniques incorporating different materials and construction practices reduce the pipeline costs, might allow district heating to become feasible in areas previously ignored. To investigate this possibility, this study uses densities below the old rule of thumb standard of 52 mw per square mile. Change in Temperature The difference between a supply temperature of 300°F and a return of 210°F set the change of temperature at 90°F. 217 These temperatures are significantly higher than the typical European temperatures of 250°D supply and 160°F return. The need for the higher temperatures is twofold. First, higher temperatures are needed in the U.S. to run absorp- tion cooling equipment; the Europeans do not provide air cooling services. Second, the higher temperature is needed to run steam generators. These generators provide low pressure steam that is used in older buildings with anti- quated heat systems. Providing the higher temperature will allow these customers to purchase the heating service with— out incurring major retrofit‘costs.8 Once the temperature parameters are set it is possible to obtain estimates of two other variables observed in equation six above. These variables are the specific heat and density of water. Capital Cost Capital cost per foot is the product of a capital recovery factor and the original cost of the pipeline. The capital recovery factor, in turn, depends on the project life and the interest rate. The project life was set at 30 years. Three interest rates, 5, 10, and 15 percent were used in the calculations. The optimal pipe size was found to be independent of the interest rate. 218 Estimates of pipeline cost vary by large amounts (see Table 31). The cause of the variance can partially be explained by construction technique and pipe materials. However, even when these factors are held constant, the data still show sharp differences. For example, in Table 31, construction technique and pipe materials are the same for columns 1 and 4, for 2 and 7, and for 3, 5, and 8. There are two standard construction techniques, field fabricated, and pre-fabricated. The field fabricated technique can be subdivided into those methods that con- struct concrete ducts, and those that pour concrete into the trench. The concrete duct field fabrication technique involves at least ten different construction steps: excavation, laying a concrete base, forming the walls, placing the pipe in the form, insulating the pipe, waterproofing, placing the drainage pipe in the duct, and a three-step covering process. (The finished product is shown in Figure 21.) The poured concrete technique, shown in Figure 22, is slightly easier to construct because it eliminates the drainage pipe, the waterproofing, and the need to place a roof over the duct.9 In both cases a steel service pipe is used. Traditionally, mineral wool has been used as the insulator. The temperature of the heating medium can be 219 10 (Steam raised to 2192°F without damaging the system. systems operate at 400°-450°F and hot water systems operate at 200°-300°F.) The pre-fabricated techniques involve only six con— struction steps: excavation (usually the trench width is only 60 percent of the width needed for field fabricated pipelines for the same size service pipe), assembly of pipes, insulation of joints, laying amuiaround the pipes, back-filling, and replacing the surface.11 The pre-fabricated pipe can be either steel-in-steel pipe or steel-in-plastic. The steel-in-steel pipe consists of steel service pipe wrapped with insulation of either calcium silicate or polyurethane. The outer mantle is 10-gage steel protected by glass fibre reinforced bitumen. The temperature range of the insulation is up to 1200°F12, (see Figure 23). The steel-in-plastic pipe consists of a steel service pipe wrapped in polyurethane. The mantle pipe is a polyethylene protective sleeve. The insulation package can be altered to be viable at temperatures of up to 248°F or 338°Fl3, (see Figures 24 and 25). The decision to use field fabrication versus pre- fabricated depends on the relative price of labor and materials. The field-fabricated technique uses more labor during construction while the pre-fabricated service pipe is more expensive. (Also, the speed of construction is faster with pre-fabricated systems, and the inconvenience 220 of and the third party expenses related to the construc- tion project are smaller.) As the pipe size increases, service pipe cost as a percent of pipeline cost increases. This fact has led several European experts to recommend pre—fabricated pipelines where the service pipe is 8" or less; and field fabricated pipe when the service pipe is 10" or more; with the 8"-10" range to be zone of indeterminancy depending on local conditions.14 Historically, in the United States, the concrete duct field fabrication was the most important method. The New York Steam Company perfected this method in the early 1900's. Almost all the pipelines in use today were constructed in that manner. The alternative techniques have been used in Europe since the early 1960's.15 Data recording the pro- portion of recent pipelines completed by construction technique in the United States are not available. However, given that the several feasability studies of district heating written in the late 1970's did not even price the alternate techniques, it seems reasonable to conclude that pre-fabricated pipelines have not been used in large numbers in the United States. For the purposes of this study, the pipeline cost estimates provided in the Piqua study were adopted. These pipelines were pre-fabricated, using steel-in-plastic pipes. This adoption provided the study with a reasonable estimate 221 of the cheapest pipeline. Later, the cost will be inflated by a factor of 1.2 and 1.4 to see the impact of higher pipe- line cost on project feasibility. Electricity Cost The cost of electricity used to pump the hot water through the system was set equal to the average price of electricity to industrial concerns in the United States in 1980. This price was 3.4¢ per kilowatt hour.16 This price was adjusted in every year for inflation. In the base case the inflation rate was set at 7 percent. Total Pipeline Cost Given the above inputs it is then possible to determine the Optimal pipe sizes for each section of the service area. Summing across sections provides an estimate of pipe needs by pipe size. Table 32 shows the results of the optimal pipe model by density classification. The transmission pipe was extended or shortened depending on designated plant location site. Origin cost of the pipeline was then determined in 1980 prices by multiplying the price per foot by the number of feet of each pipe size, and then summing across pipe size. 222 Distribution network pipe size varied from 2" to 10" diameter. In this size range the European practice is to use pre-insulated pipe. This study followed the European practice in that range. The transmission pipe was 12" and 13" in diameter. For these sizes, the European practice is to use field fabricated concrete duct pipelines.17 This study continued to use pre-fabricated pipe for the larger pipe sizes. The latter practice allowed the use of one source for all pipeline cost estimates. Net Present Value Determination District heating projects entail multi-year construc- tion programs. Revenues and operating costs can begin only after the construction program has advanced to allow the system to go into partial operation. Because of this extended time dimension, the net present value criterion was chosen to evaluate the success or failure of the projects. In this study, the pipeline system was built in four phases over a ten year period. Each section serves a population of 54,000 individuals. Peak heat demand is 8 BTU/hour and annual load is 8.2 x 1011 BTU. 3.04 x 10 This construction schedule implies that revenues and operation costs began for sections 1 through 4 in years five, seven, nine and eleven, respectively. A phase or 223 section took four years to complete, one—fourth of the phase built per year. Revenue Revenue received by the project is a function of three factors: the annual heat load given one hundred percent participation in the system, an attraction rate that allows for less than one hundred percent participation in the systems, and the price of a competitive fuel. The annual heat load was determined previously in the pipeline size model. It is based on the average tempera- tures for the New York City climate. It will be assumed that the climate for each year of the next thirty years will be identical to the average climate. The attraction rate is the percent of the potential customers who are connected to the system. The rate is allowed to move through time. An initial rate is set in the year the system starts to operate. Every year thereafter, the rate is increased until it hits a final attraction rate in the 30th year. The price of heat was set at ninety percent of the energy equivalent price of natural gas. The latter price was determined by assuming that the average boiler operates at a seventy percent efficiency rate, and that the average 224 customer purchases gas at the national average gas price for commercial customers.1 Total revenue for any given year is the product of the maximum load times an attraction rate times the energy equivalent price of gas times ninety percent. Other studies have used boiler efficiency rates of between sixty and seventy percent, attraction rates of between seventy and one hundred percent, and used the price of fuel oil instead of the price of gas.19 The reason for multiplying the energy equivalent price by ninety percent was to provide an incentive for customers to join the system. The ninety percent rate was used in the Detroit, Michigan and Hereford, England studies.20 Costs Total costs are the sum of investments; heating costs, pumping costs, maintenance. Each cost was first estimated for 1980. For all other years costs were increased by an inflation factor. Investments cost determined by the original cost estimate for a pipeline section, the construction schedule and the construction inflation rate. Original cost estimate is shown in Table 2 and the pipeline construction program is shown in Table 4. 225 Heating costs are set equal to the loss of revenue received by the electric utility due to the reduction of electricity output caused by the sale of steam. The loss of electricity output is the result of removing the steam from the turbine prior to the completion of the turbine cycle. This removal of steam causes a loss of electricity because the steam still contains energy that could have been transformed into electricity. A measure of this energy is the temperature of the steam. Further, the efficiency of a heat engine depends on the difference between the temperature as it enters the turbine compared to the temper- ature as it leaves. A very efficient system would have the sun at one end to heat the fluid, and Lake Superior at the other end to cool it. To be more precise the thermo- dynamic efficiency of a heat engine is given by the following . 21 equation: where N = efficiency T1 = temperature of the entering steam, degrees ' O Kelv1n ( K) T = temperature of the exhaust steam, degrees Kelvin The efficiency of the steam plant is dependent on not only the thermodynamic efficiency but also the boiler, turbine, 226 and generator efficiencies. The efficiency of the plant is approximately 55 percent of the thermodynamic efficiency.22 In a typical 800 MWe turbo-generator, steam leaves the boiler at 2400 PSIG and 1000°F. It enters the condenser at 1.5 in HgA backpressure and at approximately 70°F. The thermodynamic efficiency of this system is 64 percent (811°K - 294°K/811°K), and the plant efficiency is 35 per- cent. In order to raise the temperature of the heat supply water from its return temperature of 210°F to its send out temperature of 300°F it is necessary to remove steam from the turbine at 320°F and 91.1PSIG. This process reduces the thermodynamic efficiency to 47 percent (811°K - 433°K/811°K) and the plant efficiency to 26 percent.23 The additional electricity that the removed steam would have generated by continuing through the turbine to the condensers is the electricity loss charged to the district heating system. This loss can be reduced if the heating fluid is heated in multiple stages. By using the multiple stage process a portion of the steam is allowed to progress further through the turbine before it is removed for heating purposes, and in turn the steam will generate more electri- city. A two stage heating system was used in this study because a majority of the savings due to multiple-stage heating are saved in the second stage.24 227 Heat cost, which is equivalent to revenue loss, is obtained by multiplying the output loss times the electri- city price. The choice of electricity price has varied across studies. Some studies use the busbar cost of the plant producing the heat. Others use typical baseload busbar costs. The proper price to use depends on the particular pricing philosophy. Alternative pricing strategies were discussed above in reference to the Consolidated Edison Steam Cases. In the simulation model a price is quoted and compared to alternative prices. Pumping costs are a function of pumping energy require- ment and the cost of electricity. The pumping energy requirements were determined in the pipeline size model. The cost of electricity was set at the average national rate for industrial customers. Maintenance costs were estimated at 3 percent of pipe- line replacement costs. For the first ten years, these latter costs were set equal to the total actual pipeline investments. Starting in year eleven replacement costs were set equal to the actual costs as of the tenth year times the construction inflation factor. This practice results in high maintenance cost relative to other studies that set maintenance costs at 2 to 5 per- cent of the initial investment costs.25 The consequence 228 of this assumption is that any underestimation that may have occurred in other cost figures will be compensated for by an overestimation here. Results A base case analysis was determined for service areas with population densities of 10,000, 20,000, and 30,000 inhabitants per square mile. The following assumptions were made: -The transmission distance from the plant to the service area was five miles. -The attraction rate was set at .75 + .008(I-5). This rate implies that at the start of operations seventy-five percent of the customers signed up; and by the 30th year of operations ninety-five percent of the potential customers had joined the system. -The construction inflation rate was 5 percent. -A11 energy cost inflation rates were 7 percent. -Industria1 customer electric cost was $.034/ Kilowatt hours. -The busbar cost was $.024/Kilowatt hours. -Natural gas energy equivalent price was 4.77E—6 dollars/BTU. 6 The results of these trials are shown on Tables 33, 34, and 35. The net present value of the trials was 3.15 x 107, 6.52 x 107, and 7.39 x 107 for densities of 10,000, 20,000, and 30,000 inhabitants per square mile respectively. The trend that benefits will increase as density increased 229 was expected. The fall in capital, pumping and maintenance cost as a function of density are responsible for the trend. The positive results indicate that the project should be adopted. However, these results are obviously dependent on the reasonableness of the assumptions. The next seven sections will test that reasonableness through a sensitivity analysis on the selected variables. Transmission Distance The distance from the dual purpose plant to the heat service area is the transmission distance. The pipeline connecting the two principal components of the system carries the entire heat load. Consequently it is the largest and most expensive pipe in the system. Three alternative plant locations, 1, 5, and 10 miles from the service area, are compared for each population density. A distance of ten miles was considered the maximum feasible distance in the Battelle and Pine studies.27 As anticipated, the net present value of project declined as the distance increased. In only one instance did the net present value turn negative (10 miles and 10,000 inhabitants per square mile). For the higher two density cases the magnitude of the change was large, increasing by more than 140 percent as the plant moved from 10 miles to one mile away from the service area. 230 Interest Rate Changes in the interest rate, holding all other variables constant, had a larger impact on project feasibility than changes in any other variable. The range of results (see Table 37) stretches from 2.31 x 108 to -1.28 x 107. This impact is caused by the length of project, 30 years, and the timing of revenue and costs. Revenue increases in the latter years of the project due to the increase in the attraction rate and the natural gas infla- tion rate. Capital costs cease after the first ten years. Maintenance costs increase at a slower rate than energy costs. The confluence of these trends produce high nominal profits in the latter years of the project. A high interest rate would reduce to a great extent the present value of the large nominal profits earned in the final years of the project life as compared to a low interest rate. For example, a profit of 7.3 x 107 earned in the 30th year has a present value of 1.78 x 107 with a 5 percent interest rate but only a .128 x 107 present value with a 15 percent interest rate. Changing the interest rate by itself entails changing the real interest rate. The base case contains a real interest rate of 3 to 5 percent (the nominal interest rate, 10 percent to either the energy inflation rate of 7 percent 231 or the construction inflation rate of 5 percent). This real rate of interest is equal or slightly higher than the approximately 3 percent real interest rate that existed on corporate AAA bonds over the period 1960-1980.28 The comparisons shown on Table 37 change the project interest rate to 5 and 15 percent or equivalently to real rates of interest of -2 to 0, and 8 to 10 percent respectively. A range of -2 to 0 real interest rate existed in the 1970's for Treasury Bills but not for any long term bonds.29 In 1982, the real interest rate for corporate bands has hovered around 7 percent.30 Therefore, the comparative interest rates shown capture the range of real interest rates encountered in recent history. The fact that net present value of project for the higher two density cases was positive even when the interest rate was 15 percent demon- strates that high interest will not cause project cancella— tion even though they will reduce the project's value. Energy Inflation Rate The energy inflation rate adjusts the price of natural gas, the busbar cost, and the pumping costs. The adjustment to natural gas prices changes revenues. The adjustment to busbar cost and pumping cost changes total cost. Because natural gas price is the sole basis for revenue calculation, and total costs include other factors beside busbar and 232 pumping costs, increases in the energy inflation rates will increase the net present value of the project (see Table 38). The assumption that all energy prices increase at the same rate has a conservative impact on the results as compared to other studies that allow gas prices to increase faster than coal prices.31 Increases in coal prices have a greater impact on the price of electricity than increases in natural gas prices due to the fact that coal is responsible for 51 percent of electricity generation while gas is responsible for 15 percent.32 The price of electricity determines the energy costs of district heating. Therefore lower coal inflation rates will lower the energy costs and increase the difference between revenue and costs compared to other inflation scenarios. To assume that all energy costs increase at the same rate implies that not only coal and gas prices increase at the same rate but that all other inputs into electricity production increase at the same rate. While the latter assumption might seem to be heroic it was made to erase any impression that assumptions have been made for the purpose of insuring project feasibility. Busbar Cost Busbar cost is the cost of electricity at the plant. It is used to determine the revenue loss to the electricity 233 subsidiary when a plant simultaneously produces electricity and heat._ The base results case used a busbar cost of .024 cents per kwh for 1980. This is the cost of electricity at large base load plants.33 This price is equivalent to a short-run fully distributed (or actual average cost or account cost) off-peak cost. The simulation model was run using two alternative busbar prices, .030 and .036 cents per kwh for 1980. The price of .036 cents was considered reasonable for a high price because it was above the 1980 average kwh price for industrial customers.34 The results of these alternative cases are shown on Table 39. As expected higher busbar costs are associated with lower net present values. Pipeline Cost Pipeline costs are the sum of the costs of the trans- mission pipeline and the distribution network. It was first estimated as if it were installed instantaneously in year one of the project. Then the cost was adjusted to accomodate historical construction time and inflation in construction costs. Base case results are compared to two higher cost estimates. These estimates were obtained by multiplying the instantaneous installed cost by factors of 1.2 and 1.4 234 and then adjusting the new initial costs for inflation incurred during construction. The alternate cost estimates approximate the relation- ship between columns one and four in Table 31 (except at the 2-inch pipe diameter size). Column four costs are the costs used to determine the base case estimates. Column one costs represent an alternative estimate of the steel- in-plastic construction technique. The increase in pipeline costs, as expected, reduced the net present value of the project for each density level (see Table 40). However, only in one case, at a 1.4 cost factor for 10,000 inhabitants per square mile, did the net present value turn negative. Nominal Rates Changes in nominal rates refer to changes in the under- lying inflation rate. The catalyst for these changes is usually a change in federal government policy, rather than a change in a particular industry. For example, the large increases in defense spending could cause the aggregate demand curve to shift or the aggregate supply curve could shift given a reduction in the dead weight loss following the death of the disabled who have been removed from the Medicaid rolls. The consequence of such a change on the 235 simulation model would be to increase or decrease construc- tion costs, energy costs and interest rates simultaneously. The comparative results for three different sets of nominal rates are shown on Table 41. As the rates increased, the net present value decreased by small amounts. This pattern is the result of the relative increased costs in the early years leading to larger negative surpluses in the first eight years of the project, and higher relative revenues in the later years leading to higher positive surplus in last years of the project. When the new pattern of surpluses was discounted back to the initial year at the new higher discount rates, the impact of the increased costs was greater than the impact of the increased revenues, so the new present value fell. Attraction Rate The attraction rate specifies the percent of the ser- vice area heat load that is connected to the system. In the base case the percent started at 75 percent in first year of operation (the fifth year of the project life) and increased at a constant rate until it hit 95 percent in the 30th year. In the two alternative cases, the attraction rate started lower, at 65 and 70 percent, and ended up at the same rate, 95 percent. 236 The high final attraction rates are consistent with the low relative price of steam. Steam heat is always priced at 90 percent of the cheapest alternative fuel. Therefore, it would be rational for all new buildings and old buildings that have heating systems that need major renovation to join the system. A period of thirty years is probably long enough for most buildings in a service area to fall into one of the two categories. Further, the high delivery temperature of the water allows for the retrofit of old buildings at minimum cost. Setting the starting attraction rates calls for pro- fessional judgment, marketing expertise and a lucky guess. The three starting attraction rates compared in this study are below attraction rates used in other studies.35 Comparative results are shown in Table 42. For each density level, the net present value decreased as the starting participation rate decreased. These results were expected because capital and maintenance costs do not decline with the decline in customers, while revenues do. Summary The simulation model was run 45 times. In 40 cases (89 percent) the net present value of the project was positive. All of the negative cases occurred at the density level of 10,000 inhabitants per square mile. In 237 27 cases (60 percent) the value was positive by the twentieth year of the project life. This cut off date was important because first, the estimate became more uncertain the longer the time horizon, and, second, several other studies of district heating feasibility stopped in the twentieth year. In 13 cases that turned from positive to negative with the shortened project life, seven were associated with 10,000 inhabitants per square mile density level, and two of the others were associated with a 10- mile transmission distance. Three conclusions can be drawn from this feasibility study. First, district heating is feasible for service areas with a density level of 30,000 inhabitants per square mile. Second, district heating is not feasible for service areas with a density level of 10,000 inhabitants per square mile. Third, district heating may or may not be feasible for service areas with a density level of 20,000 inhabitants per square mile. The information needed to translate these conclusions into practical policy guidelines for real American Cities is not readily available. It is possible to obtain popula- tion densities on a city-wide basis. Table 19 lists the population densities of nineteen American cities. The first fifteen cities were chosen from the largest 75 American cities by choosing every fifth city. The last four cities were chosen because additional information is known about 238 them.36 If the densities shown in Table 43 were used as a policy guideline then this study would conclude that district heat is definitely feasible only in New York City, is worthy of investigation in Newark and San Francisco, and is not feasible any place else. However, very few district heating systems serve entire cities. Instead the service areas usually contain only parts of each city. Due to the fact that heat density vary within each city (for example Figure 1 shows a variance from 5 to 200 megawatts per square kilometer in the Minneapolis-St. Paul metropoli- tan area) city-wide averages cannot be used a policy guide- lines. An alternative is to examine other feasibility studies performed on real cities to obtain data for densities in possible future service areas. Here the data are only suggestive because of the limited ability to translate the figures provided into a single framework of analysis. Many studies do not provide data on design temperatures, size of service areas in terms of land mass, distribution of customer types, or population densities. Without that data it is impossible to make accurate comparisons. In three studies it was possible to make some rough estimates. Returning to Figure l,heat load densities for downtown St. Paul and downtown Minneapolis are approximately 40 and 70 megawatts per square kilometer respectively. These numbers translate roughly into population densities of 239 50,000 and 87,000 inhabitants per square mile. These densities are significantly higher than city-wide averages for St. Paul and Minneapolis reported in Table 19. Further, they are above the test densities used in this study so that one could conclude that district heating would be profitable in these areas. Second, the Detroit study provided heat demand and acreage by census tract. Summing across census tracts provided an estimated 16 megawatts per square kilometer for the proposed service area. This heat demand is equivalent to the heat demand for 20,000 inhabitants per square mile used in this study.37 Three other characteristics of the Detroit study are worth noting. First the distribution network was to be constructed using field-fabrication techniques.38 Second a large proportion of the pipe would have been less than 8" in diameter.39 Current practices in Europe dictates the use of pre-fabricated pipe for those sizes. Third the plant providing heat for the service area at Conners Creek is within the borders of the service area.40 Therefore a transmission pipe connecting the plant to the service area does not have to be built. Under these conditions this study would have recom- mended that Detroit Edison build the new system. However, the conclusion of the Detroit Edison study was to abandon construction plans unless the city of Detroit subsidized the project. 240 A second proposal of the Detroit study was to examine the feasibility of a smaller service area. The heat density of the smaller area was 19 megawatts per square kilometer or the equivalent of 30,000 inhabitants per square mile. Again this study would recommend construction of the system. It is not known if Detroit Edison ever completed the analysis on the second service area.41 The major cause of the different recommendations was pipeline costs. For pipes with an 8" diameter or smaller Detroit's costs were higher by a factor of 1.6 to 2.2 than costs used in this study. Third, the Piqua study divided the city into 52 heat zones. Heat demand and acreage was provided for each zone. It was decided to study the feasibility of district heating for a service area containing 12 zones. The heat density for the proposed service area was 54 megawatts per square kilometer.42 It is difficult to generalize for the entire country from a sample of three cities. However, the examples pro- vided show that in both large and small cities there are regions where district heating could be profitable. A comparison of nine other feasibility studies is shown in Table 44. Two curious correlations appear in that table. First, the closer to a privately-owned public utility is the performing agent (that person(s) who actually prepared the study) or sponsoring agent (that person(s) who 241 can hire or fire the performing agent; the sponsoring agent is not necessarily the agent who pays for the work) the more likely the project will be found to be not feasible. Second, the closer to either the European district heating industry or the American nuclear power establishment the performing agent or sponsoring agent is, the more likely the project will be forced to be feasible. The above comparisons imply that the self interest of the person conducting the study determines the outcome. The recent history of existing privately-owned U.S. district heating utilities has led to a disenchantment with the industry. An analysis of Federal Energy Regulatory Commis- sion data for 31 firms demonstrates the current situation. Only 15 firms earned a positive return on net fixed assets dedicated to the heating subsidiary. Their average return was 6.6 percent. For the remaining 16 firms, the average loss was 9.9 percent. Further, the steam revenues repre- sented less than 2 percent of average companies' gross revenue.43 Thus, from a company perspective, the conclusion becomes "why bother with a tiny business that will probably lose money anyway?". On the other hand, agencies and individuals tied to the nuclear power business would have an incentive to pro- mote anything that increases the economic viability of nuclear power. Given that a nuclear plant does not have stack losses (one would hope), heat from a nuclear plant has 242 a lower cost than heat from a fossil fuel plant. There- fore, a district heating business that buys heat from a nuclear plant rather than from a coal plant potentially will be more profitable. This logic appears to this author to be the only reasonable explanation for the long and extensive research into district heating sponsored by Oak Ridge National Laboratory. Regulatory Practices Given that new district heating systems are economi- cally feasible, it is incumbent upon the investigator to explain the general lack of interest in building one. Invariably, the answer is to blame the present regulatory system. I. Oker, for example, states that while economic, technological and environmental aspects of district heating are favorable, institutional barriers remain a major deterrent to implementation.44 Peter Donnelly and Isiah Sowell state that "profitable investment in such facilities hinges on resolving regulatory treatment of issue such as: (1) joint cost allocation, (2) use of innovative financing techniques; . . . (Uncertain) Regulatory treatment of each of these issues . . . has a chilling effect on attempts to promote district heating."45 These issues focus the controversy on a single question: can a district heating entity stand by itself 243 (earn the allowed rate of return without a subsidy from the electric utility)? To answer this question it is first necessary to determine the allocation of joint costs of heat and electricity generation. This allocation simultane- ously sets the legitimate costs and profits of each subsidiary. The regulatory commission must approve a particular allocation scheme. Different approved schemes could make or break a district heating project. Second, each district heating project is a multi-year endeavor that will lose money in the early years of its life and earn money in the latter years. It is therefore necessary to borrow in order to finance the early years' losses. If an electric company finances a district heating subsidiary's losses through higher than otherwise electric rates or lower than otherwise rates of return to owners, is this flow of funds automatically a subsidy? That depends on how subsidy is defined. If a subsidy occurs in any year in which a district heating subsidiary does not stand by itself, then a subsidy must be paid. However, when other definitions are used, the answer is not quite so clear. Gerald Faulhaber presented two definitions: 1. informally a subsidy does not occur "if the pro- vision of any commodity (or group of commodities) by a multicommodity enterprise subject to a profit constraint 244 leads to prices for the other commodities no higher than they would pay by themselves;"46 or 2. (formally) "Price for which the resulting revenue vector lies in the core of the game."47 The two definitions are identical because a revenue vector which is outside of the core leads to prices that are higher than would be paid if each commodity were supplied by a separate enterprise. Faulhaber also quotes a definition of a subsidy pro- vided by Harry Trebing: A more meaningful standard might relate maximum rates to the cost of a single purpose facility or system built to serve the user requirements of the particular group most affected by the upper price limit. If this group paid a rate in excess of the cost of the single purpose facility, it would be subsidizing other user groups as well as failing to partici- pate in any of the economics of the joint cost inherent in the public utility operation. While the Trebing definition emphasizes the losses incurred by the subsidizer, and the Faulhaber definitions emphasize individuals' decision to play with the group (according to Faulhaber a person receiving a subsidy inside the group might remain outside if there are potential gains from doing so), neither definition directly tackles the issue of multi-year multi-commodity projects. To do so, a corollary to the above definitions must be given. First, the net present value of the revenue requirement decreases (increases), then the wealth of the 245 customers increases (decreases). A subsidy will occur, "given a utility with two or more subsidiaries, if a flow of funds between these subsidiaries leads to the net present value of the revenue requirement for the customers of one subsidiary being higher than it would have been had the subsidiary been an independent entity."49 The implication of the corollary, given the problems of joint cost allocation and financing, can be demonstrated through an example. The example will compare a single- purpose electric utility and gas service for heat demand group to two alternative cogenerating electric and district heating systems. The difference between the alternatives centers on the joint cost allocation scheme. The technical choices available to the electric utility are shown in Figure 26. Part I details the energy cycle efficiencies for an electric plant operating either in single- purpose or dual-purpose mode. It shows that 14 kilowatt hours (kwh) of electricity must be sacrificed to obtain 62 kwh of useful heat per every 100 kwh of energy consumed. Part II illustrates the alternatives given that a boiler can instantaneously burn 3080 kw of energy. Part II alternatives 1 and 4 are based on Part I-a fuel efficien- cies. Part II alternative 2 is based on Part I-b fuel efficiencies. Part II alternative 3 is a hybrid. Steam containing 1000 kwh is allowed to expand through turbine to the condenser, producing 400 kwh of electricity, and 246 steam containing 1540 kwh is extracted at 300 degrees F, generating the additional 400 kwh of electricity and 955 kwh of useful heat.50 The demand for electricity is divided into a summer peak and winter off-peak seasons of 4380 hours each. Summer demand level is constant at 1232 kw and the winter demand level is constant at 800 kw. The single purpose electric utility would meet those loads operating at Figure 6 Part II and 4 levels. Given a price of electricity $.024/kwh, annual revenue is $213,604. It is assumed that the company is run efficiently and regulated properly so that it is earning equal its allowed rate of return, which is the true cost of capital. Therefore economic profits are zero. Gas sales occur only in the winter meeting the heat demand. This demand is set equal to the maximum heat sales that could be serviced by the electric utility's boiler. This amount to an annual sale of 38,739 MCF (see Table 46 for a listing of assumptions and definitions that generate this number). Given a price of natural gas of $3.34, per MCF annual revenue is $129,388. From this position the electric utility decides to start a district heating subsidiary. The project life is collapsed into two years. Year one represents the time period over which the subsidiary's profits are expected to be negative, while in year two, profits are expected to be 247 positive. An increase in the heat attraction rate from 50 percent to 100 percent is the cause of the change in expectations. District heating revenue is set at 90 percent of gas revenue for the relevant heat sales (50 or 100 percent of gas sales). Non-fuel costs are set at $50,000 annually. Full costs are set at $.01 kwh. The number of kwh charged to the district heating subsidiary depends on the joint cost allocation scheme. An incremental scheme charges the heat customers only for that energy used above the energy needed to serve the electric customers. A proportional scheme charges the heat customers in proportion to the energy in steam sales relative to electric sales (see Table 47 for calculations). Alternative profits of the district heating subsidiary are shown on Table 48. Using the incremental joint cost allocation scheme, the subsidiary loses money in year one but earns a positive profit in year two. Using the pro- portional joint cost allocation scheme, the subsidiary earns negativeprofits in both years. Alternative profits of the electric subsidiary are shown on Table 49. Its profits are the sum of profits earned on electric sales, profits earned on sales to the heating subsidiary and (the flow of funds to or from the heating subsidiary). The electric business, being perfectly 248 regulated, earns zero economic profits. Profits from heat sales are zero under the incremental joint cost scheme and are positive under the pr0portional scheme (heat revenue is greater than additional costs). The flow of funds to the, heating subsidiary are the reverse of heating subsidiary profits. Total profits of the electric subsidiary do not vary given the different allocative schemes; only the source of the profits varies. Assuming a 10 percent discount rate, the net present value of the project in year one dollars is $1976 to the electric utility.‘ Given these facts, should the electric utility be allowed to operate the district heating subsidiary? The answer given by a public utility commission charged with establishing just, reasonable, non-discriminatory rates must be "no" if a subsidy exists. Has a subsidy been paid in this example? The answer obviously depends on the definition of "subsidy." If a subsidy occurs when in any year funds flow from a subsidiary to another, then there has been a subsidy. If a subsidy occurs when the sum of the discounted profits of any sub- sidiary are negative, then the occurrence of a subsidy depends on the choice of joint cost allocation schemes. The proportional scheme insures negative profits for the district heating subsidiary in both years; thus there is no discount rate that would allow its discounted profits 249 to be positive, so a subsidy did occur. However, if a subsidy occurs only when the net present value of revenue requirements has increased, then the project is subsidy free. The net present value of heat customers' payment must decline because heat payments drop in both years (5 percent in year one and 10 percent in year two). The net present value of the revenue requirement of electric customers drops if any discount rate less than or equal to 24 percent is used. Given that discount rates of greater than 24 have not been used by public utility commissions (to this author's knowledge), the electric customers could gain even if rates increased by the full amount of the electric subsidiary's loss in year one. Regulatory Reform A utility commission that accepts the net present value definition of subsidy and is aware of a district heating project that has a positive net present value still has one more task: it must persuade a reluctant, skeptical utility to undertake the project. A possible solution is to institute an incentive scheme. The incentive schemes must meet three criteria: first, there must be a direct link between the incentive offered and the performance desired. Second, the size of the incentive must be set high enough to promote the 250 project but not so high that the owners benefit at the expense of the customers. Third, the incentive can produce counter-productive tendencies. The commission must have the ability to recognize and eliminate these tendencies. A scheme that increases (decreases) the utility's allowed rate of return as the BTU conversion of the steam electric plants rises (falls) will meet the first criterion. The BTU conversion rate is defined as the BTUs contained in output divided by the BTUs in the inputs. This rate is the inverse of the heat rate presently used to measure utility plant performance, giVen the kwh are converted into BTUs. (For example, a plant with a heat rate of 10,000 BTU/kwh will have a BTU conversion rate of 3413/10,000 or 34 percent; note: 1 kwh = 3413 BTUs.) An electric utility that converts its plant from a single purpose electric facility to a dual purpose facility will automatically increase its BTU conversion rate. Using the stylized facts presented in Figure 6, the conversion rate rises from 40 percent to 88 percent. The size of the incentive, the increase in the allowed rate per increase in BTU conversion rate, must be project specific. It will be a function of the potential net benefits and the size of the project relative to the utility's normal generation. If the net benefits are larger, the incentive can be small. If the project is large relative 251 to the utility, it will have a large impact on the utility's fuel efficiency so, again, the incentive can be small. The potential counter-productive tendency that this program might encounter would be the ability of the utility companies to over estimate future benefits. In this case, the electric customer pays for the incentive and the losses in stylized year one but never receives the benefits in stylized year two because year two never appears. Given today's environment, which includes the reluctance of electric utilities to expand district heating subsidiaries and so-called finance hardship of the electrics, the prob— ability of a electric utility starting a project that could cause it financial damage from irate electric customers and commissioners who feel dupped seems small. Further, com- missions are called on every day to evaluate projects whose benefits will occur in some future time period. A commission staff that has the ability to evaluate the future benefits of a nuclear power plant should be able to evaluate the future benefits of a district heating system. Commissions have the power to instigate an incentive scheme and several are ongoing today. For example, the Michigan plan includes an incentive that rises (lowers) the utility's allowed rate of return depending on the plant availability. Availability is defined as the percentage of hours that a unit would be available for generation. The goal of the plan was to increase the availability of Detroit 252 Edison and Consumer's Power plants. The desired result would be a decrease in fuel and purchased power costs. The counter-productive tendency to be monitored would be an increase in production maintenance costs.51 The plan originally established four availability ranges. Each range is associated with a particular allowed rate of return. A neutral zone between 70 and 80 percent was established. In this range the utility received its cost of capital. If availability fell below 70 percent, the allowed return dropped by 25 basis points. If the availability rate was between 80 and 85 percent, the allowed rate rose by 25 basis points; and if the availability rate was 85.1 or higher, the allowed rate rose by 50 basis points.52 Michigan Public Service Commission change the plan in 1980. It established a separate set of ranges for Detroit Edison and Consumers Power. The number of ranges increased and the size of each range was shortened. The measure of availability was altered. The new measure was set equal to the old measure plus the periodic factor. The periodic factor measures the time the plant is not available due to planned ontages. The new ranges for Detroit Edison are shown on Table 50.53 In the case of Detroit Edison, the scheme worked as planned. For the years 1977-1981, 110 million dollars of fuel and purchased power costs were saved. Profits 253 increased by 32 million dollars, and rates reduced by 78 million. Further production maintenance costs increased only according to their long run trend. One possible explanation for these results is that management, responding to incentive, paid closer attention to costs. Therefore, this example shows that x-inefficiencies can be eliminated by refocusing management attention.54 Summary The feasibility of district heating for an experimental city was studied. The criterion used to judge the feasi- bility of the project was a positive net present value. The project was found to be feasible in 40 of 45 comparisons of the simulation model. The five nonfeasible cases had the following character- istic: Each occurred at the lowest density level (10,000 inhabitants per square mile).55 Other feasibility studies were examined. Only in cases where the study sponsor was a privately-owned public utility were district heating projects found to be not feasible. It suggested that these negative results were a function of current utility experience with district heating, rather than the real potential losses of the project. To overcome utility inertia and trepidation, and to allow the benefits of the project to be reaped, an incentive 254 scheme was proposed. This scheme would allow the benefits to be shared by both the customers and the owners of the electric utility and its heating subsidiary. For it has been found that "unless every major player in this game who has a veto power over the realization of district heating and cooling will at least not lose, it is not going to fly."56 255 Table 31 Pipeline Cost Per Foot of Dual Pipe (1980 Dollars)* Pipe Diameter 1 2 3 4 5 6 7 8 2 60 139 25 43 93 4 97 186 34 75 131 6 133 222 41 109 167 8 172 305 57 143 232 10 216 360 74 181 256 12 263 415 91 203 296 686 696 641 14 108 245 16 346 523 252 374 872 771 20 1031 1126 901 24 1142 1347 946 Sources for Columns: 1. Oliker, "Economic Feasibility of District Heat Supply from Coal—fired Power Plants," p. 1064. 2. Ibid. 3. Pine, "Assessment of Integrated Urban Energy Options," p. 212. 4. City of Piqua, Power Plant Retrofit, pp. 337-338. 5. Detroit Edison Company, Power Plant Retrofit, p. 224. 6. Wisconsin Energy Office, Power Plant Retrofit, p. 4:40. 7. Ibid. 8. Ibid. *All estimates were transformed into 1980 dollars using the Environmental Protection Agency's Sewer Construction Cost Index. 256 OOH.OHe.mN emm.moe.mm omH.mmm.om Hbuoe eee.bem.H eo~.mmm.mH ee~.mmm.~H eonmHEmemae eeH.mmm.mH emm.eme.m mmo.eHo.m coHuanuumHa mmHHe OH OOH.emO.NN emm.mOH.OH omm.Ome.eH Hbuoe eeo.~mm.b ooe.Hee.O on.Hee.O :onmHEmcbpe eeH.~mm.mH emm.e~m.m emm.eHe.m cOHustpumHo mmHHe m eem.mm~.mH emO.~Ho.HH oom.mem.m Hmuoa ooe.eem.H omm.mmm.H omm.mm~.H eOHmmHEmcbaa oeH.Nmm.mH emm.e~m.m omm.eHo.m :oHuanuumHa mHHe H ooo.OH ooo.o~ ooo.om HoHHe .Om Hmm moabumHo mgcmanmn aHv conmHEmcmue mpHmch AmHMHHOQ OOOHV umoo maHHmeHm HmchHHo Nm OHQMB 257 Oo+mHm.h ho+mOm.N Oo+mNm.h oo+mO0.0 Oo+mOh.v Oo+m>O.N ho+mOm.m NH Oo+mOm.N ho+mOm.N Oo+mm0.0 oo+mO0.0 Oo+mmm.w Oo+mNO.H no+mmo.m OH Oo+mNm.mI ho+th.N Oo+mOm.O oo+mO0.0 Oo+mNm.O Oo+mOh.H no+mOO.N mH Oo+mOO.OI ho+mOO.N Oo+mOH.O oo+mO0.0 Oo+MHH.O Oo+mOO.H ho+mHO.N OH no+mOv.HI ho+mMO.H Oo+mOO.m oo+mO0.0 Oo+mNm.m Oo+mNm.H ho+mNO.N mH bo+m~O.NI bo+mOO.H Oo+mNm.m oo+mO0.0 Oo+mmh.m Oo+mHO.H no+mvN.N NH No+mOO.NI no+mmm.H Oo+MHN.m oo+mO0.0 Oo+mmm.m Oo+mOm.H ho+th.N HH ho+mON.mI Oo+mmO.m Oo+mvm.m oo+mO0.0 Oo+mOH.N Oo+mON.H ho+va.H OH no+mnv.mI OmeNm.v Oo+mmh.O Oo+mmm.m Oo+mHO.N Oo+MHH.H ho+mmm.H m ho+mOO.mI Oo+mOH.mI ho+mOm.H ho+mNO.H Oo+mOO.H Oo+mOO.H Oo+mHN.O O ho+mNO.mI Oo+mOO.OI ho+mON.H Oo+mHh.m Oo+mOm.H mo+mOm.m Oo+mOO.h n no+mOH.mI Oo+mOH.OI bo+mOH.H Oo+mON.m Oo+mmN.H mo+mO0.0 Oo+mHm.m O ho+mOO.NI Oo+mmm.OI ho+mOO.H Oo+mH0.0 mo+mOn.m mo+th.O Oo+mmN.m m ho+mNH.NI Oo+mOH.mI Oo+mOH.O Oo+mOm.O mo+mvH.h oo+mO0.0 oo+mO0.0 v ho+mOO.HI Oo+mOO.OI Oo+mOv.O Oo+mmm.h mo+mNO.v oo+mO0.0 oo+mO0.0 m Oo+mOm.>I Oo+mNO.OI Oo+mNO.v Oo+mOO.m mo+mNN.N oo+mO0.0 oo+mO0.0 N Oo+mO>.mI Oo+mOw.mI Oo+mOh.m Oo+mNO.m mo+mOO.H oo+mO0.0 oo+mO0.0 H >mz maHmHsm umou HMUHmmo mocmamu umoo mscm>mm Hmmw Hmuoa IGHMS umam mHHz .em Hmm ooo.om mm OHQME hyHmcmo GOHpmHsaom 258 ho+mOm.h ho+mOm.> no+mOm.H oo+mO0.0 Oo+mO0.0 Oo+mNO.m ho+mmm.O om bo+mN0.0 ho+mOh.O ho+mOO.H oo+mO0.0 Oo+mmm.O Oo+mON.m ho+mON.O ON ho+mO0.0 ho+mON.O ho+mOO.H oo+mO0.0 Oo+mOH.O Oo+mN0.0 ho+th.b ON ho+m>O.m ho+mOh.m ho+mNm.H oo+mO0.0 Oo+mOh.h Oo+m>0.0 bo+mHH.> ON ho+mOO.m ho+mmm.m ho+mON.H oo+Moo.o Oo+mOm.h Oo+mOH.O bo+mOm.O ON Oo+mO0.0 ho+mN0.0 ho+th.H oo+moo.o Oo+mOo.h Oo+mMO.m ho+mOH.O mN bo+mO0.0 wo+mOm.O ho+MHH.H oo+mO0.0 Oo+mOh.O Oo+mmm.m ho+mOO.m ON bo+mOO.m ho+mOH.O ho+mOO.H oo+mO0.0 Oo+mOm.O Oo+mON.m ho+mmN.m ON no+MNO.m ho+mOO.m Oo+mN0.0 oo+mO0.0 Oo+mO0.0 Oo+mmO.m ho+mm0.0 NN bo+mOO.N ho+mOm.m Oo+MHm.O oo+moo.o Oo+mOh.m Oo+mNO.N ho+mO0.0 HN ho+mNO.N ho+mON.m Oo+mOh.O oo+mO0.0 Oo+mHm.m Oo+mHO.N bo+mOH.O ON ho+mOO.H No+mNO.m Oo+mON.O oo+mO0.0 Oo+mmN.m Oo+mNO.N ho+mmO.m OH ho+mOm.H ho+mOn.N Oo+MHO.h oo+mO0.0 Oo+mOO.m Oo+mON.N bo+mOm.m OH >mz msHmst umou HmuHmmU mocmcwu umou m5cm>mm new» Hmuoa IaHmz ummm "OmscHucoo mm mHnt 259 mo+Mm0.0 ho+mOO.N Oo+mbO.h oo+mO0.0 Oo+mNm.m Oo+m>O.N ho+mOm.m NH Oo+mOO.OI ho+mOm.N Oo+mNm.h oo+mO0.0 Oo+mho.m Oo+mNO.H ho+mOO.m OH ho+mOO.HI ho+MHH.N Oo+mOH.h oo+mO0.0 Oo+mO0.0 Oo+mOh.H bo+mmO.N mH ho+mOO.HI ho+mOO.H Oo+mHh.O oo+mO0.0 Oo+mO0.0 Oo+mOO.H ho+mHO.N OH bo+mOH.NI ho+mOb.H Oo+mOm.O oo+mO0.0 Oo+mOm.O Oo+mNm.H ho+mNO.N OH no+mOh.NI no+mOO.H Oo+mOO.m oo+mO0.0 Oo+th.O Oo+MHO.H ho+mON.N NH ho+mHm.mI ho+MHm.H Oo+mOO.m oo+mO0.0 Oo+th.m Oo+mOm.H ho+th.N HH ho+mOO.mI Oo+mHH.O ho+mNO.H Oo+mON.O Oo+mmO4N Oo+mON.H ho+mOO.H OH bo+th.OI Oo+mOO.m Oo+mO0.0 Oo+mOO.m Oo+mmN.N Oo+mHH.H bo+mOm.H O ho+mON.OI Oo+mOm.OI bo+mOO.H ho+mOH.H Oo+th.N Oo+mOO.H Oo+mHN.O O no+mOO.mI Oo+mmN.OI ho+mOm.H ho+mmO.H Oo+mNh.H mo+mOm.O Oo+mOO.h n ho+mmm.mi Oo+mOm.OI ho+mON.H ho+mmO.H Oo+mOO.H mo+mO0.0 Oo+mHm.m O bo+th.NI Oo+mO>.OI ho+mON.H Oo+mO0.0 Oo+mOO.H mo+th.O Oo+mON.m m ho+mNm.NI bo+mHO.HI ho+mHO.H Oo+mnm.O mo+mOO.h oo+mO0.0 oo+mO0.0 O ho+mOO.HI Oo+mOO.OI Oo+mO0.0 Oo+mO0.0 mo+th.m oo+mO0.0 oo+mO0.0 m Oo+mON.OI Oo+mom.OI Oo+mom.O Oo+mON.O mo+mOO.N oo+mO0.0 oo+mO0.0 N Oo+mbH.OI Oo+th.OI Oo+th.O Oo+mm0.0 mo+mHN.H oo+mO0.0 oo+mO0.0 H >mz msHmHsm umou HmuHmmu mommamu umoo mscm>mm Hmm» Hmuoa IaHmz ummm mHHz .Om Hmm ooo.om Om OHQMB OuHmch COHumHsmom 260 ho+mHm.O ho+mON.h ho+MOh.H oo+mO0.0 bo+mOO.H Oo+mNO.m ho+mO0.0 om ho+mO0.0 ho+mO0.0 ho+mOO.H oo+mO0.0 Oo+mOm.O Oo+mON.m ho+mON.O ON ho+mOm.m ho+mOH.O ho+mHm.H oo+mO0.0 Oo+mOH.O Oo+mN0.0 no+th.h ON no+mNH.m ho+mOO.m ho+mNO.H oo+mO0.0 Oo+mb0.0 Oo+th.O ho+MHH.h ON bo+mm0.0 ho+mON.m bo+mOm.H oo+mO0.0 Oo+mON.O Oo+mOH.O ho+mOm.O ON ho+mOH.O ho+mO0.0 ho+mON.H oo+mO0.0 Oo+mOO.> Oo+mmO.m ho+mOH.O ON bo+th.m bo+mO0.0 bo+mOH.H oo+mO0.0 Oo+mOO.h Oo+mmm.m ho+mmO.m ON ho+mOH.m ho+mOH.O no+mNH.H oo+mO0.0 Oo+mOH.> Oo+mON.m ho+mmN.m ON ho+mNO.N ho+mOb.m ho+mOO.H oo+mO0.0 Oo+mOh.O Oo+mOO.m ho+mO0.0 NN mo+mOH.~ me+mmO.m mo+mee.H oo+meo.o oe+mm0.0 Oo+m~m.~ mo+mmO.O Hm ho+mOO.H bo+mHN.m Oo+mO0.0 oo+mO0.0 Oo+mOH.O Oo+mHO.N ho+mOH.O ON bo+mNH.H no+flmO.N Oo+mO0.0 oo+mO0.0 Oo+m>O.m Oo+mNO.N ho+mOO.m OH Oo+m>O.m bo+mNh.N Oo+mO0.0 oo+MO0.0 Oo+mOm.m Oo+mON.N ho+mOm.m OH >mz msHmHsm umou HmuHmmo mocmcmu umou msam>mm Hmmw Hmuoa Ichz ummm "OmseHucoo an mebe 261 me+mHm.~I mo+mem.m me+mme.H eo+mee.e be+mee.m Oo+mme.~ mo+mom.m mH mo+mHm.mI mo+eme.~ me+Mmm.m oe+mee.o Oo+m0H.m me+m~m.H mo+mme.m OH me+mem.MI mo+mem.H bo+mm~.m oo+moo.o eo+mem.e eo+mmm.H me+mmm.m mH mo+mom.eu mo+mem.H ee+mmm.m ee+mee.o ee+mOO.O Oo+mOO.H mo+mHO.~ OH mo+mHm.OI mo+mmm.H eo+mm~.m eo+meo.o ee+mmH.O ee+m~m.H me+m~O.~ mH mo+m~m.mu mo+mmO.H ee+mmm.m oo+meo.o bo+mem.m me+mHO.H me+mem.m NH mo+mmm.mi mo+mmm.H Oe+mHO.m eo+mee.o Oo+mem.m ee+mom.H me+mmo.~ HH mo+mem.OI mo+mm0.0 mo+mmm.H Oe+mem.m bo+mHO.m Oo+me~.H mo+mOO.H OH me+mem.bi me+mem.H me+mHm.H bo+mmm.m ee+mmH.m ee+mHH.H ee+mmm.H m mo+mmm.ei me+mHm.HI me+mmo.m mo+mmm.H ee+mom.~ ee+mmo.H ee+mH~.m m mo+mmm.mi mo+mOH.HI me+mom.H mo+mmm.H Oe+m~O.m mo+mbm.m eo+mem.e m me+mHH.mI mo+mmO.HI mo+mmm.H me+mOO.H Oe+mOm.H me+mem.m Oo+mHm.m O me+mmm.ei me+mmm.HI mo+mmb.H mo+mmm.H oe+mmm.H me+mmH.m eo+mm~.m m mo+m~m.mi me+mmO.HI mo+m~O.H mo+mHm.H ee+mHH.H eo+moo.e eo+moe.o O mo+mmm.mi me+mmm.HI mo+m~m.H mo+bm~.H mo+mO~.m ee+mee.e oe+moe.e m mo+mmH.HI Oe+mom.OI Oo+mem.b eo+mmm.m mo+mmO.m oe+meo.o oo+mee.e m be+mem.mi me+mem.mi ee+mem.m Oo+mme.m me+mom.H oe+mee.e oo+meo.e H >mz msHmHsm umoo HmuHmmu mocmcmu umou macm>mm Hmm» Hbuoa IeHmz ummm mHHz .vm Hmm OOO.OH OuHmch coHuMHsmom mm mebe 262 Oo+mOH.m Oo+mO0.0 Oo+mOH.N oo+mO0.0 Oo+mOO.H Oo+mNO.m Oo+mO0.0 Om Oo+mNO.N Oo+mmN.O Oo+mOO.N oo+mO0.0 Oo+mOm.H Oo+mON.m Oo+mON.O ON Oo+mON.N Oo+mOO.m Oo+mOO.H oo+mO0.0 Oo+mON.H Oo+mN0.0 Oo+mO0.0 ON Oo+mOO.H Oo+mON.m Oo+mNO.H oo+mO0.0 Oo+mHN.H Oo+mO0.0 Oo+MHH.O ON Oo+mHO.H Oo+mO0.0 Oo+MNO.H oo+mO0.0 Oo+mOH.H Oo+mOH.O Oo+mOm.O ON Oo+mH0.0 Oo+MO0.0 Oo+mNO.H oo+mO0.0 Oo+mOH.H Oo+mOO.m Oo+mOH.O mN Oo+mOO.m Oo+mHH.O Oo+mmm.H oo+mO0.0 Oo+mmO.H Oo+mOm.m Oo+mOO.m ON mo+mO0.0 Oo+mOO.m Oo+mOO.H oo+mO0.0 Oo+mO0.0 Oo+mON.m Oo+mON.m mN Oo+mOH.OI Oo+mOO.m Oo+mOm.H oo+mO0.0 Oo+mNm.O Oo+mmO.m Oo+mO0.0 NN Oo+mOO.OI Oo+mOH.m Oo+mON.H oo+mO0.0 Oo+mO0.0 Oo+mNO.N OO+mO0.0 HN Oo+mOm.HI Oo+mOO.N Oo+mNN.H oo+mO0.0 Oo+mO0.0 Oo+mHO.N Oo+mOH.O ON Oo+mOO.HI Oo+mOO.N Oo+mmH.H oo+mO0.0 Oo+mNN.O Oo+mNO.N OO+mmO.m OH Oo+mNm.NI Oo+mOO.N Oo+mOO.H oo+mO0.0 Oo+mM0.0 Oo+mON.N Oo+mOm.m OH >mz maHmusm umou HmuHmmo mocmcmu “moo mscm>mm Hmm» Hmuoe IGHmz ummm "OmsaHuaou mm mHQma 263 Table 36 Net Present Value - Transmission Distant Density Pop. per sq. mile 30,000 20,000 10,000 Transmission Distant 1 mile 1.00E8* 9.09E7 5.87E7 5 miles 7.39E7 6.52E7 3.15E7 10 miles 4.15E7 3.24E7 -2.75E6 *E8=108 Table 37 Net Present Value - Interest Rate Density Pop. per sq. mile 30,000 20,000 10,000 Interest Rate .05 2.31E8 2.18E8 1.65E8 .10 7.39E7 6.52E7 3.15E7 .15 1.80E7 1.17E7 -l.28E7 264 Table 38 Net Present Value - Energy Inflation Rates Density pop. per sq. mile 30,000 20,000 10,000 Energy Inflation Rate .09 1.30E8 1.21E8 8.73E7 .07 » 7.39E7 6.52E7 3.15E7 .05 3.43E7 2.57E7 -7.69E6 Table 39 Net Present Value — Busbar Cost Density pop. per sq. mile 30,000 20,000 10,000 Busbar Cost $/kwh .024 7.39E7 6.52E7 3.15E7 .030 7.09E7 6.22E7 2.86E7 .036 6.80E7 5.93E7 2.57E7 265 Table 40 Net Present Value - Pipeline Cost Density pop. per sq. mile 30,000 20,000 10,000 Cost Factor 1x 7.39E7 6.52E7 3.15E7 1.2x 5.95E7 4.88E7 8.67E6 1.4x 4.51E7 3.29E7 -1.37E7 Table 41 Net Present Value - Nominal Rates Density pop. per sq. mile 30,000 20,000 10,000 Nominal Rates of inflation (construction, energy, interest rate) .05,.07,.1 7.39E7 6.52E7 3.15E7 .07,.09,.l2 7.31E7 6.42E7 2.97E7 .10,.12,.15 7.15E7 6.22E7 2.61E7 266 Table 42 Net Present Value - Attraction Rate Density pop. per sq. mile 30,000 20,000 10,000 Attraction Rate .75 + .008 (I-5) 7.39E7 6.52E7 3.15E7 .70 + .01 (I-5) 6.96E7 6.09E7 2.72E7 .65 + .012 (I-S) 6.53E7 5.66E7 -1.13E7 267 Table 43 City Population Rank and Density: 1975 City Rank Density New York, NY 1 24,964 Manhattan 62,953 Brooklyn 34,257 Bronx 32,900 Queens 18,182 Staten Island 5,655 Houston, Texas 6 2,744 San Antonio, Texas ~ll 2,935 San Francisco, California 16 14,637 New Orleans, Louisiana 21 2,840 Denver, Colorado 26 5,090 Cincinnati, Ohio 31 5,283 Toledo, Ohio 36 4,528 Newark, New Jersey 41 14,450 Baton Rouge, Louisiana 46 6,146 Tampa, Florida 51 3,138 Wichita, Kansas 56 2,800 Richmond, Virginia 61 3,856 Shreveport, Louisiana 71 3,020 Minneapolis, Minnesota 34 6,813 St. Paul, Minnesota 52 5,355 Detroit, Michigan 5 9,675 Piqua, Ohio ** 3,276 **Not Available Source: U.S., Department of Commerce, County and City Data Book: 1977, p. 804. 268 .mHQHmmmm .Emummm GOHudn IHHumHO may now mocmch HmmHOHcaE nqu mHnHmmmm “mocmcHw mHm>HHm anB mmocmpmfiso IMHO OSOHHsuHom Hmecs mHQHmmmm .meHmmmm Hoz .mHnHmmmm uoz oHso .mstm .omHz .uHOHmm \mHHH>mm:MO .omH3 .Omncmmuw .omH3 .comHOmz .non .pHouumo OHSO .OUOHOE OHQO .msva mo OHHU OOHmmO Omumcm mumum chcoomHB eomem uHouuma camHem opmHoa coHumum acme IHmexm OQHHmmcham OOoHocnomB mo musvHumcH MHOHomo aemaa baa Hmzom CHmcoomHz ..muoo w0H>Hmm UHHnsm chaoomHz ..oo OHHuomHm Ode mmw acmemz AmucmuHSOGOU wum>Humv mHHOHw paw .cmfinocHz .nuHEm Hmuamu IHSmaoo wum>HHmv UHOHMHOEEOO mmuoww can puma .m mmEMO .H mpHHHeHmbmm COflUMUOHH pammd mcHHomcomm vamm¢ OGHEHOMHmm Hammd OcHHomcomm can OuHHHQHmmmm cmmzamm QH£OQOHHMHmm was OO OHQMB 269 cumuumm OHSHMHmmEmp OuHo ego» zmz HHH; OmoHocsomB mo muspHumcH mpummssommmmz .OGH IummcHOcm HmmHosz ACOHQMu:HmmmHUV .mHQHmmmm OuHo HmucmEHHmmxm mo ucmfiunmmmo maHm UHmHmw .O .O.z .comumumm .n.z .xubamz .n.z .NuHo ammume .mH .mmdom aoumm Ououmuonmq HMGOHumz .MH .mcmmHHo Bmz cm>mnxoonm mo .OHHMU .mmHmm:¢ moq mmmOOHmEmO HHm3om .O .mm .MHQQHOOMHHsm OHOUMHOQMH can nunozmeummm .anHmmmm .w.z .xHow 3mz HmcoHumz am>msxooum .m .xomeHmm .O .O moammd ucmEmon>wo can AEHHM OCH .M.z .mam3o soummmmm mmumcm IHmmcham anwmzm mo .mHnHmmmm paw cmeoonm mumum xuo» 3wz OOOOHQEOV muwm mHme .O .Ewummm GOHHSQHHHOHU Ahcmmfioo map How moan OGHHmmchcm sme ICHO HmmHOHCSE .ccHz .Hdmm .um Onoumuonmq Imzm wo OOOOHQEOV QHHB wHQHmmwm ImHHommmccHE HchHpmz mOUHm MOO cmmnmz Hmuwm .m OuHHHQHmmmm QOHpmooq pammd ucmm¢ OQHHomcomm OcHEHomnmm "Umchucou OO OHQMB 270 OHOHMHOQMH HaaOHumz mmpHm HMO On woamch xHo3 OHOGHEHH #32 OOOO Imam «mocmummcou UmoH xmmm nuHB Hm3om GMUHHwfid map Hucmqumcoo .mHnHmmmm OuHo HmpcmEHHmmxm um emum>HHmU Hmmmm Ampm>HHmv meHHo .H .O OHHHHQHmmmm GOHHMOOH ucmmfl ucmmd OCHuomcomm OGHEHomem “UODCHOCOU OO OHQMB 271 Table 45 Single Purpose Electric Utility Revenue Hours x Load x Price = Revenue Summer 4380 1232 $.024 kwh $129,508 Winter 4380 800 $.024 kwh $ 84,096 (365x24) = 8760 $213,604 Table 46 Gas Sales Plant Capacity 3080 kw Plant Heat Efficiency 62% Pipeline Distribution Efficiency 95% Hours 4380 Conversion Rate Home Boiler Efficiency Heat Content per MCF Price of Gas 1 kw = 3413 BTU 70% 106 BTU $3.34 per MCF 272 Table 47 Heat Cost Coal Cost = $.01 kwh I. Incremental Heat Cost a) Year 1 2540 kw input in dual purpose operation 2000 kw input while generating electricity 540 kw incremental input Cost = Hours x Input x $/kwh $23,652 = 4380 x 540 x .01 b) Year 2 3080 kw input in dual purpose operation 2000 kw input while generating electricity 1080 kw incremental input Cost = Hours x Input x $/kwh $47,304 = 4380 x 1080 x .01 II. Proportional a) Year 1 Output Electric output + Heat output 1755 = 800 kw 955 kw + Heat cost proportion =175§ = .54 Cost Hours x Input x Proportion x $/kwh $60,076 4380 x 2540 X .54 x .01 273 Table 47 Continued: b) Year 2 Output = Electric output + Heat output 2710 = 800 + 1910 Heat cost proportion = %%%% = .7 Cost = Hours x Input x Proportion x S/kwh $94,432 4380 x 3080 X .7 x .01 274 Table 48 District Heating Profits Profits under the incremental joint allocation scheme a) Year 1 Revenue = Potential gas sales x attraction rate x discount 129,388 x .5 x .9 equals 58,224 Revenue.......58,224 Heat cost.....23,652 Other costs...50,000 Profits ...... —15,428 b) Year 2 Revenue = Potential gas sales x attraction rate x discount II. 129,388 x 1.0 x .9 equals 116,449 Revenue.......116,449 Heat cost..... 47,304 Other costs... 50,000 Profits ....... 19,145 Profits under the proportional joint cost allocation scheme a) Year 1 b) Year 2 Revenue........58,224 Revenue.......ll6,449 Heat costs ..... 60,076 Heat costs.... 94,432 Other costs....50,000 Other costs... 50,000 Profits ..... ..-51,852 Profits ....... -27,983 275 Table 49 Profits of the Electric Utility I. Profits under the incremental joint cost allocation scheme a) Year 1 Economic Revenue Costs Profit Electric Business 213,604 213,604 0 Heat Sale 23,652 23,652 0 Flow of Funds + 5,428 - 15,428 Total 237,256 252,684 - 15,428 b) Year 2 Economic Revenue Costs Profit Electric Business 213,604 213,604 0 Heat Sales 47,304 47,304 0 Flow of Funds 19,145 0 19,145 Total 280,053 260,908 19,145 Net present value of economic profits assuming a 10 percent discount rate ‘ NPV = -15,428 + 17,404 = 1,976 276 Table 50 The Detroit Edison Company Availability Incentive Provision System Availability (ECAR) Equity Return Plus Periodic Factor Incentive 100% - 92.01% + .50% 92.00% - 90.76% + .40% 90.75% - 89.51% + .30% 89.50% - 88.26% + .20% 88.25% - 87.01% + .10% 87.00% - 81.01% - 0 - 81.00% - 80.01% — .05% 80.00% - 79.01% - .10% 79.00% - 78.01% - .15% 78.00% - 77.01% - .20% 77.00% - - .25% Source: Michigan Public Service Commission, Exhibit D: Availability Incentive Clause Filing Requirements, Detroit Edison Case Number U-6006, p. 2. 277 S (Rt/(Cg Pm: _. 0.:ou ‘ nu. Specmmo Hmm-MI. Ian. 13.-amen —. Ham In we: no.» i I '——- -.- -—.« -.“" 3- 4 046vamua Han-I. 1.41?! “Rx \ :‘JINIMUM .5 72' '19.: m" . ' In; Cacnuuc To» 3 . V' ' )' . I mmmmmmo-——+mp L, in 0312:. i I.) . <4: i ' H' K 3 :J . —-—-—-[ 0 . i N ' : METALLIC Cantu? Raw 3~4rn——I-/T: ' ' — g'» '0 . ”Li" JerrJO-Ir '. ‘ . Km. _._—“'3: 1,: Cm. Duo {6‘ , ‘ . ‘fi' DRAIN Pm: -/ Figure 21 Insulated Pipe in Concrete Enclosure V“; Dzflwc: PIP: ‘ '.\"— NA 7:}! QAOOF Cavcnmo \ CA LCIUM ‘ \‘ '\_' 51LICATE INSULAT'OU \ ”_\ [EN GAG/o: CAS!NO Alvuutan AIR SPACE —-‘ - S pecan SuP’ORT ‘—"' SLIP RING Anngsrcfl Figure 22 Insulated Pipe in Steel Conduit 278 558 we: PIP: /- Pounce Caavcne r: Foam /N5ULATION -- 41A 'SPAC: Figure 23 Insulated Pipe in Poured Concrete r-- 5.:R we: PIP: A\ o _ \ .P-Iorzcnvc 31.8w: ( ii ,3 / ,USUMTKW " ‘ 1‘. 7’ ii /NSULAT.'.'0 .9115 A Ly FEA521g_ Cououg; TYPvca-t Ca~srAucrIa~ 34010 0~ 0:7AILS Soprano 6r M. L. SHJLJJN Paasrtc: Co. Figure 24 279 ‘ g .: '4 ”97‘“\LOOSE M /W ._’<‘~. ‘5"' '4: fl ’. . O '¢x§ -.1 41e~9¢ V c ‘e 731.." . 6 r ‘3 b '- 9 .. (’1, s" a , ' “q. f g .5 6". é a 0 . Q 0% ?- ’3' v A O a C ..v g Q I: ... . 0‘. .. '. C ‘. o .. BACKFILLED.-JL_‘.-v ; ,.-' a PACKED . .. Q .0. ... O Q‘. .0 .0 .....,.' o. O C /—\:‘ ' b _ -~—-- ‘9 SUPPLY/ RETURN -* MAINS Figure 25 \‘PIF-‘E BACKFILL EDDWG Insulated Pipe - Construction Completed 280 I. Fossil fuel fired electric plant energy cycles a) Conventional: Heat rejection To cooling water at 100°F Fuel Energy 100 kth V Mechanical Losses Electrical Energy 2kwh < 'P—4 40 kwhe I\ Cooling water Stack Gases 48 kwh 10 kwh b) Waste Heat utilization Heat rejection at 300°F Fuel Energy 100 kth Mechanical “ Losses Electrical Energy 2 kwh < 26 kwhe Useful heat Stack losses 62 kth 10 kwh II. Alternative stylized instanteous input/output choices of the hypothetical electrical utility Input Output 1. 3080 kwt 1. 1232 kwe 2. 3080 kwt 2. 800 kwe 1910 kwt 3. 2540 kwt 3. 800 kwe 955 kwt 4. 2000 kw 4. 800 kw t e Figure 26 Electric Utility Technology REFERENCE CHAPTER IV 1Great Britain, Department of Energy, District Heating Combined with Electricity Generation in tne United Kingdom Energy Paper Number 20, pp. 64-66; Pine, "Assessment of Integrated Urban Energy Options," pp. 90—91. 2 Pine, "Assessment of Integrated Urban Energy Options," p. 74. 3Ibid., pp. 74-75. 4Ibid., p. 101. 51bid. 6Ibid., p. 103. 7Broders, Potential for District Heating: An Historical Overview, pp. 10—11. 8Detroit Edison Company, Power Plants Retrofit, p. 213; Erik Wahlman, "Methods for Development of Modern District Heating Systems" in Combined Heat and Electric Power Genera- tion for U.S. Public Utilities: Problems and Possibilities, ed. Roland Glenn and Richard Tourin (New York: New York State Energy Research and Development Authority, 1977), p. 23. 9Diamant and Kut, District Heating and Cooling for Energy Conservation, p. 219. loIbid., p. 211. 281 282 11Ibid., p. 219. 12Wisconsin Energy Office, District Heating and Cooling Systems for Communities Through Power Plant Retrofit and Distribution Networks, prepared for the U.S. Department of Energy Contract No. EM-78-C-02-4981 (1979), p. 4:38. 131bid., p. 4:41. 14Peter Robinson, "Transmission and Distribution Net- works and the Consumer - The Potential for Development," in Combined Heat and Power, ed. W. Orchard and A. Sherratt (London: George Godwin Limited, 1980), p. 177; Mikkelsen, "Development of District Heating in Denmark," p. 756. 15Mikkelsen, "Development of District Heating in Denmark," p. 762. l6U.S., Department of Energy, Monthly Energy Review, April 1982, p. 88. l7Mikkelsen, "Development of District Heating in Denmark," p. 756. 18 . . . The national average gas price for CommerCial customers is calculated by dividing revenue by sales for commercial customers. For 1980, the price was $3.34 per MCF; See American Gas Association, Gas Facts (Arlington, Virginia: American Gas Association, 1980), p. 75. 19EUS, Incorporated, Dual Energy Use Systems - Dis- trict Heating Survey, pp. 4:11-13. 20Detroit Edison Company, Power Plant Retrofit, p. 206; Price, "Hereford Combined Heat and Power Station," p. 430. 21Hans Thirring, Energy for Man (New York: Harper Colophon Books, 1976), p. 55. 22Ibid., p. 68. 23Pavlenco, G. and Englesson G., Allocation Methods for the Separation of Electrical and Thermal Cogeneration Costs, Oak Ridge National Laboratory Report Number ORNL7TN-6830/P12, p. 3.1. 283 24John E. McConnell, "Maximizing Byproduct Energy," in Combined Heat and Electric Power Generation for U.S. Public Utilities: Problems and Possibilities (New York: New York State Energy Research and Development Authority, 1977), P. 9. 25 p. 87. Pine, "Assessment of Integrated Urban Energy Options," 26 lMCF 1 _ $4.77 $3'34/MCF x 106BTU x 77" 106BTU 27Pine, "Assessment of Integrated Urban Energy Options," p. 43; Craig McDonald, An Analysis of Secondary Impact of District Heating from Existing Power Plants, p. 2. 28Paul Meyer, Monetary Economics and Financial Markets (Homewood, Illinois: Richard D. Irwin, Inc., 1982), p. 115. 29Charles Cathcart, Money, Credity and Economic Activity (Homewood, Illinois: Richard D. Irwin, 1982), p. 115. 30Federal Reserve Bank of St. Louis, Monetary Trends, February 1983, p. 12. 31Detroit Edison Company, Power Plant Retrofit, p. 237. 32U.S., Department of Commerce, Statistical Abstract 1982-83, p. 581. 33U.S. Department of Energy, Office of Reactor Deploy- ment, Update: Nuclear Power Program Information and Data, April/June 1982, p. 1. 34U.S. Department of Energy, Monthly Energy Review, April 1982, p. 88. 35EUS, Incorporated, Dual Energy Use Systems - District Heating Survpy, p. 4.11-13. 36U.S., Department of Commerce, County and City Data Book: 1977, p. 804. 284 37Detroit Edison Company, Power Plant Retrofit, p. 401. 38Ibid., p. 217. 39Ibid., p. 221. 401bid., p. iii. 41Ibid., p. xvi. 42City of Piqua, Power Plant Retrofit, pp. 188-189. 43EUS, Incorporated, Dual Energy Use Systems - District Heating Survey, p. 8.2. 44I. Oliker, "Economic Feasibility of District Heat Supply from Coal-fired Power Plants," American Power Conference 43(1981), p. 1068. 45Peter Donnelly and Isiah Sowell, "Regulation Induced Uncertainties Seriously Impair Investment in District Heating," Factors affecting Power Plant Waste Heat Utilization Workshop (Atlanta: Tennessee Valley Authority, 1978), p. 6. 46Gerald Faulhaber, "Cross-Subsidization: Pricing in Public Enterprises," American Economic Review 65(December 1975), P. 966. 47Ibid., p. 969. 48H. Trebing, "Competition: An Asset to Utility Regulation," Public Utility Seminar of the American Market- ing Association (Chicago, 1967). 49Robert Loube, "Relationships with Other Utilities," in State and Local Regulation of District Heatinggand Cooling Systems: Issues and Options, ed. P. Kier (Argonne, Illinois: Argonne National Laboratory, 1981), p. 33. 50Hill, Power Generation, pp. 86-87. 285 51Daniel Demlow, "Putting Incentives in Public Utility Regulations," Tenth Annual Conference of the Institute of Public Utilities (1978), p. 14. 521bid., pp. 15-16. 53Michigan Public Service Commission, Exhibit D: Availability Incentive Clause Filing Rquirements, Detroit Edison Case Number U-6006, p. 2. S4Mathew McLogan, Commissioner, Michigan Public Service Commission, interviewed 14 December 1982. 55The variable values were: one, transmission dis- tance, 10 miles; two, interest rate,.015; three, energy inflation rate, .05; four, pipeline cost, 1.44; five, attraction rate, .65 + .012(I-5). 56EUS, Incorporated, Dual Energy Use Systems - District Heating Systems, p. 4.7. 57Wisconsin Energy Office, Power Plant Retrofit, p. 4:39. 58Ibid. 591bid., p. 4:43. 6OIbid. 61City of Piqua, Power Plant Retrofit, p. 191. CHAPTER V CONCLUSION Most economists will agree that if it is possible to make at least one person better off without making anyone worse off, social welfare has increased. This study shows first, that such a possibility exists, and second, that the transformation of the possibility into reality mandates a change in the present system of regulation. The source of the potential gain lies in the existence of an x-inefficiency. In particular, the present system of electric and heat supply is more expensive than an alter- native mode. Today, electricity is usually generated at power plants that dissipate two-thirds of the energy input into the atmosphere. Only one-third is transformed into electricity. Natural gas, where it is available, is the cheapest form of heat supply for most individuals. Alternatively, both heat and electricity can be generated from the same plant. The plant is connected to both an electric grid and a heat supply pipeline network. It has been shown, under a set of reasonable conditions, that the second alternative is cheaper than the first. That is, the 286 287 savings in energy costs associated with cogeneration are greater than the expense of constructing a new pipeline network. The net savings, calculated in net present value terms, is the measure of the x-inefficiency. Harvey Leibenstein offered three causes for its exis- tence. "These are: a) contracts for labor are incomplete, or b) the production function is not completely specified or known, or c) not all inputs are marketed, or if marketed, are not available on equal terms to all buyers."1 This study offers a fourth cause based on the theory of bounded rationality. It is that humans will purposefully ignore possible benefits in order to accomplish more limited satisficing goals; and with the passage of time, it will no longer be necessary to ignore the benefits because their existence will be forgotten. The utility companies built electric companies. They could have built full-service energy empires. It is also necessary to explain the existence of the small and declining district heating industry in the United States. One explanation, consistent with the theory of bounded rationality, is that the industry was used as a loss leader. It gathered in customers for the electric utility. Once it had served its purpose, it was ignored and left to. decline. The alternative hypothesis is that at one time district heating was profitable, but now it is not. The cause of 288 this reversal was the completion of the interstate natural gas pipeline network in the early 1950's. With the intro- duction of gas, district heating companies lost customers to gas companies, revenues dropped relative to cost, so profits fell. In order to substantiate this hypothesis, it is necessary to show: first that natural gas and steam are substitutes, second, that the decline in the price of natural gas relative to steam was sufficiently large to induce fuel substitution by a large number of steam customers, and third, that evidence of fuel substitution exists. Two commodities are considered substitutes "if com- pensating variations in income keep the consumer on the same indifference curve, an increase in price of commodity one will induce the consumer to substitute commodity two ..2 for commodity one. Then 3q2 >0 To test the 3P1 u = const hypothesis that steam and gas were substitutes, a series of demand curves was estimated. In only 14 of 244 estimated steam demand curves was the gas coefficient positive and significant. This evidence will not support a claim that steam and gas are substitutes. However, it simultaneously will not support the claim that steam and gas are not substitutes. The estimations, if the demand curves model reality accurately, can only estimate the uncompensated demand curve. Therefore the coefficient reported reflects 289 the response of the quantity of steam demanded to a change in price. The response includes both an income and sub- I stitution effect. If the income effect overwhelms the substitution effect, then substitutes can appear to be complements or independent. Even if the demand curves were estimated incorrectly or if the income effect overwhelmed the substitution effect, implying that steam and gas were substitutes, it still would be necessary to show that substitution of fuel inputs took place. Data on the number of customers who switched from steam to gas are not available prior to 1969. However, implications can be drawn from the available data. It is known that the price of gas was higher than the price of steam in terms of energy units purchased prior to 1970 in the majority of cities studied. Customers incur trans- formation costs when switching from steam to gas. Labor and insurance costs associated with gas heat are higher than those associated with steam heat. Thus, few customers had an incentive to switch. An examination of customer trends reveals the following set of facts. The number of customers served by the eleven major cities peaked in 1954 and has declined steadily since then. The percent of total industry customers served by the eleven major cities rose steadily from 1950 to 1969 and continued to rise during the seventies. For the entire 290 \ industry, new customers added exceeded the number of customers lost to other heat suppliers in every year prior to 1973.4 From this set of facts one can conclude that the decline in customers served was due to the loss of potential customers within the service area rather than a loss of customers to a competitive fuel, and that potential cus- tomers within the service area preferred steam heat to alternative fuels at least until 1973. This combination of evidence discounts the thesis that customer substitution to natural gas explains the low profits of the district heating industry and that, in turn, low profits explain the retarded state of the industry. This line of reasoning exposes another unanswered question. Why did the industry remain within its old boundaries, when its customers moved? Here the availability of natural gas could provide an answer. That is, while natural gas could not penetrate the existing service area, it did provide a ring around the service area. In ful- filling this function, the availability of natural gas led to the downfall of the industry. Again, evidence contradicting the latter hypothesis is available. This evidence supports the notion that lack of interest, rather than lack of profits, led to the demise of the industry. 291 First, the U.S. industry did not avail itself of techniques developed in Europe. These techniques-~hot water distribution, pre-fabricated pipe, and trash burning-- reduce the cost of district heating and allow the costs to remain low when the service area is extended. Second, there is no record of established firms attempting to extend service areas. Executives who were interviewed stated that their companies did not investigate potential extension of service into urban renewal areas during the 1950's and 1960's. Third, the only new system built during the 1960's was built due to the persistence of a real estate developer. The developer insisted that burning gas in a central boiler and distributing heat via steam pipes was cheaper than distrubuting gas to the individual buildings where it would burn in smaller boilers. Finally, a series of feasibility studies provide examples of potentially profitable new service areas. If the industry had pursued these possibilities, it might have fulfilled its potential. Instead the companies did nothing. The Institutional Setting The district heating industry in Europe is viable and growing. Government ownership and/or promotion is often cited as the crucial reason for the European success. The 292 government finances the projects. In so doing, it provides the financial resources that every project needs in the developmental years, and it subsidizes the projects through lower than market interest rates. While this study recognizes district heating projects' need for financial backing, it shows that the backing need not be a subsidy. In Europe, it is recognized that heating via district heating is provided at a cost significantly less than alternatives. European executives believe that their companies would earn significant profits if they were allowed to behave according to private ownership standards.5 District heating projects in the United States also need financial backing during their developmental years. Here, governments are reluctant to finance utilities directly. However, commissions do allow one group of utility customers to finance projects that serve another group. These financing schemes usually occur across time. It could also occur when electric customers finance the development of district heating. The justification for such action is that electric customers will receive reduced rates in the future. Finally, the electric utility reluctance to take on new district heating projects must be addressed. A plan to overcome that reluctance was proposed. The plan would allow the utility to keep part of the net benefits through higher allowed rates of return. 293 Why should a utility receive a bonus for simply per- forming its legitimate tasks? The bonus is needed to crack the shell of self-imposed restrictions. The commis- sion holds up the bait of higher returns to utility execu- tives so they will recognize and invest in profitable projects. 294 Table 51 Changes in Customers Served: All Cities, 1950-1978 Customers Lost to Year New Customers Other Heat Suppliers* 1950 984 81 1951 849 54 1952 676 70 1953 529 47 1954 521 44 1955 510 44 1956 448 59 1957 455 73 1958 355 50 1959 275 71 1960 278 41 1961 258 42 1962 291 56 1963 247 70 1964 231 80 1965 305 56 1966 218 82 1967 224 57 1968 352 19 1969 241 54 (45) 1970 203 46 (38) 1971 133 70 (56) 1972 130 60 (55) 1973 31 58 (42) 1974 102 86 (22) 1975 42 154 (50) 1976 73 128 (30) 1977 85 73 (16) 1978 67 95 (23) Source: "Annual Business Report for 1978," Proceedings of the Seventieth Annual International District Heating Association (Pittsburgh: International District Heating Association, 1979), pp. 1-2. *The numbers in parenthesis are the number of steam cus- tomers that switched to gas. REFERENCE CHAPTER V lHarvey Leibenstein, "Allocative vs. X-Efficiency," American Economic Review 56(June 1966), p. 412. 2Henderson and Quandt, Microeconomic Theory, 3rd ed., p. 31. 3Ibid., p. 32. 4See Table 51, p. 294. 5Larsson, "Development Plans for a Low Temperature Heating System in Odense," p. 128. 295 APPENDIX 296 AOO.HHO AOO0.0 AOO.HO AH.HmHO Am.OOO .HOHHV OO.H mm OO. «OO.Nm OOO. HO.N «O.NOOI OO.OHmI «OOON Hmpmmnoom Am.OmO HOH.. AO0.0 AOOOO AN.OOOO AHOOOO mO. mm Om. 0.0m OOH. OO.H H.OmO O.NOH H.OommI mHHOOMGMHOcH A0.000 Ammm.O AOO.NO A0.0000 HO.NHNO AOOHOO mO. mm Om. «0.0NH Omm.I OON.OI 0.00m OO.mHmI OOOH coumom A0.0NO ANN.O Hmm.v HONO. AOOOO AOOOOO OO. mm OO. t0.0HH OON. «OO.H O.ONHHI O.OmmI H.OOOmI pHoupmo HOm.NO ONO. AO0.0 A0.000 A0.0HO ANONO OO. mm OO. OO.H ONOH. mN.H N.NmI OO.OOI OO.HOI omeoa AONHO AO0.0 AO0.00 AOHOHO AOOOO AOOOONO OO. mm OO. Om.OOHH OOO. OO.NO «HNOO OO.OOONI «ONmOOHI xHow 3mz HOONO HOOH; HONNO 8mm. Hmm: HNmOO HO.H mm OO. OO.OON OOO. HO. O.OOHI «O.OHOI «OOOOI OHsmHmOmHHnm HOON; OOO; HOHHO HmHNO 3.3. 0.000 OO.H ON OO. «HN.HI «OHO. OOO.O H.OHI «O.OHHI N.ONN msva AHO.H. AmOO.V AOO0.0 AOm.OO AN0.00 AmmOO Om. mm mO. tH0.0 «OHO. OOO. «N.OH OH.N «O.OOOI Ouoocou moHp eoHuba bemHo m O m N H o muHo ImHubpm Ipmmbo IHmmmoo m m m mm mm m .3 .Q OO O GOHumH my on oz m m Imuuou OOIOOOH mmumswm ummmH OHMCHUHO "0:0 H0602 Nm OHQMB 297 HO.HHO .OO0.0 A0.00HO AH.ONV AOONHO OO.H mm OO. «N.OOI «ONH. «O.OOmI «O.OOI OO.HNOO OHDQOHHHOO H0.00 AO0.0 Am.mNO A0.0NO HOOOO m.H mm NO. «O0.0H OOOH. ON.Om OO.OONI 0.0HO mHuummm AOO.HO HmH0.0 A0.00H. HO.NOO AOOHO OO.H mm OO. «H.Om ONO. OO.ONOI OO.OHI OO.OONHI Hm>cmo Hm.mv HOm0.0 A0.00HO AN.ONO AOOO. OO.H mm OO. «0.0H OOO. OO.OOOI «O.NOI O.NOH cmusnmuuHm HO0.0 ANH0.0 AO0.000 AN.OmO A0.00HO OO.H ON OO. OOO.mH OOOO. «O.NOmI «0.0NN «O.OOOI ommHa cmm ANO.mO HON0.0 A0.000 AO.mNO HH.HNmO OO.H om HO. «Om.O «OOO. O.ONI «O.HOI OO.OOI coux< AO0.00 AOO0.0 AN.OOO AH.OO. AOOOO OO.H mm OO. «O0.0H HmO. «H.OOOI «H.OOHI 0.0HN couwmo HON.O Hmo.v AH.mo~O HO.NOO HOOOHO OH.H mm OO. Om.m «OOH. «O.OOOI «O.OOHI HmOH mHsoq .um H0.0mO HOOH.O AOOOV A0.00HO ANOHOO OH.N mm ON. Hm.O NmH.I OO.OONI 0.00 OONN OGMHm>wHU “0.0H. HOmH.. A0.0NNO HOOHO HOOOHO OO. mm OO. «0.00 NmH.I 0.000 N.OHNI OOONI mHoEHuHmm OUHH COHum> pcmHo O O N H O OuHU Imemum Iummno IHHHmoo m m mm mm m .3 .o H0 O :oHamH m» on m m ImHHou HemscHucouv Nm meme 298 .mnouum eumecmum UmmeHqu may mum mmmmnucmumm :H mHmQEdz .moameHmcoo mo Hm>wH OOO may um oumu Eoum ucwummmHe OHucmonHGOHOO H0.0. HOo.O HO0.00 HO.m-O AN.OHHO ANONHO mm. om om. OO.HN mmo.I mmm.I m0.0MH OH.OmI mo.mmOI mOHabm Ocmuo AO~.mO HHmo.O Hmm.v HOm.mm. .mO.mHO HO.NmNO OO.H om mm. Om~.I OOOo. «Om. OO.OONI «O0.00 m.OmO muebHum moHu aoHub> uan6 m O m N H o OuHo ImHumum upmmbo IHOHmoo m m m mm mm m .3 .o HO O :oHumH my on oz e m ImHHou HpmseHueooO mm me69 299 Hm.Om. .ON.H. HHmO Am.MHO. HHmm. AmmOH. OO.H mm Om. H.ON Ommm. OH.I H.mOO O.OOHI O.OH mHHoemchOeH Hm.mmc HOOH.. Hmm.O. Ammme Hm.OOHO HOOONO OO.H mm OO. OO.OON OmH. Hm.O H.OOO O.OmHI «NOOOI eoumom Am.O~O AOH.. HOO.O HOOOO lemme HeeOH. OO.H mm OO. Om.OeH OmH. mHO. O.Nmmi m.HeOI N.OHmI pHouume HOH.~O .OHe.. AOO.. HO.HmO He.OHO HOHH. OH.H mm HO. mm. «OOO. OH~.I O.OOI «O.OOI mmH oOmHoe HO.HmO HOoe.O HOH.OO HOOOHO HHOO. AOHHO. mH.H mm em. OOm mm. Om.O~ 0.000 «H.OmmHI OOOONHI Hue» zmz Hm.Om. HmH.. ANO.~O HO.HOO. HO.HOHO HO.OeO. mO.H mm mm. OO.OON me. me.O N.OOI em.~eOI «O.OOOOI mHseHmObHHsm Hmm.. AOoe.O AOm.mO Hm.mH. Hm.mHO 10.mO. OO.H mm Om. OOO.HI OOHe. Oe.O m.HHI «m.OOHI 4m.~O~ mseHm HO0.0 Hmoo.e Hmo.O Hmo.mO Hmm.mO HmH.OHO OO.H mm OO. «mm.m «NHe. OOO. OO.m Om.H OmO.OOI Onoocoo moHH :oHpm> HcmHo O O m m H o OuHo ImHumum Iammno IHOOmoo m m m Om mm m .3 .o O0 O eoHumH me no oz e e ImHHou muma Omauommcmue pusoHOImcmunooo OchD OOIOOOH mmnmswm ummmH OHMGHUHO ammo Hweoz mm OHQMB 300 HOm.mO Hmo.O NO. AO.mNO .O.mmO HOmOO NO.H Nm OO. OO.HN NH. HOH. O.mm 4m.OOHI OOOI mHHummm HOO.HO HNH0.0 Hm0.0 HOOH. Ho.mO. ANOH. mO.H Nm Om. «m.Om OHO. OOo.I em.mNOI H.NO- ONOOI pm>cmo ANO.HO HOOo.O OO.H O.mmH 0.0m HOOOO mO.H Nm OO. OO.HH «OOO. OO.H OO.NONI OO.OHHI omm emusbmuuHm Hom.O HMH0.0 Hm.OO. HO.mmO HomH. HO.H MN mm. OO.OH ONOe. OO.OeOI «O.OmN OOmmI ommHa cam Hmm.H. HOHo.O .OOo.O HmOHO HH.ON. HmOHO Hm.H mN OO. OOO.mI OMOo. Hmo. em.NONI O.OOI Ome couxO HOO.m. HOmo.O HO0.0 HO.mmO Hm.Om. HOOOO MO.H Nm Nm. OO.OH HOo. HO.N 4m.NOOI em.OOHI 0.0mH coummn HO.mHO HOo.O HOMO.. HO.MOm. Ho.OO. HONOO Om.H Nm em. OO.HN «Omm. .4OO.H «N.Hmmu N.HHHI HOOI mHsoH .um HO.NOO HOH.O Hmm.mv HOOOO AmOHO HmmOO. mN.N Nm Hm. O.OHI NNN.I MNO.I 0.00 O.OOI OONO OcmHm>mHo AN.OH. HOo.O HOOH.H. HomHO HH.mmO ONO . OO. Nm mO. 4H.OO OOOH. mHm. HON O.OOHI OmOHI mpoeHuHmm HO.HHO HOo.O HOm.. Hm.HmHO Hm.OOO HmOHH. OO.H Nm Om. OO.ON eOo. «mOO. «H.OOOI «m.NONI mOON “mummzoom OOHH GOHHm> uamHo O O m N H o muHU ImHumpm Iummno IHOOmoo m m m Om mm m .3 .o H0 O coHumH mH no oz m m IOHHOU AOmsaHuaoo. mm mHnma 301 .mHOHHm numeqmgm OmumEHumm on» mum mmmwnpcmumm cH mHmQEdz .moameHmcoo mo Hm>mH OOO may um OHON ECHO pamHOOMHO OHHGOOHMHCOHOO Hm.OO AOHo.O Hmm.N. Hm.OHHO HO.NO. HmmHO HO.H mN OO. 4m.Om OHo. NN.m m.NOI N.Omi Ommmi mOHmmm Osmao ANO.m. .NNe.O AOom.O HO.HO. AH.OHO HOomO NO.H mN OO. mmN. Ommo. tOmm. OO.OONI Om.mO O.Hmm mpcmHuO HO.NHO HOOO.O AOO.. HO.mOH. H0.0NO AOOOHO mO.H Nm mm. 4m.OmI OmHH. OOO.HI .m.ommi «H.NO- OmOHO mpsanHumm OOHu COHpm> ucmHo O O m N H O OuHo ImHubum Ipmmno IHmOmoo m m m Om mm m .3 .m mo O cpomH ON mm oz m m ImHHoo .OmsOHucooO mm mHnme 302 HONO; No; HOO.: HmNN. HO.HOHO HHmmO OH.H mm OO. «OON. OO. HO.HI «O.OOOI OO.HOm OOONI HO>OOO AOOH.. AHOm.O HOH.OV .ONN. AOOHO AONmN. HO. mm OO. OO.I OO.I “OO.OI OO.OmmI 0.00H «ONOmH coumom AOH.O .OH.V A0.00 ANmmv AOmNO AOOOHO OH.N mm ON. OHO.I mH.I OH.N O.OONI 0.00 OOON OOOHO>OHU HOO.V HO0.0 HO0.0 Hm.OmO Hm.OmO ANOOO OO.N mm mO. OOO. OOO. OOO.I «O.NOOI OO.HOHI «OOOO Hmummnoom HO0.0 ANON.O HO0.0 AOOOO HOOmV HOONNV OH.H mm mO. «OH.I OON. «OO.H «HONNI O.ONNI «OmmO uHOHuOO HHH.O HOO.HO HON.OHO AHOOHO AOOHHO AOOOOOO mO.H mm OO. OmmO.I OO.H «O0.0m «NOHmI NOOHI ONOOm HMO» 3mz HO0.0 AmN.O, .Hm.HO AH0.0000 HO0.00HO AHOHO. mH.H mm OO. «OON.I HmN.I «ON.OH OH.NOm OH.NOHI .OmNO OHOOHOOOHHOO HOH0.0 AOO0.0 HO0.0 ANO.N. HOO.mO HNH.OOO HH.H mm OO. OOHH. «OHO. HOO. «OO.OI Hm.OI «ON.OOHI OHOOOOU mOHu OOHum> ucmHo O O m N H O OHHU ImHumpm Iummno IHmmmoo m m m Om mm m .3 .9 MO # GOHHOH mm DO 02 m m IOHHOU OOIOOOH mmnmswm ummmH OHOOHOHO "039 HOOOE Om OHQOB 303 .mHOHHO Onwvamum OOHOEHumO on» OHO mmmmnucmnmm OH mHOQEdz .OOOOOHMOOO mo HO>OH OOO Onu um OHON Eoum HOOHOMOHO OHucmOHNHcOHOO Ame.. AOOm.. HO0.0 HOOHO HNNO AOHOO OO.H mm OO. OO.I «OO.N emo. Om.NOOI 4ON.OmI «OOON ampsbmupam Hmo.o HN0.0 HO0.0 Ammo HmOO Ammm. mm.H mm mm. OOOH. ONH. OOH.HI OO.OOOI Om.mOHI «OHOH commmm HmOo.. HO0.0 HH0.0 HNmHO Hmmm. AOHHHO Hm. mm mm. Omo.u OONm. NO.I «O.OOOI eommHI OOONO mHsoH .HO AmNo.O Hmo.c HMNN.O A0.000 HONHO AOOOO OO.H mm mm. «mOH.I mo. NOH.I 4m.NmmI «O.mHN «HmmO mpoeHuHmm AOmo.O Ame.o HOHN.. .H.OOO Hm.mNO HOOmO OO.H mm OO. OOON.I OOOH. ONOO. «H.OOmI OH.NOHI OONNH musanpumm HOHO.. HOe.O HOm.O HmNO Home HmmOO NH.H mm OO. moo. Omo. 4NN.NI N.mN OO.OONI «mmNN mHHummm moHu eoHum> ucmHo O O m N H o NHHo Imemum Iummno IHmmmoo m m m Om mm m .3 .o O0 O coHumH mm mm oz m m IOHHOU HOmseaucooO OO meme 304 HOHo.O HNo.O lem.. ANHH. O0.000 NO mm. «OH. «Oo. «OO.HI ON.OmOHI Om.OmO Hm>cmo HO0.0 HO0.0 Hm0.00 HmOm. HOmHO NO Om. HH. «ON.H mO.OI O.OHOI OO.HOO coumom OOH.O HOO.. Hmo.OO AOOOO HOHHO NO Om. OH. «OH. Om.NI 0.0HH OH.OHN OcmHm>mHo HOO.. HOO.. HH0.0 HN.mOO .0.000 NO mm. OOO. OOO. «mH.NI «H.OOOI H.NO umummgoom HOO.. HOO.. HO0.0 OOOH. .OOO NO mm. OOO. OON. «OO.NI OO.OOOI N.OO pHouuma HHH.. HO.HO AN0.0. HOOOH. AHOOO NO mm. «OO.I O.N «O.NO OHmNOI OOO- Ono» 3mz HNo.O HOO.. HOO.. HOOO. .NmO NO mm. OOOH.I ON.OI «0.0H «mNHH OO.HHI OHOOHOOOHHOO .No.O HNo.. HHH.. AN.NO H0.0. NO mm. ONH. Heoo. Hooo. ONO.OI OH.OI Ouoocoo OOHOO> ucmHo O O O N H muHo Ipmmno IHOOmoo m m m om Om O0 O coHOmH mm on oz m m IOHHOU Opma OOEHOOOOOHB uuDOHOIOOwHQOOU OOHmD OOIOOOH mmnmsvm ummmH OHOOHOHO mm OHQOB "039 HOOOS 305 .mHOHHO Oumwcmuw OmumEHumO On» OHO mommnuconmm OH muwnfidz .OOOOOHMOOO mo HO>OH OOO man an OHON EOHO OOOHOMMHO OHHOOOHMHGOHOO HOO.. HOO.. ANO.HO HHOHO O0.0NO NO mm. OO.I ONO. «OO.O OH.meOI HOONI OmuenmuuHm Ame.. Ame.c HOO.. OOOH. HOMO NO mm. OOO. «ON. OOe.NI OO.ONOI N.OO coummm ANe.. HOo.O AOO.O OOOHO Hm.NOO NO mm. «mo. OOO. «OO.H- Om.OOOI O.HHHI mHeOH .HO HOe.O OOH.O OO0.0 HOOHO Hm.mOO NO mm. me. OmO. OOH.HI OO.NeOI Om.NNO muoeHuHmm HeH.. Amo.o .NN.. AH.OO. HO.mH. NO mm. OON.- «em. OH.H ON.ONOI O.HOI musanHpmm AHo.. HOe.O HHO.O O0.000 HH.mN. NO mm. OOO. OON. «ON.HI OO.OH OO.OOI mHuummO OOHum> OemHo O O O N H OuHo Iummno IHOOwoo m m m mm mm O6 O eoHpmH Om on oz m m IOHHOU HOmsaHucooO OO mHnma 306 AOoo.v ANo.v OHN.V O0.0HO AmO.omv AHONV moo.l «OH.o «OO.HI Om.O «O.OOHI «0.000H OHuumOm AHNO.V AOHO.V OmH.HV O0.00HO HO.HHHV HO.HONV «ON. moo. Om.l «O.OOmI «N.NOO «OmONI HO>GOQ HHH.V OOH.V HOO.mv HO.Nmm. Hm.OHNV AmNmHv OO.I “OO.I OH.N H.ONNI m.MOI «m.MOOO UGOHO>OHU HO0.0 HOO.. HOO.. O0.00V HO.NOO AHOOV Oo. Ooo.l HO.I «O.OHOI «H.OOHI «N.ONOm Hmummnoom HHH.V AHN.V AOm.Nv O0.0mHV AH.OMHV AOOOHV HH.I OHm.OI OOO.mI «m.OHmI «O.mON «OmONH coumom «AOo.V HON.V «OOO.V AOOOV AmHOV AMOOHV OOH. MHH. OO.N «NOHMI 0.00N «mmOO uHOHuOQ AOo.v OOH.HV Am0.0V AOOOV ANmOV AOmNNNV «OO.I HON. «m.om «HNOmI «OOHNI «HHOmO xHOM 3OZ AOo.V ONH.V HOO.. AO.mMNV Am.mOHv AOONNV OMN.I NH.OI «O.MH OO.OOO O.ONHI «OMOm OHSQHOOOHHSQ ANHO.V Amoo.v Amo.v AOo.NV OmH.OV O0.0NV «NOO. «Ho. «OOH. «NO.OI No.HI OOO.OOI GHOOCOU Om Om Om Nm Hm om OuHo mm GD 02 Gm mm OOIOOOH mmumsvm ummmH OONHHOHOOOO "039 HOOOZ Om OHQOB 307 OHOO. u Emummm map How OHOOOmIm OOHOOHOE OOOOum>HOmOO mo HOQEOZ .mHOHHO Oumvcmpm OOHOEHHOO may OHO mmmmnucmnmm OH mumnadz .OOCOUHHCOU MO HO>QH Mom 03“ “.0 OHUN EOHM UGOHOHMflU \AHHVGMOHWHGOHWO. ANO.O HOO.. AHO.O OO.OmO Hm.ONO HONOO .OO.I «OH. «NO.N .O.NNOI .O.mmi .HmmH OmusnmuuHm HOO.O HNO.O HOO.O NO HHOO AHNNO .NH. «OH. .HH.HI OO.OONI .N.NmHI .HOON coumme ANO.O HOO.. ANO.O AOOO OOONO OONN. .OO.I .NO. HH. .O.OOOI OOOHHI .ONOO mHsoH .uO lmHO.O HOO.O HOH.O Om.HOO AHOO HOOOO OOH.I NO. mO.I .O.OOOI .m.HmN .N.ONNO muoeHuHmm ANO.O HOO.O .mH.O Hm.OOO .O.ONO AOmNO .OH.I .OH. .mO. .m.NOOI .N.OOI .O.ONO musanupmm Om Om Om Nm Hm Om OuHo mm on oz Om mm AOOOOHHOOU. Om OHOOB 308 HOO0.0 AN0.0 OOH.O HN.mHO O0.0HO «mmO. OOmH. «mO.I O.ONI O.ONI mHuummO HH0.0 HH0.0 HO0.0 HH.mOO O0.000 «ONH. “OOO. OO.I OOOOHI «0.000 Hm>cmo HOH.O OHN.O Hm0.00 HOmHO HONHO OH. «OO. OOO.OI O.OOOI «0.000 coumom HHH.O HO0.0 HO0.00 HOOOO OOOH. mO. mOO. OO.I 0.00N OO.OON OcmHm>mHo HOO.. HO0.0 HOO.. O0.000 O0.000 OHO. OOO. OOO.HI OO.HOOI 0.00 Hmpmmnoom HO0.0 ONH.O Hm0.0 HOOOO HOHOO OO.I OOO.H «NO.H OOONOI O.mON uHouumo HO0.0 HOO.HO HO0.00 HOOOO HOOOO OOO.I NO.N OO.OO OONmOI O.NOH OOOH 3mz HH0.0 OmH.. Hm0.0 O0.00NO HH.OOO OOH.I «OO. OOH.OH 0.0mm 0.00 OHOOHOOOHHOO .H0.0 OHOO.. HO0.0 HOO.HO HOO.NO OmO. OOOO. OO. ON.OHI ON.H Ouoocoo Om Om Om Nm Hm OuHo am am oz Om Om mama OOEHOmmOmHB quOHOIOOOHOOOU mchD mhlhvmfl mmhmgmum #mmmq UONHHMHOCOU Om OHQOB ”039 HOOOS 309 NOO0.0 Nm u mcoHum>HOmno mo HOQEOZ Emummm How mumsvmlm OOHAOHOS .mHOHHO Oumwcmum OOHOEHHOO may OHO mommsucmumm OH OHOQEOZ .OOOOOHOOOO mo Hm>OH OOO way an OHON SOHO HOOHOMMHO OHHGOOHOHOOHOO ANO.O HO0.0 ANO.O O0.0mO .O.OHO NOO.I OmN. OON.N .N.ONOI O.ONI OOHOOOHOHO HO0.0 ANO.O HNm.O OOOH. .O.OOO .ON. .NN. N.H- OO.OOOI m.ON eopmmm ANO.O HOO.O ANO.O HO.HOHO OO.OOO OmOO. OOO. OO.OI ON.HOOI O.NOI mHsoH .HO ANO.O Am0.0 HOO.O AOOO O0.000 NO. «OO. OOO.I OO.OOOI .OOO muoerHmm HOO.O AN0.0 OOH.O Hm.OOO Hm.OHO OOH.I .ON. OOm. OO.NNOI «O.ONI musanpumO Om Om Om Nm Hm OuHo am no 02 «m mm .OmaaHpaooO NO mHnme Hmo.v AOHO.V OOH.HV AOONV OOOHO AHOmO OO.H ON Om. OOH. OHO. OO.I OOOOHI O0.00 OO.OOI Hm>cmm OOH.O .m0.0 ANN.OO OOON. HONOO HmmmNO OO. ON NO. OH.I OO.I OO.NHI O.OOOI OOOOH OOHOOH :oumom HOH.O ANN.O HO0.00 HOOOO HOOOO HOOONO OO.N ON HO. OO.I OO.I ON.O OOOHI H.OOHI «OOOO OOOHm>mHo HOH.O ONH.O HO0.0 O0.000 ANONO HOOOHO OO.H ON Om. OO.I NHO. NN.I «O.OHOI H.OONI «OmOO umpmmnoom Hm0.0 HOO.. HOm.O HmHOHO .OHOO HOOOOO OH.H ON OO. «HN.I OO. .NO.H HOOHI H.NO «NONm uHoHuma HHH.O AN.NO O0.0HO HHOOHO HOOHOO HmOomOO o mH.H ON Nm. OHO. mN.N OO.OO «OOHOI OHOO OOOHI xuow 3mz n HOO.O ANN.O ..OO.HO AONOO AOOOO AmOOOO OO.H ON Om. OON.I mO.I OO.OH O.OOHH 0.0NO OHHO OHOOHOOOHHOO . HOH0.0 HOO0.0 HOmH.O HOO.NO HO0.00 HO.HOO OO.H ON Om. «OO. .OOO. «OO.H ONO.O OO.N OO.HOHI Ouoocoo moHp OOHOO> uemHo O O O N H O OHHO IOHHOHO nummno IHOOmoo m m m Om mm m .3 .O m0 O coHumH mm on oz m m IOHHOU OOIOOOH OOHOOOO HOMOH OHOOHOHO "O39 HOOOZ Om OHQOB 311 .muouum cumwcmum wmumfiaumm may mum mmmmsucmumm cm mHmQEdz .mocmoamcoo mo Hm>ma wom may um oumu Scum ucmuwmmwc wapcmoHMAcmwmm Amo.v Amo.v Amm.. AmHH. Amm. Ammm. ON.H ON mO. *Oo.- *NN. «NH.N Nm.Nmmn o.moa «ommfi smusnmupfim Amo.. Avo.v Amm.v Amway Ammav Ahhmv mN.N ON om. OH. Omo. Nm.- mammal *mmm- *OOmN copmma Ammo.. Amo.v Amm.v Ammav Amomv AHmNHV NO.N ON Nm. *mN.- moma. ,mmO.N Om.m «mmmN m.mmm- mflsoq .um Amo.v AOo.V AmN.. .mmv AmONV .NNm. ON.H ON mm. “mN.- Omo. Nm.- *N.mOmu m.mHm *mmmm muoefluamm Amo.v Amo.v Ama.v .mOV .O.moav Ammm. om.N ON Ho. *mN.- *OH. “mo. «H.mmmu *m.Ommu *HOmN musnmfinumm AHo.. Amo.. AOm.v .m.mNV Av.omv ANNv. mN.N ON Hm. Noe. «mo. *mO.N- «H.Nm mm.mamu «mmON mauummm mONu somum> ucmflo m m m N H o muflo -mflpmpm nummno -Nmmmou m m m m m m .3 .a mo * conmN mm an oz «m mm ImHHOU Awmscfiucoo. mm magma 312 Amao.. Amao.v Amm.. AmmN. AONN. mN mm. “ma. «mmo. Nm.- *mmmau m.HMN- um>cmo Amm.v Amm.v .m.m. Amam. AmHOV oN mm. mac. *m.H Hm.mu N.HON- Hmm coumom .mm.uv ANN.V Amm.m. Amflmv Aoaqv 0N mm. mN.o mO.- ON.m- mN.u *m.mmm camam>mau Amo.v Amo.v ANm.. “O.va Amva mm mm. «mm. «ow. «mo.NI «N.vmhl «v.0Hm Hmummnoom .mH.. ANm.. AmN.H. .mNONV Amomv oN mm. «ON.- “o.N «OO.N m.Ommu O.OON “Nouumo .ma.. Aoa.Nv AoO.m. .mONH. Ammmmv 0N mm. «mm.. «mO.m mN.Nm «comm- «ommO xuo» 3mz .mo.v AOm.. Acme. Amomv Ammmv oN mm. «MN.u Nm.- mm.mN *Omma Hmm «Namamcmaflnm AmNo.v .moo.. AmN.. AOm.mav Aom.mv oN mm. *Omo. moo.- «HO.H mm.m- OO.m wuoocou coaum> ucmflo m m m N N muflo nummao -Nmmmoo m m m om mm mo * cONumN mm on oz m m Imuuou mumo Umfiuommcmue wudoHOImcmnnoou mafima NOIOva mmumswm pmmmq >HMCHGHO mm magma “039 Hmvoz .muouum cumucmum UmmeHumm may mum mmmmnuamumm cm mumnesz .mocmwflmcoo mo Hm>ma mom map um oumu Scum #:mumwumw wapcm0flmmcmHmN 313 ANNH.. ANom.v ANN.NV AmHONV Aoomv mN mm. «ON.- Nom.H «o.m m.Ommu 0.00N amusnmpuflm Amo.v Amo.v Amm.v ANNNV Aom. 0N mm. «ONN. NNN. «OO.N- O.Nman *O.mma coumma .mo.. Amo.. Amm.v AomH. .oan mN mm. .mmH.- ON. mm.N «.mm- «mNmN mfisoq .um Amo.. ANH.. Amm.v Amm. ANQNV oN mm. NNo.- HNN. «mOm.u *m.Nmmu «mmaa muoeflpamm AOH.. .qo.v AmN.. AmNN. ANOHV mN mm. mom.u Nam. *mo.a «m.mNmu N.mmu musnmfluumm Aao.v .oo.. Aoo.v AN.Nm. Am.omN. 0N mm. «mm. *mN. mam.au mm.mN m.mman mapummm coflum> ucmflo m m m N N muNo nummno uflmmmoo m m m mm mm mo # cofluma mm on oz m N IGHHOU Aomscfluaoo. mm magma 314 Amoo.v Ahao.v Ammm.v Am.NHV Ao.hmv AHNNV $00.! *mmo. «mm.ml «mm.NN *val «m¢mm magammm Amo.v Avao.v Awm.v Ammmv Am.oaav Amvv. *mH. moo. Hm. {vmval m.NHH vmml Hm>cm0 Amo.v ANN.V Amm.mv Ab.HmHV Am.mvmv Aomomv «vm.l mm.l «v.0HI «m.omml «NBNH «mmmva c0#m0m AHH.V Ama.v Amm.mv Ammmv Ammvv AHHHNV NH.I ¥bm.l av.v *mmmai v.mmml «wmmh ficmHm>mHU Amo.. Amo.v Amv.v Aavv Aamav Avmmv Hoa.l Hooo. mm. «v.moml m.mHNI «mvmm Hmummnoom Amo.v Adm.. AOO.V Aaoaav Amonv Amavmv «mba.l mmH. «mo.N «Nhoml v.mmv «vam uflouuma Amoa.v Adv.Hv Amo.mv Ammaav Awhmmv Awmmomv «Nb.0| mN.H *o.mv «mmmml *mem Nmmma xhow 302 Amo.v AmH.v Amm.v Amomv Amom. Avmmmv «mm.l mmo. «v.vH «mhma «m.mhm mmmm aflnmamcmaflam Amao.v Amoo.v Amwa.. AmH.N. on.o. AH.mmv ¥hmo. «moo. *HH.H «mo.h vm.v *o.vNHI GHOUGOU mm vm mm mm am on wvfiu mm QQ 02 ”um mm NOuova mmumswm ummmq Umufiamnmqmw ow magma "039 Hmcoz mm H mGOHum>Hmmno mo umnfidz no. u EwumMm How mumswmlm unmwwz .muouum Unmccmum cmpmeflumm may mum mmmwnucmumm :H mumnEdz .mocmcflmaoo mo Hm>ma mom map pm oumu Scum wcwummmwc maucmoflmmcmflm« 315 .mNo.. Amo.. AOO.. Amoav AmOV AOOm. Nm.- *mmH. NN.H *Omma mod *vmma smgsnmpuNm Amo.. ANo.. Amm.. AmHNV Aomv Ammmv mm. «mm. NN.- «mHNN- moom- «mNmm coummo .mo.. Amo.. .mm.. Ammav Ammm. ANmoa. «mH.- *mNN. gaO.H N.NH «Nmma mmm- mflsoq .pm ANNo.. .mo.. AmN.. Am.Nm. .mHN. ANNm. .oNH.u mmoa. «mm.u mN.NNm- N.MON “mNOm muoefluamm AmOo.. Aamo.. AmmN.. Am.mmv Am.mmv ANom. *oN.u «NH. «mm. *mmm- «mmm- *mOOH musnmmuumm N mm mm mm Nm Hm om uNu mm 00 02 mm mm Acmscflucou. om magma 316 ANo.o Amo.o Aom.o Ao.NNo Aooao *Nmo. *HNN. Om.- m.Nm- moan mauummm AoNo.o Amoo.o Aoo.o AHoNo AmO. «mma. *NNo. mHN. *oomau m.oHH- um>cmo AoN.o AoN.o AoH.oo AmoNo Amaoo Noo.- «mo.H «HO.O- oom- «mONN geomom 1mN.o ANN.o AOo.oo Aoooo ANmmo mma. Hmo.u mm.N- oHN «moO oomam>mao Aoo.o Aoo.o .mm.. AOoo Aooao «OoN. «mmo. «mN.N- «mNO- «moo “mummnoom ANOo.. AoH.. AmHO.o Amao. AmHNHo *mom.- «om.N ooo.m oomu omm ufiouumo AmoN.o Amm.No Ama.oo Aoooao .momNo mHo.- NO.m o.oo mmmo- oNOO xuo» zmz AoNo.. AHoH.. Ammo... AmoN. ANmNo «OHN.- Nmo. *o.oH oomoa *Nmm mfiomflmomaflnm ANNo.o ANoo.o Amma.o AOm.No AON.oo «moo. «moo. «NN.H «mo.m mo.o- ouoocoo mm om mm Nm Hm muoo Em on oz flm mm mama cmEuommcmue unsouolmcmunoou meow: Nhlnvma mmumsvm #mmmq wmuflamuwcmw "039 vaoz Hm magma 317 mm H mCOHHMKVHmeO MO .HGQEDZ mmmm. n Empmhm Mom mumsvmlm Gwynwfim3 .mHOHHm UHMUGMHm Uwflmgflflmm 03¢ 0H6 mmmwfiflcmhmm CH mhwgz .mocmcflmaoo mo Hm>ma mom man an ouwu Eoum pamHGMMNU haucmofimwcmwm« AoNo.o Aamo.o AmHo.o AoNN. 1N.ooo Omo.- «mmN. *mo.H mooo- mooN amusnmupflm ANmo.. AoNo.o .mNO.o. ANoNo Am.mOo .OHm. «moN. «mm.N- oma- «NmN aopmmo Amo.o Amo.o ANm.. AoNNo Amoao *OmN.- mood. «mo.a O.mo- *OmmN mflsoq .um Aomo.o Aomo.o Amam.o .O.moo ANmNo moo. «mmN. «Hoo.- mooo- «oNo muoefluamm Aomo.o Ammo.o Amma.o Ao.Nmo Ao.Nmo .NHN.- momN. «omo. «mmm- om- musnmonumm mm om mm Nm Nm muoo am no oz om mm Aomscmuaouo No magma BIBLIOGRAPHY BIBLIOGRAPHY Alchian, Armen and Allen, William. Exchange and Production Theory in Use. Belmont, California: Wadsworth Publishing Company, 1959. American Gas Association. Gas Facts, 1980. Arlington, Virginia: American Gas Association, 1980. . Gas Facts, 1981. Arlington, Virginia: American Gas Association, 1981. . Gas Facts, 1954. New York: American Gas Association, 1954. . Gas Facts, 1950. New York: American Gas Association, l950. Anderson, Douglas. Regulatory Politics and Electric Utilities. Boston: Auburn House Publishing Company, 1981. Auckland, James. James Madison University. Interview, May 1980. Balestra, Pietro. The Demand for Natural Gas in the United States. Amsterdam: North-Holland, 1957. Berndt, E. and Watkins, G. "Demand for Natural Gas: Residential and Commercial Markets in Ontario and British Columbia." Canadian Journal of Economics 10 (February 1977). Berndt, E. and Wood, D. "Technology, Prices and the Derived Demand for Energy." The Review of Economics and Statistics 57(August 1975). Blaug, Martin. The Methodology of Economics or How Economists Explain. Cambridge: Cambridge University Press, 1980. Bohi, Douglas. Analyzing Demand Behavior. Baltimore: John Hopkins University Press, 1981. 318 319 Broders, Martin. Potential for District Heating: An Historical Overview. Oak Ridge, Tennessee: Oak Ridge National Laboratory, 1981. Bushnell, Morgan and Orr, Fred. District Heating. New York: Heating and Ventilating Magazine Company, 1915. Cathcart, Charles. Money, Credit and Economic Activity. Homewood, Illinois: Richard D. Irwin, 1982. Christensen, L., Jorgenson, D., and Lau, L. "Transcendental Logarithmic Production Frontiers." American Economic Review 51(February 1973). City of Piqua. District Heating and Cooling for Communities Through Power Plant Retrofit and Distribution Network. Prepared for the U.S. Department of Energy Contract No. EM-78-C-02-4976 (1979). Cochrane, C. and Orcutt,G. "Application of Least-squares Regressions to Relationships Containing Autocorrelated Error Terms." Journal of American Statistical Associa- tion 44(1949). Collins, John F. "The History of District Heating." District Heating, July 1976. "Decadenflxfl' District Heating, January 1945. "Decadenia." District Heating, January 1946. Demlow, Daniel. "Putting Incentives in Public Utility Regulation." Tenth Annual Conference of the Institute of Public Utilities. 1978. Denmark. Danske Elvaerkers Forening. Dansk Elforsyning 1978. . Ministry of Commerce. Act on Heat Supply Act No. 258 of 8th June 1979. . Ministry of Commerce. Danish Energy Policy 1976. . Ministry of Commerce. Danish Energy Report 79. . Ministry of Commerce. First Report from the Heat Plan Committee. DeRenzo, D. J., ed. European Technology for Obtaining Energy Solid Waste. Park Ridge, New Jersey: Noyes Data Corporation, 1978. 320 Detroit Edison Company. District Heating and Cooling Systems for Communities Through Power Plant Retrofit and Distribution Network. Prepared for the U.S. Department of Energy Contract No. EM-78-C-02-4975 (1979). Diamant, R. and Kut, D. District Heating and Cooling for Energy Conservation. London: The Architectural Press, 1981. "District Energy Suppliers in the United States and Canada." District Heating, January 1977. Dixan, R. "District Heating in Eugene, Oregon." District Heating, April 1960. Domino, Francis. Energy From Solid Waste: Recent Develop- ments. Park Ridge: Noyes Data Corporation, 1951. Donnelly, Peter and Sowell, Isiah. "Regulation Induced Uncertainties Impair Investments in District Heating." Factors Affecting Power Plant Waste Heat Utilization Workshop. Atlanta: Tennessee Valley Authority, 1978. Dow, Alex. "Speaking to Engineers." Bulletin of National District Heating Association, July 1939. Edgar, C. L. "A Brief History of District Heating in Boston." Proceedings of the Twenty-second Annual Conference of the National District Heating Association. Pittsburgh: National District Heating Association, 1931. Bus, Incorporated. Dual Energy Use Systems - District Heating Survey. Palo Alto: Electric Power Research Institute, 1980. Faulhaber, Gerald. "Cross-Subsidization: Pricing in Public Enterprises." American Economic Review 65(December 1975). Federal Reserve Bank of St. Louis. Monetary Trends, February 1983. "43rd Annual Meeting Now History." District Heating, July 1952. Friedman, Milton. Essays in Positive Economics. Chicago: University of Chicago Press, 1953. 321 Geiringer, Paul. High Temperature Water Heating. New York: John Wiley and Sons, Inc., 1963. "General Services Administration's District Heating System in the Nation's Capitol." District Heating, Fall 1972. "Generating Stations of the New York Steam Corporation." Bulletin of National District Heating Association, April 1939. Govan, Francis. High Temperature Water for Heating and Light Process Loads: Report No. 37 for the Federal Construction Council. Washington: National Academy of Sciences, 1959. Great Britain. Department of Energy. District Heating Combined with Electricitineneration in the United Kingdom Energy Paper Number 20. . Department of Energy. The Structure of the Electricity Supply Industry in England and Wales. . Secretary of State for Energy. The Combined Heat and Power Group. Combined Heat and Electric Power Generation in the United Kingdom Energy Paper Number 35. . Standing Group on Long-term Co-ordination. Subgroup on Energy Conservation. Experts Group on District Heating. Existing Combined Heat and Power District Heating Systems. Grierson, Relph. "Cost of Service Study in Basic Prepara- tion for a Steam Rate Case." Proceedings of the Forty-fourth Annual Conference of National District Heating Association. Pittsburgh: National District Heating Association, 1953. Halvorsen, Ralph. "Energy Substitution in U.S. manufactur ing." The Review of Economics and Statistics 59 (November 1977). "Have you Heard." District Heating, October 1959. "Have you Heard." District Heating, July 1951. Heating and Ventilating Research Association. District Heating: A Survey of Current Practice in Europe and America. London: National Coal Board, 1967. Henderson, James and Quandt, Richard. Micro-Economic Theory, 3rd ed. New York: McGraw-Hill Book Company, 1980. 322 Hill, Philip. Power Generation. Cambridge: MIT Press, 1977. Hoffman, J. D. "Notes on the Design of Central Station Hot-water-heating Systems." The Heating and Ventilating Magazine 7(1905). Houthakker, H. and Taylor, L. Consumer Demand in the United States. Cambridge: Harvard University Press, 1970. Hughes, Thomas P. "The Electrification of America: The System Builders." Technology and Culture 20(January 1979). International Energy Agency. Energy Policies and Programs of IEA Countries 1979 Review. Paris: Organization for Economic Co-Operations and Development, 1980. Jacobs, Foster. "Princeton University's Central Utility Plant." District Heating, Summer 1972. Jebb, William. "Central Station Heating and Cooling in Hartford." District Heating, October 1961. Jenkins, Normann. "District Heating in the U.K." Energy International, July 1978. . "District Heating Conference Presents Progress and Problems." Energy International, August 1979. Johnson, Tom. My Story. Edited by Elizabeth Hauser. New York: B. W. Huebsch, l9ll. Kahn, Alfred. The Economics of Regulation. 2 Vols. New York: John Wiley and Sons, Inc., 1971. Kahn, Alfred. "Concurring Opinion," in Consolidated Edison Company of New York, Inc. - Steam Rates: Case 26794. New York Public Service Commission. Karkheck, J., Power, J., and Beardsworth, E. "Prospects for District Heating in the United States." Science, March 1977. Kegley, Charles and Kegley, Jacquelyn. Introduction to Logic. Columbus: Charles E. Merrill Publishing Company, 1978. 323 Kirkpatrick, Milton. "Update on Nashville Thermal." Proceedings of the Ninth Biennial Solid Waste Confer- ence of the American Society of Mechanical Engineers. Washington: n.p., 1980. Kmenta, Jan. Elements of Econometrics. New York: Macmillan Company, 1971. Koutsoyiannis, A. Theory of Econometrics. New York: Barnes and Noble, 1973. Kronberg, Thomas. "Organization and Enforcement of Heat Planning." Papers of the Second International Total Energy Congress. San Francisco: Miller Freeman Publications, 1979. Larson, Lennart. "Development Plans for Low-Temperature District Heating in Odense." In Combined Heat and Power. Edited by W. Orchard and A. Sherrett. London: George Godwin Limited, 1980. Larsson, Kjell. District Heating: Swedish Experience of an Energy Efficient Concept. Stockholm: Swedish Board of Trade, 1978. Lehner, Urban. "Many Steam-Heat Companies Bemoan Loss of Customers." Wall Street Journal, 10 January 1973. Leibenstein, Harvey. "Allocative vx. X-efficiency." American Economic Review 56(June 1966). Lipsey, Richard and Steiner, Peter. Economics, 6th ed. New York: Harper and Row, 1981. Lonnrath, Mons, Stern, Peter and Johansson, Thomas. Energy in Transition. Stockholm: Secretariat for Future Studies, l977. Loube, Robert. "Relationships with Other Utilities." In State and Local Regulation of District Heating and Cooling Systems: Issues and Options. Edited by Philip Kier. Argonne, Illinois: Argonne National Laboratory, 1981. Lucas, N. J. D. "CHP and the Fuel Industries." In Combined Heat and Power. Edited by W. Orchard and A. Sherrett. London: George Godwin Limited, 1980. MacAvey, Paul and Pindyck, Robert. "Alternative Regulatory Policies for Dealing with Natural Gas Shortages." Bell Journalof Economics and Management Science 4(Autumn 1973). 324 Martin, H. L. "Sales Promotion in Boston." Proceedingg of the Twenty-second Annual Conference of the National District Heating Association. Pittsburgh: NDHA, 1931. McConnel, John E. "Maximizing Byproduct Energy." In Combined Heat and Power Generation for U.S. Public Utilities: Problems and Possibilities. Edited by Roland Glenn and Richard Tourin. New York: New York State Energy Research and Development Authority, 1977. McDonald, Craig L. et al. An Analysis of Secondary Impacts of District Heating from Existing Power Plants. Richard, Washington: Battelle Pacific Northwest Laboratories, l977. McGraw, Richard L. "Keynote Address." Proceedings of the Sixty-ninth Conference of the International District Heating Association. Pittsburgh: IDHA, 1978. McIntyre, Robert and Thorton, James. "Urban Design and Energy Utilization: A Comparative Analysis of Soviet Practice." Journal of Comparative Economics 2(Decem- ber 1978). McLogan, Mathew. Commissioner, Michigan Public Service Commission. Interview, 14 December 1982. Meyer, Paul. Monetary Economics and Financial Markets. Homewood, Illinois: Richard D. Irwin, Inc., 1982. Michigan Public Service Commission. Exhibit D: Avail- ability Incentive Clause Filing Requirements. Detroit Edison Case Number U-6006. Mikkelsen, Walter L. "Development of District Heating in Denmark." In Total Energy: Proceedings of the First International Total Energy Congress. Edited by Eric Jeffs. Copenhagen, 1972. Mikkelsen, Walter L., and Hammer-Sorenson, Fl. “National or Localized Plans for Combined Heat and Power Develop- ments." Papers of the Second International Total Energy Congress. San Francisco: Miller Freeman Publications, 1979. Mill, John Stuart. Principles of Political Economy. Edited by Sir William Ashley. Furfeld: August Kelly, l976. Miller, Raymond. Kilowatts at Work. Detroit: Wayne State Press, 1957. 325 Morrison, George. "Preparation of Annual Rate Analysis of the New York Steam Corporation." Proceedings of the Forty-first Annual Conference of the National District Heating Association. Pittsburgh: NDHA, 1950. Muir, Neil. "District Heating and Cooling in Sweden." In Combined Heat and Electric Power Generation for U.S. Public Utilities: Problems and Possibilities. Edited by Roland Glenn and Richard Tourin. New York: New York State Energy Research and Development Authority, 1977. . "District Heating: The Water Approach." In Combined Heat and Electric Power Generation for U.S. Public Utilities: Problems and Possibilities. Edited by Roland Glenn and Richard Tourin. New York SERDA, 1977. National District Heating Association. District Heating Handbook, 3rd edition. Pittsburgh: NDHA, 1951. . Proceedings of the Seventeenth Annual Conference. Pittsburgh: NDHA, 1926. . Proceedings of the Twentyrfirst Annual Conference. Pittsburgh: NDHA, 1930. . Proceedings of the Twenty-third Annual Conference. Pittsburgh: NDHA, 1932. . Proceedings of the Twenty-fourth Annual Confer- ence. Pittsburgh: NDHA, 1933. . Proceedings of the Thirty-ninth Annual Conference. Pittsburgh: NDHA, 1948. Needham, Douglas. The Economics of Industrial Structure, Conduct and Performance. New York: St. Martin's Press, 1978. Nerlove, M. "Estimations of the Elasticities of Supply of Selected Agricultural Commodities." Journal of Farm Economics 38(1956). "New District Heating-Cooling Plant in Nashville." District Heating, Fall 1972. New York Public Service Commission. Consolidated Edison Company of New York, Inc. - Steam Rates: Case 26794. 326 . Consolidated Edison Company of New York - Steam Rates: Case 27276. New York Steam Corporation. Fifty Years of New York Steam. New York: New York Steam Corporation, 1932. Neuffer, Hans. "Potential of District Heating in the Federal Republic of Germany." Proceedings of the Sixty-ninth Annual Conference of the International District Heating Association. Pittsburgh: IDHA, 1978. Oliker, I. "Economic Feasibility of District Heat Supply for Coal-fired Power Plants." American Power Confer- ence 43(1981). Orr, Fred. "Report of the Heating Research Committee." Proceedings of the Seventeenth Annual Conference of the National District Heating Association. Pittsburgh: NDHA, 1926. Pavlenco, G. and Englesson, G.. Allocation Methods for Separation of Electrical and Thermal Cogeneration Costs. Oak Ridge National Laboratory Report Number ORNL/TM-6830/P12. Pine, Gerald. "Assessment of Integrated Urban Energy Options." Ph.D. dissertation, Massachusetts Institute of Technology, 1978. Pond, Oscar. Municipal Control of Public Utilities. Columbia University Studies in Social Science 65. New York: AMS Press, 1968. Price, M. E. "Hereford Combined Heat and Power Station." Papers of the Second International Energy Confegence. San Francisco: Miller Freeman Publications, 1979. "Public and Private Sellers of Heat in the United States and Canada." District Heating, April 1962. Public Service Commission of Indiana. Standards of Central Station Hot Water Heating Service: Order No. 4082. "Report of the Operating Statistics Committee." Proceedings of the Twentieth Annual Conference of the National District Heating Association. Pittsburgh: NDHA, 1929. . Proceedings of the Twenty-fifth Annual Conference of the National District Heating Association. Pittsburgh: NDHA, 1934. 327 . Proceedings of the Thirtieth Annual Conference of the National District Heating Association. Pittsburgh: NDHA, 1939. . Proceedings of the Thirty-sixth Annual Confer- ence of the National District Heating Association. Pittsburgh: NDHA, 1945. "Report of the Statistical Committee." Proceedings of the Sixty-fourth Annual Conference of the International District Heating Association. Pittsburgh: IDHA, 1973. . Proceedings of the Sixty-ninth Annual Conference of the International District Heating Association. Pittsburgh: IDHA, 1978. . Proceedings of the Seventieth Annual Conference of the International District Heating Association. Pittsburgh: IDHA, 1979. Reynolds, William C. Energy from Nature to Man. New York: McGraw Hill Book Company, 1974. Riley, Jack. Carolina Power and Light Company: 1908-1958. Raliegh: Carolina Power and Light Company, 1958. Robinson, Peter J. "Transmission and Distribution Networks and the Consumer." In Combined Heat and Power. Edited by W. Orchard and A. Sherratt. London: George Godwin Limited, 1980. Rodousakis, John. U.S. Department of Energy. Interview, 8 January 1983. Root, Thomas. Senior Engineer, Detroit Edison Company. Interview, April 1979. Scherer, F. M. Industrial Market Structure and Economic Performance. 2nd edition. Chicago: Rand McNally College Publishing Company, 1980. Seiter, J. E. "Sales Promotion Developments in Baltimore." Proceedings of the Twenty-second Annual Conference of the National District Heating Conference. Pittsburgh: NDHA, 1931. Simon, Herbert. Administrative Behavior. New York: MacMillan, 1947. . Models of Man. New York: John Wiley and Sons, 1957. 328 . "Theories of Bounded Rationality." In Decision and Organization. Edited by C. Radner and D. Radner. Amsterdam: North-Holland Publishing Company, 1972. . "Rationality as Process and Product of Thought." American Economic Review 68(May 1978). . "Rational Decision Making in Business Organiza- tions." American Economic Review 69(September 1979). Smith, Bruce. Technological Innovation in Electric Power Generation 1950-1970. East Lansing: Michigan State University, 1972. Steag. Annual Report 78. Essen, Germany: n.p., 1979. Stigler, George. "The Economics of Information." Journal of Political Economy 69(June 1961). . "The Xistence of X-efficiency." American Econo- mic Review 66(March 1976). Strickland, Allyn. Government Regulation and Business. Boston: Houghton Mifflin Company, 1970. Stureman, R. V. "Advantages and Disadvantages of Steam and Water Heating." Proceedings of the Twelfth Annual Conference of the National District Heating Association. Pittsburgh: NDHA, 1921. Sweden. The Swedish Institute. Fact Sheets on Sweden. May 1979. Taylor, L. D. "The Demand for Electricity: A Survey." Bell Journal of Economics and Management Science 6 (Spring 75). Thirring, Hans. Energy for Man. New York: Harper Colophon Books, 1976. Trebing, Harry. "Competition: An Asset of Utility Regula- tion?" Public Utility Seminar of the American Market- ing Association, 1967. . "Motivation and Barriers to Superior Performance Under Public Utility Regulation." In Productivity Measurement in Regulated Industries. Edited by Thomas Cowing and Rodney Stevenson. New York: Academic Press, 1981. 329 Ueker, A. B. "University of Michigan's Washington Street Heating Plant." District Heating, July 1973. United Nations. Economic Commission for Europe. Combined April 1982. Production of Electric Power and Heat in Moscow. Seminar on the Combined Production of Electric Power and Heat, November 1978. . Economic Commission for Europe. Contribution of Urban Waste to the Combined Production of Heat and Electricity. Seminar on the Combined Production of Electric Power and Heat, November 1978. . Economic Commission for Europe. District Heating in Sweden. Seminar on the Combined Production of Electric Power and Heat, November 1978. . Economic Commision for Europe. Obstacles to Increased Combined Heat Power Production. Seminar on the Combined Production of Electric Power and Heat, November 1978. Department of Commerce. Bureau of Census. County and City Data Book: 1977. . Department of Commerce. Bureau of Census. Historical Statistics of the United States Colonial Times to 1970. . Department of Commerce. Bureau of Census. Statistical Abstract 1982-83. . Department of Energy. Monthly Energy Review, . Department of Energy. Office of Reactor Deploy- ment. Update: Nuclear Power Program Information and Data, April/June 1982. . Federal Power Commission. The Gas Supplies of Interstate Natural Gas Pipelines. 1973. . Statutes at Large. Vol. 84. "P.L. 9l—GO4." 31 December 1970. United States Conference of Mayors. "District Heating for Treaton, New Jersey." City Currents, November 1981. Urbancik, George. Senior Engineer, Baltimore Gas and Electric Company. Interview, February 1980. 330 Wahlman, Eric. "Methods for the Development of Modern District Heating Systems." In Combined Heat and Electric Power Generation for U.S. Public Utilities: Problems and Possibilities. Edited by Roland Glenn and Richard Tourin. New York: New York SERDA, 1977. Wainwright, Nicholas. History of Philadelphia Electric Company. Philadelphia: Philadelphia Electric Company, 1961. Whitt, William. "Conference Keynote Address." District Heating, July 1977. Wilber, Charles with Harrison, Robert. "The Methodological Basis of Institutional Economics: Pattern Model, Storytelling and Holism." Journal of Economic Issues 12(March 1978). Wisconsin Energy Office. District Heating and Cooling Systems for Communities Through Power Plant Retrofit and Distribution Networks. Prepared for the U.S. Department of Energy Contract No. EM-78-C-02-04981 (1979).