Amazing articles on just about every subject...

Electric Energy, Wholesale and Retail

( Originally Published Early 1900's )

IT IS NOT clear to me that the meeting of an Institute Section is the occasion to discuss a subject which is seven-tenths accounting, two-tenths business expediency and only one-tenth engineering; in which proportion, approximately, those factors enter into the making of public utility rates. I would not, of my own motion, have offered this talk, but I could not graciously decline the invitation extended to me; and if I am to speak on the stated subject, I am more than glad to have an audience made up largely of men having an active interest therein, and entirely of engineers whose habits of mind will permit them to follow my presentation with ready comprehension. I assure you that what I am offering is what your Secretary asked for, and if you don't like it please place the blame on him, in his official capacity.

General Principles Underlying Public Utility Rates

I ask you to listen first to statements of the general principles controlling rate making of any and all public utilities. Some of these principles are self-evident. Each statement has the approval of high authority.

(a) A public utility may charge only a reasonable rate for the service rendered.

(b) The rate charged must cover the cost of the service as near as it can be computed; and a provision for depreciation. It should also cover a reasonable profit on the fair value of the property de-voted to the service.

(c) The costs of service include the outlays directly pertaining to the service immediately under consideration, together with a fairly apportioned share of the general expenses of all the services rendered by the utility.

(d) The provision for depreciation must not only be sufficient to make replacements as they become necessary, but must be such that, at the end of any given term of years, the original investment remains as at the beginning. That is to say, the rate must provide for accruing depreciation.

(e) A reasonable profit is that amount which, under normal conditions, is received by capital similarly invested, and sufficiently high to encourage investors to put their money into the enterprise.

(f) A utility is not required to fix a separate rate for every individual service performed, neither to make such minute distinctions that rate making becomes impossible. On the contrary it is required to make reasonable adjustments and classifications, grouping together services of similar cost and value, and fixing a standard rate for each group or class.

(g) In making such classifications the utility must consider the differences in cost of service, and the differences in value of service. The rule is positive that each class of service must pay its own cost, plus a properly apportioned share of general costs, but the rule is equally positive that the utility need not secure the same percentage of profit on every sort of business.

(h) The utility is forbidden to perform any service at less than cost. It is presumed that deficits arising from performing service at less than cost must become a burden upon other service performed at a profit; which would be unjust. Neither may a utility demand an extortionate price for service. It may not make the necessity of the customer the measure of its charge.

(i) Between the limits that no service may be performed at less than cost, and that in no case shall an extortionate price be charged, the measure of the differing percentage of profit that may be earned in different classes of service is the value of each different class of service to the customers receiving it.

Summing up these principles in a concise, although not complete statement, it may be said that the rate charged for any class of service must pay the cost of performing the service, plus the accruing depreciation of the equipment required for the service, plus some return—more or less—upon the investment. The variation between class rates will, therefore, arise, firstly, from difference in class costs, including depreciation and, secondly, from difference in the return upon the investment. Pure theory would suggest that the percentage of profit collectible from each and every class of service should be the same. Commercial experience and court decisions based upon observed facts recognize that this is impossible—that it is a counsel of perfection. There are plenty of illustrations—every corner grocery store, every department store, every mail order house, every factory making more than one line, knows that margins of profit cannot be made alike and that a surplus on one line goes to make up a deficit on another.

Even State and Federal Governments in assessing taxes, fail to lay exactly equal burdens. The just and the unjust, the thrifty and the spendthrift, the worker and the slacker, all alike complain with truth that the tax burden is unequally distributed—that neither the cost of the service which they receive from the State or the Nation, nor the value of the service to the recipient, namely, the taxpayer, controls the laying on of the tax burden. Be that as it may, the law requires that a public utility shall collect, from each reasonably constituted class, all the cost of serving that class; and that it shall not collect an exorbitant profit from any class. The much defamed rule of charging what the traffic will stand, when applied honestly and intelligently, comes very near to giving a perfect apportionment of the reasonable return. It applies, for instance, to the collection of a full fare from the short distance passenger on a street car, who may be saving himself steps; and equally to the collection of no greater fare from the workman who travels 10 miles, night and morning, between the cottage which he has bought (or is buying) and the bench at which he does his share of the world's work. Interpreted rightly—either according to the Sermon on the Mount or according to the Proverbs of Solomon—the rule of charging what the traffic will stand means that the person served shall pay according to his profit by the service; that his gain shall be the measure of the price; that if he profits much he shall pay liberally, but if he receives only an absolute necessity he shall pay little. It is written in the Sermon that With what measure you mete it shall be measured back to you—Matt. VII, 2. And in the Proverbs that There is that scattereth and yet increaseth, and there is that withholdeth more than is meet but it tendeth to poverty —Prov. XI, 24. If the rule is or has been unwisely or unfairly employed, the error is in the application, and not in the rule.

Why Rates Differ

So much for the principles; let us proceed to practice. Your Secretary suggested that you, many of you being responsible for the electric supply in hundreds of horse power to large factories, would like to know wherein and why that service differed in cost and in price from electric supply at the other extreme, say, to a few lamps in a private residence. Comparison might be made between any two of the classes separately scheduled in the rate book, but the suggested comparison is one which will interest more people and will be the most complete illustration of the facts and the con-trolling principles. Each of the two classes compared is of great importance. One consists today of 88 large industrial establishments employing many thousands of hands. The other class consists of over 87,000 residences—each of them the home of a Detroit family. I speak of the Greater Detroit—the city area and the zone of villages immediately surrounding it. I do not include any figures from the two-score municipalities north of the Eight Mile Road and west of the Greenfield line which are served by the same electric system.

Will you please set these facts in your mind. The number of residences served in the Detroit area is almost one thousand times the number of factories which buy untransformed electric current from us. The number of accounts to be rendered monthly for the one group is less than 90, and for the other group nearly 90,000. The earnings from the factory group for the current calendar year will be, in round numbers, $1,200,000 and from the residence group $1,500,000. The purchases of current by the 90 customers will be but Q0 per cent less than the purchases of the 90,000 customers. So that the contrast between wholesale and retail business could not have a more striking illustration.

Please do not attempt to average these figures. Not all of the 90, neither all of the 90,000 customers, have been connected to the lines for the entire year. The average annual bill will, there-fore, be somewhat higher than earnings divided by customers. I will give you later some figures based on the accounts of 1914, using the loaded average number, for that year, of customers in each class. The final 1915 figures will not differ enough to affect any general conclusions which may be drawn.

Systems of Distribution

Each of these rate classes is supplied with electric current generated in the same power houses by the same turbo-alternators. The current for both classes leaves the power house through the same meters and over the same cables, and is transmitted to the same substations. The divergence in the method and character of the service begins at the substation. All of the factories under consideration purchase untransformed alternating current, 60-cycle, three-phase, at approximately 4400 volts. They buy it in blocks from 100 kilowatts upwards. Some take over 1000 kilowatts. The current delivered to them is maintained at the correct frequency with great exactness but the voltage variation plus or minus is considerable. To be precise, the plus variation of voltage is considerable. Minus variation means inferior service, whereas plus variation is tolerated or is even acceptable. Each factory has its own equipment of transformers. Some factories have converting apparatus to change the alternating current to direct current. All of them have, of course, switch gear and such regulating devices as are required by their operating conditions. Every one of them has a competent electrician; most of them a competent electrical engineer who is responsible for the distribution, regulation and utilization of the current within the premises. All of the transforming, converting and regulating equipment is owned by the customer. All of the cost of transformation, conversion, regulation and supervision is borne by him. The electric company's investment and responsibility ends with the delivery of the high tension alternating current at the customer's terminals.

The line from the substation to the large customer is invariably of the simplest kind. Usually but not always there is a three-phase regulator on the circuit in the substation. There is always an oil switch with overload release. There is in most cases a short length of underground cable over which the current leaves the substation, but the greater part of the distributing circuit is overhead and of the simplest possible construction. The electric company's investment for the service of these customers is, therefore, a pro rata part of the cost of the generating station, and of the trunk line cables and the substation space; and the special investment in substation switch gear and in the 4400-volt distributing circuit.

At the same substations the supply to residence customers has to be stepped down to half voltage—or (it may be) converted to direct current. The circuits serving residences wander in and out through tree-lined streets and wire-infested alleys, and costly experience has shown that in a city now and always noted for its shade trees, and from time to time afflicted with competing telephone companies—not to speak about private wires—the use of 4400-volt, three-phase, distribution in residence districts means unreliable service and excessive danger to life and property. Of course underground wires would have advantage under such conditions, but the public will not pay for them. Please consider the risks in other ways that the American public tolerates or even welcomes—the risks of traffic by rail or motor, the fire risk which to a European visitor seems criminal—and my saying that the public will not pay for underground wires will be recognized as an exact statement of the case. A quick calculation indicates that to supply electricity to our residential districts by underground mains would somewhat more than double the necessary selling price per kilowatt-hour—that is to say, the minimum return on the new expenditures in cables and conduits, plus depreciation, plus taxes, would be equal to the entire present earnings from that class of service. The result, of course, would be that many people would go without electric light. Query, would the advantage be worth the price?

Please do not slight that reference to taxes. The tax burden on the local company this year is over $1,300 for every day of the year including Sundays and holidays. And a tax plan which assesses property values presses very heavily on a public utility whose in-vestment is at least five times its gross earnings.

To return to our muttons: The current received at the sub-stations and destined to be used in residences is transformed to 2300 volts, single-phase, or converted to 250 and 125 volts direct current. The cost of apparatus for this first transformation or con-version, and the operating losses in this apparatus, are, of course costs assignable to residence service alone. The distribution at the lower voltage over many-branched circuits calls for more special investment and for further losses of energy. In brief, with the solitary exception of circuits for street lighting, the circuits distributing current to residences are the most costly in proportion to their capacity.

The direct current distribution to residences is a comparatively small item. It survives from the earlier Edison days, and has such merit that it is inadvisable to do away with it. Of course the drop on a low tension current network at time of peak load is large, but the total distribution loss in the year compares favorably with that of a house to house alternating current service.

The 2300-volt alternating current circuits deliver current to scattered transformers. So far as possible these supply, and are linked together, by secondary networks; but in new residence subdivisions this is impossible. The iron or hysteresis losses of these service transformers are, in the aggregate, a large item in the distribution expense, Please remember that there are 90,000 residence customers to be served, and please consider the area over which these customers are scattered, and do your own imagining as to the aggregate of the iron losses. These iron losses are a large part of the distribution costs of the electric company, whereas the iron losses of the transformers in the large factories are not only a smaller percentage of the energy but are on the customer's side of the meter.

Secondaries and service connections from transformers to customers are a considerable cost item, both for investment and maintenance. And 90,000 meters make a very large item indeed. The ordinary residential service meter is small, and is low priced in comparison with the primary current meter with instrument transformers which is required for the factory customer, but the cost of 90 primary meter installations in factories is not a circumstance to the cost of 90,000 meters installed in 90,000 different dwellings.

Customers' Accounts

Now for exact figures based upon the closed accounts of the year 1914, and the average number of customers in each class during that year. The annual bill of the average residence customer for 1914 was $19.50; that is, the bill for the entire year. This figure will be lower in 1915 because of relatively greater number of cottages and small apartments connected. The annual bill for the average customer buying unconverted current in 1914 was $15,185.

The average cost of keeping a customer's account open in 1914 was $6.60. That is to say, the cost of testing, reading, calibrating and repairing meters, making, recording and collecting customers' bills, answering inquiries about service and bills, and a lot of similar fussing; depreciation and taxes on meters; and, finally uncollectible bills—all of which expenses increase very closely in proportion to the increasing number of customers—amounted to $6.60 per customer. In the making of all these expenses the residence customer is an average customer. He asks more questions. He blows more fuses. He requires more duplicate bills. He makes, proportionately to business done, more work for inspectors and repairmen than any other class of customer. Also, his proportion of uncollectible bills is high.

I don't mean to say that the public in its home life is less honest than in its commercial life. But the residence customer's business methods are not so good and if he defaults he seldom has any tangible assets, even if the loss in any one case were large enough to warrant a lawsuit. The large factories in Detroit practically never default. They pay their bills promptly. They know what their bills ought to be, and they check their meter readings from week to week or from day to day exactly as they check the weight or the count of raw material delivered in their receiving sheds. If their electric bills vary from month to month they do not descend upon you with a protest that this cannot possibly be so, because the family habits are as unchangeable as the laws of the Medes and the Persians. The factory superintendent, or the electrical engineer, knows exactly what department ran overtime, or what additions were made during the month to the motor equipment. Those 90 factory customers make less trouble to the dollar of investment, or to the dollar of earnings, or to the number of bills rendered, than any other group of customers to whom we sell. It would not be fair to say that $6.60 per customer is the whole cost of keeping open a factory account. But please notice that $6.60 is 34 per cent of the total annual payment by the average residence customer and also notice that an allowance of $100 per annum, say two-thirds of one per cent, would be ample to take care of the metering, accounting, etc., for the factory. You will, I think, agree with the dictum accepted by all engineers who have studied customer costs, that these costs, at two-thirds of one per cent, are a negligible item in the supply of factory power, whereas at 34 per cent they are a controlling cost in the distribution of current to dwellings:

Investment and Operating Costs

We can now make a useful comparison between the investment required by wholesale business as exemplified by the factory customer and retail business as exemplified by the residence customer. Each of you who keeps production or selling costs knows that the quick method of comparing one line of business with another is to construct two comparable sheets for the total output or total sales of the factory during a given period; one cost sheet on the assumption that the entire output or total sales had been of the one class of goods and the other on the assumption that the entire output or sales had been of the other class. Here are some salient figures arrived at by a similar comparison. The total earnings from electrical service during 1914 were $5,461,966. The number of customers (loaded average) was 86,865. Had the same amount of business been done with primary current customers alone, the number of customers would have been 359; or had all the customers been residence customers, their number would have been 280,103.

Had the system been built to serve only 359 primary current customers, it being premised that the power plants, trunk lines and substations remained the same, the following constructions would have been unnecessary.

Substation equipment $1,032,670.00
Storage batteries 260,400.00
Conduits, underground 1,067,110.00
Poles and fixtures, overhead 832,610.00
Conductors, underground 906,430.00
Conductors, overhead 1,108,940.00
Line transformers 472,490.00
Lamps, arc and incandescent 94,580.00
Meters, (reduction) 1,012,220.00
Office, storeroom, meter
and department miscellaneous 600,000.00
equipment; reduction


Per contra, we would have had to add to the substation equipment and to overhead lines serving primary customers at a cost of about $1,500,000; which means that the fixed charges and depreciation and taxes and maintenance and repairs upon $6,000,000 worth of property would have been saved in 1914, had all our customers been of the factory power class. Conversely, it means that had our year's business been done with residential customers alone we would have been required to provide additional equipment costing not less than $11,000,000. I am omitting the awful details. Neither I nor any other electric light man wishes to think of what would happen if we had no connected business except residence lighting. When one of us has been very naughty indeed, and a nightmare is the punishment which fits the crime, that is the dream which is visited upon him.

I brief the comparison thus—the difference between $6,000,000 minus and $11,000,000 plus, being $17,000,000, is the increase of investment, outside of power houses, required to serve retail instead of wholesale customers. And lest you fancy that I am talking millions in a purely academic way, let me say that I personally have approved estimates and signed vouchers for something over thirty millions of real money spent by my Company for additions or betterments.

Turning from the investment side to the operating costs, the conditions of our comparison require the use of identical costs for production of the current, so that load factor variations are excluded from consideration. An item of $11,672 storage battery costs would vanish. Distribution operating expenses, $74,541, would also vanish. Customers premises expense, inspection, fuse renewal, etc., amounting to $66,324 would be no longer required. Neither an item of $141,124 renewal of incandescent lamps. And general office expenses would be cut by something like $850,000. The sum of these operating expenses is $1,143,661.

In giving you these figures from our 1914 cost sheets I have touched only the high places in the capital and the expense accounts. I have not attempted to carry the comparison down through the hundreds of sub-accounts to which a light and power company is required to distribute its expenditures. I have said enough, however, to show you that the cost of service to equivalent groups of wholesale customers and of retail customers differs so largely as to warrant a very great difference in the price of service without at all taking into consideration the other occasion for difference, namely, the value of the service to the customer. Or, in other words, the mere difference in out-of-pocket costs, before considering the rate of return upon the investment, is such as to warrant a very large difference in unit price.

Earnings and Rates

As to the return upon the investment, year by year for many years past there has been prepared by me, or by my assistants for me, a careful distribution of costs, item by item, of each of those hundreds of capital and operating items to each of the several classes of service for which we make and file rates. Just to prevent someone thinking something mean, I say here that we file all our rates. This annual analysis conforms to the dictum of the Supreme Court of the United States that the outlays exclusively pertaining to a given class must be assigned to that class and the other expenses must be fairly apportioned. Against the class costs, so found, we set up the class earnings. Of course the first step of such an analysis is to allocate to each class its share of the property. Deducting out-of-pocket expense for the class, from its earnings, leaves a remainder which ought to be a reasonable return upon the value of the property—that is to say, upon the present value of the investment. I pointed out to you at the beginning of this talk that the principles—and, for that matter, the court decisions—are explicit, that we are not required to earn the same rate of return from each class of service. But naturally such studies, continued for many years, tend to that end. If a class of service shows a large margin of profit, that class puts itself in line for a reduction of rate. On the other hand, if a class of service shows no profit, or a loss, the rate is likely to remain stationary until the all around reduction of unit costs, following the increase of annual production and sales, puts that class of service over on the right side of the books. Be that as it may, we know what each class earns; and it goes without saying that the sum of the net earnings of all-classes must be enough to pay the annual return upon all the capital in the business. In a healthy concern, and in a normal year, it must be enough to make a small addition to the surplus, over and above the amount distributed as interest and dividends. The average rate of return upon the total capital and the return by classes, are easily calculated and it will then appear whether class return, set against the investment required by each class, exceeds, or equals, or falls short of the average return. In 1914, residence service fell short, and sales of primary current earned a little more than the average. Please fix that in your mind. The net rate paid by the residence customer in 1914 was 5.87c per kilowatt-hour. Nevertheless, for the causes set before you in detail and for another cause which I have touched upon but have not gone into, namely, the poor load factor of residence lighting, that class of service failed to earn the average return upon the value of the property employed in it. Whereas, sales of primary current to large power users, at an average price per kilowatt-hour of 1.12c, paid somewhat above the average rate of return upon the value of the property therein employed.

Lest you should imagine vain things, let me say that the 24-hour a day service of certain primary current customers and the very long shifts worked by others, tended to make the load factor of that class exceptionally good during 1914—a condition which, much as we may desire it, is not likely to be perpetual. Also, coal was cheap in 1914, and the cost of fuel is an appreciable item in long-hour service. Coal is going to cost more hereafter. On the other hand, the present bad load factor of residence service is being bettered by the use of domestic appliances of which the homely flatiron is the most important. Do you realize that the flatirons which we know to be in use in this town are the equivalent of an installation of 55,000 horse power in motors? If on some Tuesday morning every one of these flatirons were put into commission at once we would surely notice it at the power houses—we do notice it now in the substations in the residential district. Furthermore, although there are entirely too many houses still being built away out in the fields and the marshes, there is much filling up of the nearer residential districts of the City so that the costs of mains per kilowatt of residence demand may some day be reduced from the present absurd figure to one more reasonable.

In Conclusion

I do not fancy that either the large factory or the small residence —the wholesale customer or the retail—is complaining today either of our service or of our price. It is not an accident that residence service is sold at a very small margin over cash cost, with the result that more homes are electric lighted in Detroit than in any other city of its class. It is part of a long established policy adopted after consultation with the official as well as the unofficial leaders of the community. Neither is it an accident that in Detroit a greater proportion of the large factories use central station electricity than in any similar city. They use it because it pays—and they are very sure that it pays, these cost experts of Detroit workshops. And as the moral of this story is in the application of it, I close by saying that we want to serve well our community both in its shops and in its homes; that we have had fair play from Detroit and we have played fair; and that we believe and intend that the mutual good will of the past and the present will continue into the far future.

Home | More Articles | Email: