on a single receiving apparatus, we see no basis for the amendment offered by NCTA nor any logical need for a change in Section 110 (5) of the bill to conform it to the provisions of the legislation on a cable television.

APPENDIX A

COMMENTS OF ROBERT W. CRANDALL AND LIONEL L. FRAY

INTRODUCTION

In their statements before the Subcommittee on August 1, 1973, representatives of the cable television industry, and their economic consultant, Bridger Mitchell, argued in part that the growth of the cable television industry would be impaired even if the very low schedule of compulsory copyright fees proposed in

S. 1361 were adopted. In addition, they made other statements regarding the industry's economics.

We believe some commentary from a different perspective would be of benefit to the Subcommittee. Our comments are organized topically as follows.

1. Economic Analyses of Copyright Fee Payments

2. Subscriber Penetration

3. The Effects of New Cable Services upon Estimated Profitability

4. Owners' Expectations of Cable Profitability

5. New Systems and Franchises in the Top 100 Markets

6. Effects of the Definition of a Cable Television System

7. The Effect of Modifying the Proposed Fee Schedule

8. Availability of Evidence for the Determination of Copyright Fees

1. ECONOMIC ANALYSES OF COPYRIGHT FEE PAYMENTS

The analysis submitted to the Subcommittee which relates to the economic effects of the schedule of copyright fee payments proposed in S. 1361 consists of two studies: one by Bridger Mitchell dated September 30, 1972, Cable Television under the 1972 FCC Rules and the Impact of Alternative Copyright Fee Proposals; the other by the authors of these comments, dated April 25, 1973, The Profitability of Cable Television Systems and Effects of Copyright Fee Payments. These two studies arrived at sharply different conclusions.

In his testimony before the Subcommittee on August 1, Mitchell summarized his study and its gloomy conclusion that a cloud of prospective low profitability hangs over the future development of the cable industry, and that even the very modest schedule of copyright fees proposed in S. 1361 in payment for the most essential ingredient in the cable television business-programming—would darken the cloud to the point where the rains would fall and the cable industry would be virtually washed away.

We do not in general take issue with Mitchell's methodology of constructing a "financial model" of a typical cable television system as a basis for reaching public policy conclusions of the sort under consideration by the Subcommittee. Indeed, the model we employed in our analysis is very similar to his. But we differ sharply with Mitchell in specifying the values of several parameters which are inserted into the model and which dictate the low estimates of profitability which he obtains. Two of these parameters-subscriber penetration and revenue growth-are critical in any calculation of profitability, and we believe that more optimistic values of these numbers, based upon projections developed by the cable industry itself, are a far more satisfactory basis for predicting the future profitability of cable investment. Utilizing these data and making a few other relatively minor modifications, we conclude that cable television systems could expect to earn enough profits to insure attracting all the necessary capital required for the industry's growth, even after payment of copyright fees substantially higher than those proposed in S. 1361.

Although our study was made widely available at the time of its publication some months ago, Mitchell did not dispute the evidence which determined these key differences underlying the two studies' respective conclusions, nor did he present new evidence which might affect them. We are therefore not able to under

stand how he continues to adhere to the conclusions of his earlier study which he reiterated before the Subcommittee.1

We might note in passing that the fees proposed in S. 1361 are based upon gross revenues. If, however, the Subcommittee wished to see profitability or ability to pay as a determinant of the amounts of copyright fee payments, the best single indicator we know of is penetration. This variable far outweighs all the other ones commonly used. But the fact remains that the accurate determination of system profitability is a complex function of many variables having to do with system location, availability of off-the-air signals, density, costs, etc. For this reason, a simple formula will almost always produce results which in many cases will be inequitable.

2. SUBSCRIBER PENETRATION

Subscriber penetration is the most important single determinant of cable system profitability. As used by both Mitchell and us, penetration is defined as the ratio of subscribers to the total number of homes passed by the trunk cable; that is, the potential immediately available to a cable system without laying more trunk cable. If penetration is low, only a few of the potential subscribers are contributing revenues to amortize a cable system's relatively large investment in capital equipment. In such a case, the profitability of the system would clearly be low or negative. If penetration is high, however, very substantial revenues will be generated relative to a cable company's investment. In such cases, profitability can be extremely high.

Mitchell's study projects that mature penetration near the edges of the top 100 markets-that is, the level of penetration achieved after many years of operation and promotion of the system services in the local market-will be 34 to 45%. and in the center of such markets only 22% to 35%. These projections are derived from an academic study conducted by his colleague, Dr. Rolla Edward Park of The Rand Corporation. This study is based upon a confidential sample of 63 systems whose identity have not been revealed. Moreover, there has been no verification of the predictions of this study in any publication. Finally, even its author does not utilize its results as literally as Dr. Mitchell in applying to real situations. For example, Park has predicted that Dayton will realize a subscriber penetration rate of 40 percent of homes passed, considerably above the 22 to 35% which Mitchell projects for central cities in the top 100 markets.

While Dr. Mitchell used the predictions of Park's model, we chose to use the industry's own predictions since the latter are based upon informed judgments of the potential appeal of cable services now beginning to develop, but which are totally ignored by Park. The NCTA's own survey to multiple system owners, furnished to the CCO by the NCTA, reveals that these firms expect much stronger demand for their services and they anticipate mature penetration of 65%.

How can these predictions diverge so dramatically with those reproduced by Mitchell from Park's study? The answer is quite simple. Even at present there is a strong movement towards nonbroadcast offerings by cable systems. Some are now offering special channels of recent motion pictures. Other systems are beginning to offer sports channels on a similar basis, and in testimony before this Committee it has been suggested that NFL teams may even begin offering their "blackout" games on a pay-cable basis to local patrons unable or unwilling to purchase a stadium ticket. Clearly, these services and many others which will develop shortly will have the effect of making cable much more attractive to households-even to those able to receive three network signals with clarity. We agree with Mr. Foster of the NCTA that academic studies of cable demand are not likely to provide good predictions because future conditions will be dif

1 The only attempted rebuttal was offered by Mr. Foster (pp. 30-31 of his testimony) in quoting Mr. Kagan's published remarks, whose only discernible point of possible substance related to the "vulnerability" of our report as exemplified in our statement that most systems do not publicly report their financial statements thus making precise cost estimates difficult. Mr. Kagan apparently failed to read the detailed section on operating costs which followed, for not only do we utilize reported cost data, but we present cost data for a sample of systems comprising 781,000 subscribers, more than 12 percent of the industry at that time. Mr. Kagan reports on this effort by noting that "Because of this, the study concluded, it was not possible to work with precise CATV operating cost data."

There is no substantive criticism of our operating cost analysis in Mr. Foster's testimony, perhaps because we choose to present results based upon Dr. Mitchell's cost parameters.

ferent from those today. We suggest as one alternative the survey of his own members.

While we employ the NCTA estimates of subscriber demand in projecting the profitability of cable systems, we also utilize a more conservative assumption for the top 100 markets-55.3% of homes passed, the datum which the NCTA claims is representative of 1972 conditions. As a final calculation, we even utilize a very low 40% subscriber penetration assumption, and we find that copyright fees can be paid by systems in the top 100 markets with this very low subscriber penetration if revenues per subscriber grow by at least 2% per annum. Since few cable systems will be built in anticipation of subscriber penetration of less than 40% of homes passed, this provides a lower limit to anticipated profitability which is in the range of 14.3 to 19.2%. We cannot disagree with Mitchell's pessimistic projections for middle-market systems if they are connected to only 22 to 35% of homes passed. We simply do not believe, however, that profitability estimates should be based upon subscriber penetration levels of systems which even by Mitchell's own assertion will not be built.

3. THE EFFECTS OF NEW CABLE SERVICES UPON ESTIMATED PROFITABILITY

It is now a well-accepted principle of public-utility regulation that all of the capital facilities required for the generation of joint products should not be allocated to only one of these products or services. It would be equally falacious to attribute all of the profits of a very profitable cable system to leased channels and to assign the cost of the capital facilities required to transmit these channels entirely to the retransmission of broadcast signals offered by the cable owner. For this reason, we have calculated the profitability of cable systems under the assumption of modest revenue growth from new services. To do otherwise would understate the ability of these systems to pay mandatory copyright fees.

In his testimony, however, Mitchell admitted that he did not include the revenues or costs from new services such as leased channels or pay cable in his study because the "programming shown on such channels is fully subject to the present copyright law." While we do not doubt that cable owners will be required to compensate copyright owners for this additional programming in proportion to the added net revenues generated, we cannot agree with Mitchell in excluding these revenues from the analysis of overall cable profitability. These leased or pay channels will utilize the same capital equipment as the retransmitted broadcast signals and will contribute to amortizing this investment. Therefore, it is invalid to exclude such ancillary activities from calculations of the rate of return on total investment in cable plant. Mitchell should exclude a pro rata share of the capital base from his more narrow calculation of “profitability" or alternatively, include the net revenues to be expected from these new services. To do otherwise would be tantamount to assigning all of the profits from cable operations to leased channels or to pay-television services while treating the retransmission of broadcast signals of "loss leaders."

While we are unable at this time to provide precise estimates of future revenue growth, we have calculated the profitability of all systems under three different assumptions about the rate of growth in revenues per subscriber:

(a) 0% per year

(b) 2% per year

(e) 4 per year

These assumptions are probably conservative in light of the potential for new cable services and the possibilities for growth in fees for traditiona! basic services. This range of alternatives was chosen because it illustrates the difference between Dr. Mitchell's assumption (0%) and the projections implicit in cable owners' acquisition prices for completed systems in 1971-1972. These prices reflect the anticipation that cash flow will grow at a rate equivalent to 9% in perpetuity-an anticipation which is consistent with growth in revenues per subscriber of more than 3% per year with no increases in costs.

The effect of allowing revenues per subscriber to grow at 4% per year is considerable. On the average, this growth rate leads to an increase of 9 percentage points in the annual rate of return available to system owners. A more modest 2% growth rate increases the estimated rates of return by approximately 5 per centage points.

4. OWNER'S EXPECTATIONS OF CABLE PROFITABILITY

Perhaps the most dramatic evidence that Dr. Mitchell's profitability projections are too low comes once again from the cable television industry. In 19711972, a number of these firms were engaged in an active policy of acquiring existing cable systems at prices which averaged $325 per subscriber at maturity. Since the reproduction costs of these systems is usually less than $200 per subscriber, at least $125 of the $325 purchase price must reflect anticipated profits in excess of the cost of capital.

There appears to be some evidence that the rate of return before taxes required for the construction of cable systems is 15% per annum (although Mitchell states that returns in the range of 9% to 13% are likely to attract equity investors). Mitchell projects rates of return of less than 15% for all but large systems outside the top 100 markets. Even in these markets, he predicts that systems of 10,000 subscribers will realize only 13.6 to 17.0%. But the terms under which the average system was purchased in 1971-72 reflect expectations of at least a 20% annual return on capital before taxes. It is quite clear from this evidence the price knowledgable buyers were willing to pay in the market place-that the industry members do not believe Dr. Mitchell's low estimates of profitability.

5. NEW SYSTEMS AND FRANCHISES IN THE TOP 100 MARKETS

Much of the NCTA testimony centers on Dr. Mitchell's projections for the top 100 markets. It is his prediction that systems in the center of the top 100 markets will earn a rate of return of 2.4 to 10.4% per annum before taxes even without copyright payments. Clearly, if he is correct, there will be no construction of cable systems in the top 100 markets with or without copyright fee payments.

Surprisingly, however, there has been considerable activity in the center of the top 100 markets. Between 1970 and 1972, the number of franchise applications in the central cities of the top 100 markets increased from an average of 1.5 per city to nearly 3.0. In Chicago, 17 applications were pending in 1972. In Milwaukee, 14 were outstanding while in Portland, Maine, 13 were pending. In three other cities, 10 separate franchise applications had been received by municipal authorities. Since the beginning of 1973 another 10 franchise applications have been received by the regulatory authorities in the central cities of the top 100 markets.

Despite the freeze on imported signals in the top 100 markets until early 1972, 22 new systems began construction in the central cities of the top 100 markets between 1970 and 1972. The new cable rules enunciated by the FCC in 1972 should allow other franchise holders to begin construction once the municipal and FCC administrative processes are completed.

This strong evidence of investor optimism in even the center of the top 100 markets is hardly consistent with Dr. Mitchell's projections of low profitability for these systems. Once again, we conclude that the cable television industry does not accept Dr. Mitchell's conclusions and that its members believe that the future is bright for cable investment in even those areas with three or more strong local signals.

6. EFFECTS OF THE DEFINITION OF A CABLE TELEVISION SYSTEM

The definition employed for a cable television system in S. 1361 could have a substantial effect upon the magnitudes of copyright fees paid, and the Subcommittee should be aware of these effects.

Under the informal definition which has been used by the cable television industry since its inception, and employed by the NCTA as well as by statistical services such as the Television Factbook and others, a cable system has been considered as being under single ownership, generally having a single head end, and operating over a contiguous geographic area. The discussions held heretofore between copyright owners, representatives of the cable industry, the Subcommittee staff, and others, have been in terms of this definition. The economic analysis peerformed, including that by Mitchell, has also employed this definition. In his testimony on August 1, however, Mr. Foster proposed an alternative definition as follows. "A 'cable system' is any facility providing a cable service which in whole or in part receives signals transmitted by one or more broadcast stations licensed by the Federal Communications Commission and simultaneously distributes them by wire or cable or radio to subscribing members of the public within a political subdivision within which the facility operates (emphasis

supplied)." It is the latter part of this definition which has a significant effect upon the amounts of copyright fee to be paid because many systems operate over more than one political subdivision. Thus, by the conventional industry definition, there are about 3.000 systems; but by the definition proposed by Mr. Foster, there would be about 6,000 systems-each system being split into two "systems" on the average. If the alternative definition were made operative in the bill, the effect on the amounts of copyright fee paid by the cable industry would be to reduce them by one half.

An example makes clear why this would happen. Suppose a cable system with quarterly revenues of $160,000 operated over two adjoining townships, and obtained $80,000 in revenues from each one. Under the conventional definition, this system would pay $4,000 each quarter, or 2.5% of its total gross revenues. Under the alternative definition, however, this system would be viewed as two "systems," each with revenues of $80,000. Each of the "systems" would pay only $1,200 per quarter and together, they would pay $2,400 or only 1.5% of

revenues.

7. THE EFFECT OF MODIFYING THE PROPOSED FEE SCHEDULE

The copyright fees proposed in S. 1361 vary from 1% to 5% of gross revenues in five discrete steps. The Chairman of the Subcommittee has requested that we analyze the effect upon cable system profitability of fees which are twice as high, i.e., which vary from 2% to 10%.

The average fee paid under the schedule proposed in S. 1361 would be nearly 2% of total cable revenues under current conditions. This average would approximately double to 4% of cable revenues if the fee schedule were increased to 2-10%. The effect upon profitability would be de minimis as the following excerpt from Tables E-3 to E-7 of our study demonstrates:

As may be observed, the effect of doubling the fee schedule is to reduce rates of return by less than 1 percentage point. Average returns are still far in excess of the 10-15% required to attract investor capital.

8. THE AVAILABILITY OF EVIDENCE FOR THE DETERMINATION OF COPYRIGHT FEES In his August 1 statement before the Subcommittee, Mr. Foster argued for the establishment of an initial fee schedule for compulsory copyright license for cable television systems by Congress (p 26). His major reason was that "sufficient empirical data simply does not presently exist to permit arbitrators to fairly establish an initial fee schedule" but would be available in three years when the arbitrators would convene under the proposed bill. We believe that a great deal of economic evidence presently exists or can readily be obtained, sufficient to permit arbitrators to make an informed determination as to just and reasonable fees.

The amount of financial data presently available from the cable television industry includes financial statements of hundreds of cable companies, some of them publicly held. In most cases, these companies maintain financial records by system, samples of which could be obtained under proper circumstances by an