It seems to me that the necessity for copyright protection from the point of view of the publisher arises out of the very significant difference in costs that attend original publication, and those who may imitate the original publication once it has come into existence. There are two sources of this cost. One is that an original publisher incurs rather substantial risks when he publishes a book. It is virtually impossible to predict just what the commercial value of a manuscript is going to be when it is translated into a book. Some books sell well and some just do not sell very well.

I could add parenthetically that I specialize in writing books that do not sell very well.

For this reason there is a risk that an initial publisher bears which an imitator does not bear. For example, if there was no copyright at all, an imitator could just pick out the best sellers and manufacture them, and he therefore could avoid the risk that comes in the initial stage of publishing some that just simply do not turn out well.

So for the publication of original works, it seems to me that the copyright is essential to protect the original publisher.

There are also certain prepublication costs that an original publisher incurs that imitators do not incur. There are costs of galley proof, of page proof, of negotiations with the author-indeed, one of the statements that is submitted to this committee, I believe, contains a quote that a single high school book will require an investment of $50,000 before the first copies are available for sale. Well, this is a cost that the original publisher incurs but one that an imitator does

not.

Now, it follows from this that the greater the amount of the difference in the costs between the original publisher and anyone who elects to copy the works of an original publisher, the greater the incentive to try to circumvent the copyright. Or stated another way, the lower the cost of copying an original work, the greater is the incentive to circumvent the copyright.

It is fairly clear from these studies that I helped coordinate for the two book publishing organizations that the incentive to circumvent copyright has grown much stronger due to certain technological developments in the copy field. A study that was carried out by Arthur D. Little, Inc., shows that over the past two decades, the cost of copying a page has declined dramatically as copying technology has advanced over the past couple of decades. Some of the studies carried out by the National Opinion Research Center in the University of Chicago showed that in our elementary and secondary schools, indeed in colleges, copying machines are used with considerable frequency in copying such material as textbooks, test and answer sheets, and music scores, although copying is usually limited in these cases to several pages.

The recent and projected advances, therefore, in copying technology, I would argue, pose immediate problems for public policy toward the copyright law, the subject that this committee is inquiring into, but for the more distant future, I would submit that larger problems would be presented by the application of computers and microimages and their technological descendants to the process of information storage, retrieval, and dissemination.

The present state of technology suggests that the computer will affect conventional publishing in too distinct ways: The initial ver

sions of some types of information that are now reduced to writing; copyrighted and published, will very likely be computerized, thus bypassing conventional publishing althogether; and secondly, the contents of published books will be stored in computers and, once stored, will serve as a substitute for additional printed copies.

Now, the reduction of literary and scientific creations to computer tape for dissemination by experts carries with it the risk of dulling the incentive copyright now affords author and publisher. Under the conventional system of publishing, the author and publisher are rewarded in accordance with the actual number of copies sold, and there is no problem in accounting for the copies. But as computer systems take over the function of "satellite" publishers, accounting for the number of copies becomes much more complex. Unless a system is created for the purpose there will be no accounting control even for the printed copy output of books, and it will be even more difficult to account for intangible copy "printouts" over closed-circuit television, as microfilm, and in other visual forms. This possibility of leakage may reduce the incentive of authors to create and publishers to publish.

While the computer "revolution" may reduce some of the inventives to create intellectual property, it may also provide something in the way of a new incentive that at least in part, offsets the reduction.

In an economic sense we usually argue that when something is substituted for something else, it is because it is more efficient or, in terms of cost, it is lower in cost. Therefore, to the extent that computer tape becomes a substitute for books, it may create an incentive to create certain works that otherwise may go without being created. Now, a likely developinent that is related to this is that the computer manufacturers themselves will increase their demand for intellectual property.

The demand for computers will increase as the volume and variety of services the computer can satisfactorily perform increases. In the past, success in the computer manufacturing industry has depended largely upon the computer manufacturer's ability to provide the requisite servicing, including programing.

This raises the very important question of how computer manufacturers, some of whom have already acquired publishing houses, will obtain the intellectual property they program. Will they enter into agreement with authors who retain copyright, or will they employ authors as they now do programing and systems specialists so that the author's creations are considered work made for hire? Were the latter to materialize, it would pose an important issue of concentration of control over the creation, storage and dissemination of knowledge and information. Under the present system of copyright, control is dispersed over tens of thousands of authors and about 900 publishing houses. Almost the entire output of high-speed digital computers is in the hands of about eight companies.

In summary, the essentail thrust of my analysis is that the technological revolution in information storage and dissemination is apparently upon us. It brings with it both great opportunity and great challenge. The spectacular advances in copying machines have made it possible to reproduce excellent copy almost instantly and at a cost per page of a few cents-or in sufficient volume, of a fraction of a

cent. The benefit of this to society, especially to the process of instruction in schools and colleges, is enormous.

The new advances in the technology of reprography establish strong economic incentives to infringe the copyright or to argue for its modification to permit freer use of copyrighted material. The surveys cited in this analysis indicate that the incentives to infringe and circumvent the copyright are already at work in the schools and colleges and will very likely become stronger as new technological advances

occur.

The question cannot be answered so simply in terms of economic welfare, but there are economic solutions to the problem. Under the new technology the incentive to use selections from textbooks, reference works, and other instructional material has greatly increased because the cost of reproducing them for classroom distribution has been substantially reduced. This implies that the demand for parts of books has increased, possibly but not certainly at the expense of demand for additional copies of the entire book.

If this is in fact the economic issue posed by the new technology in reprography, there is little reason to suppose that the market mechanism cannot resolve this supply-demand problem. Suppliers of books-authors and publishers must design some method whereby parts of books or, more practicably, the rights to reproduce parts of books, may be sold. Unless they do this, either present methods of selling books and the copyright will frustrate the new technology, or the incentive to circumvent the copyright will lead to unlawful reproduction or legal reproduction through copyright revision. Not only is neither of these solutions socially desirable but, in addition, neither is likely to produce an optimum allocation of our resources to the production of knowledge or optimum patterns of such knowledge. In brief, the copyright to a book should not obstruct extraction of the maximum use of a book, but neither should the increased incentives to use books in away different from past patterns lead to emasculation of the property right. The conflict of seller and buyer can be resolved here, as can most other buyer-seller conflicts, through the pricing system.

The economic issues posed by the electronic computer are more complex and less definable; the observable trends have been too short in duration and the technological possibilities too much a matter for speculation to point with certainty to what the future might bring. But two developments appear reasonably certain: there will be an increasing demand for "publishing" the contents of such intellectual creations as books in the form of computer tape, and for this reason the contents will increasingly be "read" from visual-image substitutes for the present printed page. It is possible also that copies printed out from computer tape will be substituted for both visual-image computer output and conventional published copies.

While these problems are, as has already been indicated more complex than those posed by the new means of reprography, they are not fundamentally different. Nor do they appear to require radical solutions. The new technology in information storage and retrieval, as in the case of all other innovations in history, requires market and institutional adjustments. As the demand for books in the form of tape increases, publishers, at least at present, are in the best position

to fulfill it and should do so. Otherwise, effective use of the new technology will be hampered. But there appears to be no fundamental difference between the social necessity of protecting intellectual property rights in film and tape and that of protecting such rights in visible and tangible printed books. Similarly, there is no fundamental distinction to be drawn between reproducing copyrighted material on a Xerox machine and reproducing it from tape or film as visual images or printed copy.

If one violates our laws or private property and dulls the incentive to create, so does the other. There is, therefore, a need to clear up any uncertainties that may still exist about whether intellectual property created, stored, and retrieved in one form is on legal parity with that created, stored, and retrieved in other forms. If the copyright is economically justifiable for one, it surely should be justifiable for the other. In any case, the form the property takes should have relatively little bearing on the copyright principle.

I think, Mr. Chairman, that that summarizes the written statement. (The prepared statement of Professor Markham follows:)

STATEMENT OF JESSE W. MARKHAM, PROFESSOR OF ECONOMICS AT PRINCETON UNIVERSITY

I am Jesse W. Markham, Professor of Economics at Princeton University. For the past ten years I have devoted considerable research and study to economic issues involving intellecutal property, especially the economics of patents and copyrights. Since its beginning, I have been a member of the research staff of the Patent, Trademark and Copyright Institute of George Washington University. Since 1961, I have been Economic editorial advisor to one of the better known publishing houses. I have written, and hold copyrights to a few books myself. Let me add in all candor, however, that the issues of copyright are complex and there is a good deal about them I do not know.

I appear here today on behalf of the membership of the American Book Publishers Council and the American Textbook Institute. My reason for appearing is that I acted as coordinator and economic advisor to both of these organizations in a comprehensive study of the present and future effects of information-transfer technology on book publishing.

The findings of these studies appear in An Economic-Media Study of Book Publishing and are pertinent to copyright law, the matter before this committee. Since copies of this study have been sent to this committee, I do not plan to present all the findings here; I will discuss the more important ones.

Much of what I shall say centers on the implications the vast strides in the new information technologies hold for copyright. There is no question but that the new technologies create important problems for the institution of copyright; however, I wish to make it clear at the outset that the remedy is not to hold back technical progress but to adapt our legal institutions to better accommodate it. The social benefits of the more efficient methods of creating, packaging, storing and disseminating knowledge and information are immense. It is clearly not in the best interests of a democratic society that the use of these new and more efficient information technologies in any way be frustrated or retarded by institutional obstacles. On the contrary, in this area perhaps more than any other technical progress should be given all the encouragement our legal institutions can provide. In the simplest possible terms, the economic function of copyright is the incentive it provides those with the requisite ability to create intellectual property and the incentive it provides publishers, recording firms, and similar enterprises to package and distribute these intellectual creations. This statement puts the case for copyright much too simply, so I would like to elaborate and clarify what it means.

First, I do not wish to imply that the creation of all intellectual property depends upon the existence of copyright protection. Clearly it does not. Some of the classics of literature, science, and music to which each generation has access as a cultural inheritance were created with little if any consideration for copyright protection. However, the public would surely be less familiar with these creations if it were

not possible to copyright certain versions of them, and treatises on them. Second, the many scholars, scientists, and artists who may not be stimulated directly by the financial incentive afforded by copyright are very often stimulated by the incentive of publication; the copyright, in providing publishers the incentive to publish, indirectly provides scholars the incentive to create.

Finally, in the case of textbooks and similar instructional material, it is reasonably safe to assume that both author and publisher are motivated by the prospects of financial reward. The writing of textbooks is generally considered to be more a matter of creatively assembling existing knowledge for purposes of effective instruction than as a matter of creating new knowledge. Those who write them may sometimes do so for purely professional reasons but income from royalties is their principal incentive, and profit rather than prestige generally provides the publisher with the incentive to publish. The financial regard of both author and publisher is greatly dependent on copyright.

In sum then, while all intellectual creations do not come about because of copyright, it is clear that copyright plays a very important role in stimulating creative activities and in the packaging, publication and wide distribution of the results that follow from it.

I have asserted that copyright protection is essential for the publication and distribution of scholarly works; I now wish to demonstrate why this is so. The necessity for copyright protection arises from the substantial differences in costs incurred by the original publisher and by those who copy or imitate the original publication. (Hereafter, I shall use the word publication to include all creations subject to copyright.) This difference in costs is attributable to 1) the risks attending the publication of a manuscript that has not yet undergone the commercial test of the marketplace, and 2) certain prepublication costs incurred only by original publishers assuming these risks.

The original publisher incurs risks because the commercial value of an untested manuscript when carried through to a salable manufactured copy can be predicted only in terms of broad ranges. It is therefore inevitable that while some books turn out to be best sellers, some barely break even, and some lose money. In the absence of copyright protection other publishers and copiers could avoid such risks by confining their copying to books that turned out to be commercially successful. In avoiding the risks they would avoid the attending costs of bearing them.

There are also prepublication costs that fall exclusively on publishers assuming the risks of first publication. Data supplied by the ATPI to Subcommittee Number 3 of the House Judiciary Committee on May 26, 1965, show that "a single high school textbook will require on investment of $50,000 before the first copy is available for sale." This investment includes such items as advance to authors, editorial and contractural costs, book design, galley and page proofs, and initial inventory. In addition, original publishers must pay royalties to the author. An essential economic justification for copyright, therefore, is that it provides publishers with an incentive to publish original material by providing an opportunity to recover the additional costs it entails.

It follows from this that the greater the amount of the difference in costs between original publications and copying, the greater is the need for copyright. Stated another way, the lower the cost of copying an original work, the greater is the incentive to circumvent the copyright. It is fairly clear from the detailed information developed in the studies I helped coordinate for the ABPC and the ATPI that the giant technological strides in copying machines and in information storage and retrieval systems built upon the electronic computer have created strong incentives to circumvent copyright. There is every reason to believe that future technological strides will be equally as impressive, and bring with them corresponding incentives to circumvent copyright.

A study carried out by Arthur D. Little, Inc., shows that over the past two decades, copying costs per page have declined dramatically as copying technology has advanced. It is now possible, where a modest sacrifice in quality is acceptable and where no binding is required, to copy on common offset paper twenty or more copies of a standard text book at a cost per copy measurably below the price of the textbook. Studies carried out by the National Opinion Research Center of the University of Chicago show that in elementary and secondary schools and in colleges, copying machines are used with considerable frequency in copying such copyright material as textbooks, tests and answer sheets, and music scores, although copying is generally limited to several pages.

Recent and projected advances in copying technology pose immediate problems for public policy toward copyright in that they dramatically reduce the cost and