First Monday

Is copyright necessary? by Terrence A. Maxwell

Is copyright necessary? by Terrence A. Maxwell
Copyright is a legal mechanism for promotion of useful knowledge. However, it is not the only means society could use to encourage information dissemination, and several alternative models have been suggested over the last 200 years. This article provides the results of a dynamic simulation of the publishing industry in the United States from 1800 to 2100, and tests the impact of different protection schemes on the development of authorship, the publishing industry, and reader access. It closes with a discussion of intellectual property information policy decisions that can be currently made, and their likely impacts on domestic and international copyright protection.


Modeling publishing relations
Model assumptions
Model results
The future of copyright
Discussion and conclusion





Copyright is a limited monopoly granted by the U. S. Constitution. Both before and after the Constitution’s creation in 1787, policy–makers debated its necessity and effectiveness in meeting the policy goal of promoting and disseminating useful knowledge (Jefferson, 1995; Walterscheid, 2002). During the nineteenth century, the focus of the argument shifted to the rights of authors to compensation against the right of members of the public to purchase inexpensive copies of works to aid in the dissemination of knowledge (Greeley, 1840; Carey, 1868). More recently, authors such as Stephen Breyer (1970) and Lawrence Lessig (2001) have questioned the economic and societal efficiency of copyright.

The American debate about the necessity of copyright has revolved around several issues, including the concept of authors’ natural rights, the balance between author’s rights and the public domain, and the economic impact of copyright on the protection of information industries and the diffusion of knowledge through society.

This paper focuses on the third area, particularly the economic impacts of the copyright regime on authors, the public, and publishers. It presents a dynamic simulation model (Ventana, 2000; Sastry and Sterman, 1992; Senge, 1990) of the publishing industry based on historical relationships between authors, publishers and the marketplace. It does not model the production and sale of individual works, nor does it study the publishing industry relative to other media industries. However, within the framework of print publishing, the model provides a platform to analyze the impact of high and low levels of copyright protection and other factors on the relative economic conditions of authors, publishers and readers over time.



Modeling publishing relations

In determining the necessity of copyright as a tool for the dissemination of knowledge and a method for recompensing authors for their intellectual effort, it is necessary to determine its impact on the following groups: authors, publishers, and the general public. Each has different economic goals and constraints on their behavior that affect their decision–making with respect to the creation, production, sale and purchase of information products.

Authors, for instance, are generally motivated both by a desire to see their writings disseminated and the necessity to maintain a high enough standard of living to support themselves and their families. Publishers hope to achieve a profit through the production and distribution of books and other information products. This they do by selling as many products as possible at the highest possible price. They also seek to maximize the use of their production equipment, reduce the costs of production and distribution, and generate capital income to expand capacity. In contrast, the public is motivated to purchase books and other information goods at the lowest possible price, maximizing the value they receive when purchasing information products (Greco, 1997).

These economic players and their sometimes competing, sometimes complementary, desires represent a system of transactions found in the book publishing industry, within which the copyright regime plays a critical role. The expressed goal of copyright within such a system is to protect authors’ expression by providing protection against others taking their works without compensation, while allowing for maximum dissemination of knowledge for the benefit of society. This philosophy was articulated in a Supreme Court decision handed down in 1984, in which it was stated:

"The monopoly privileges that Congress may authorize are neither unlimited nor primarily designed to provide a special private benefit. Rather, the limited grant is a means by which an important public purpose may be achieved. It is intended to motivate the creative activity of authors and inventors by the provision of a special reward, and to allow the public access to the products of their genius after the limited period of exclusive control has expired" [1].

Understanding the complexity of the publishing system within which copyright operates is not, however, limited to understanding the varying goals of authors, publishers and the public. The system’s complexity, and hence the difficulty in determining the impact of high and low levels of copyright protection, lies also in the fact that the variables necessary to model these relationships interact. These interactions are summarized below.

Author interactions

Given authors’ desires to achieve a living wage, we can expect that the total number of authors in a society will be affected by the amount of royalties paid to authors and the demand for new works by publishers and the public. Higher average royalties paid to authors will attract more writers into the profession, while lower average payments will tend to drive people into other professions. Yet even if the average level of payment does not allow for recompense at a level necessary for full–time employment, a high demand for new works and the greater likelihood of publication may attract people into the field of authorship, In addition, the promise of copyright protection may help attract people into the writing field, if they feel their efforts will be adequately protected against piracy for a long enough period to extract a payment sufficient to repay them for the effort generated in developing their work.

Publisher interactions

Publishers, even more than authors, seek to maximize profits. In doing so they, too, are buffeted by competing priorities and variables. Maximizing profit is achieved by selling books at the highest price the market will bear, while minimizing both fixed costs (e.g. administration, plant) and variable costs (e.g. printing, distribution, payment to authors). They are also driven to improve their production capacity by investing in new equipment that will allow for more production at a lower per unit cost, and seek to increase their size and market share to a level that discourages other firms from entering the publishing field. Given these requirements, publishers are affected by a variety of factors, including the availability of author works and the prices they must pay for them, the size of the market and its level of saturation, the number of other publishing firms operating in their field, funds available for capitalization of new equipment, and the technological capacity of their current equipment. These factors, in turn, help to determine the number of volumes publishers produce each year and the average number of copies of each volume they print. The publishers’ goal is to maximize both net profit and utilization of human and physical resources (Greco, 1997).

From the publishers’ perspective, the role of copyright is to clarify ownership of a work by an author, and thus ensure that the proper transfer of publication, distribution and related rights can be effected by way of contracts. It also provides a mechanism for enforcement against infringing acts by others.

Public interactions

As the consumer of information goods, the public is the ultimate arbiter of a work’s commercial success or failure. On a general level, the public’s ability to purchase a book is based on its ability to find a volume in its area of interest, literacy level, the book’s price and the availability of surplus income for book purchases.

The public’s perception of copyright protection is conflicted. On one hand, copyright is seen as a mechanism to provide a just reward for authors’ labor, consistent with communal notions of fairness. On the other, individuals seek to access information at the lowest cost. In this regard, copyright, if viewed as a mechanism for increasing the purchase price for consumers or as a means for controlling the marketplace, is seen as at best a nuisance and at worst a barrier to the free flow of information. The degree to which copyright might be viewed as a positive or negative value would depend on the degree to which it is perceived as impeding the flow of low–cost information products.

In short, the relationships and key variables in the publishing industry can be summarized in the following Figure 1.

Figure 1: Relationships in the publishing industry.

In order to adequately model the relations outlined above, we require a simulation environment with a requisite level of complexity to capture multiple interactions over time. To achieve this, we utilize a system dynamics computer environment (Ventana, 2000) designed to model the impacts of policy options on complex systems, where the endogenous structure of the relations among parts of the system generates behaviors that may not be apparent by studying dyadic and linear relations among a limited number of variables. In order to model the relations noted above, a multi–variable model, containing linear and non–linear equations, was developed (A list of the model variables is included in the Appendix). The following figure provides more detail about the causal interactions among variables in the model, and the direction of impact a variation in one key variable will have on others to which it is connected. For example, an increase in capital investment will lead to an increase in production capacity, which accounts for the positive sign at the head of the arrow connecting these two parts of the model. In contrast, an increase in market saturation will decrease the average price of books. Hence a negative sign is appropriate for this connection.

Figure 2: Summary of causal relations among model sectors.

Assuming the relationships identified above are accurate, it is possible to quantitatively model their interactions, and identify the impacts of different policies on the behavior of this system. The following provides an analysis of the impact of intellectual property protection on the development and payment of authors, the size and health of the publishing industry, and the capacity of readers to access works. It is based on an historical view of publishing industry development based on data about American publishing during the nineteenth century (Bradsher, 1966; Lea, 1885; Lehmann–Haupt, 1939; Tracy, 1913; Unger, 1998).

The choice of an historical perspective is due to the following reasons. First, the U.S. Constitutional rationale for providing monopoly protection was based on the assumption that providing a limited monopoly for authors would lead to "the progress of science and the useful arts" [2] and provide a means for knowledge promotion within a market society. Therefore, modeling the effects of high and low levels of copyright protection would provide a test of the underlying assumptions behind the establishment of copyright monopolies. Secondly, we have historical data available on the growth of American publishing in the nineteenth century, literacy and population rates in the United States during this period, and the timeline of technology development. These provide important baseline information for model specification and allow us to test model behavior against reality.



Model assumptions

In analyzing the impact of copyright on the publishing industry, it is necessary to make certain assumptions about the state of the publishing industry in 1800, the first year of the simulation. The following table provides an overview of some key variables. These variables provide a starting point for analysis and the model continually generates new conditions endogenously as the simulation advances.


Table 1: Model Assumptions based on 1800 data
(Sources: Bradsher, 1966; Charvat, 1968; Clark, 1960; University of Virginia Library, 2002)

Number potential American authors300
Number potential English authors for import500 with no international copyright,
200 with international copyright
Living wage requiredUS$100 per month
Length of time to produce a new work8 months
Base royalty rate16.66%
Number of major publishers18
Base number of copies in production run1,036
Base retail price per copyUS$1.50
Fixed cost per 1,036 copy production runUS$192
Variable per copy costUS$0.34
Technology improvements production capacityBased on availability of new volumes to support knowledge diffusion
Base U.S. literate population, 18001,890,000
Market growth3%/year
Funds available for book purchasesUS$2.00 per month (1/50th of income)


Since the assumptions above are based on fragmentary sources and data, it is important to determine the level of impact changes in any variable listed above would have on the overall behavior of the model. Sensitivity analysis was performed using multivariate sensitivity simulations (Ventana, 2000). The analysis indicated that the model was most sensitive to changes in the base retail price, variable per copy costs, and market growth rates. The base assumptions behind these variables are documented in U.S. Census data and the letters and journals of Mathew Carey, the largest American publisher in the 1800 time period (Bradsher, 1966; University of Virginia Library, 2002). Although other baseline variables were able to generate some variation in particular sectors of the models, they were not at a level capable of dramatically changing the results of policy simulations.



Model results

Given these assumptions and the relationships between variables discussed above, it is possible to model the effects of a variety of different copyright scenarios on the publishing system, and analyze the impact on the number and condition of authors, publishers, and the reading public over time. For the purposes of this analysis I have chosen a one hundred year timeframe, spanning 1800 to 1900, and tested three basic scenarios, consistent with the most extreme positions found in the nineteenth century copyright debate.

Authors: During the nineteenth century, the majority of authors who were vocal about copyright felt that works should be perpetually protected; English authors should be given the same treatment as Americans with respect to copyright protection; and, the position of authors should be protected against publishers (Andrews, 1872; Sedgwick, 1882; Eggleston, 1882). With regard to the last, this would mean that ‘work for hire’ provisions that established the legal fiction whereby publishers could be considered authors, resulting in a diminishment of the stance of romantic authorship, would not be supported.

Publishers: Like authors, publishers advocated strong copyright protection for American authors, but for nearly a century were not unanimous in their support for protection of foreign authors. Unlike authors, publishers argued that they should be deemed authors for the sake of copyright, supporting ‘work for hire’ provisions (Holt, 1888; Putnam, 1909; Brylawski and Goldman, 1976; Putnam, 1915; Bradsher, 1966; Exman, 1965).

Public: At its most extreme, the public stance regarding copyright would have supported no copyright monopoly either for domestic or foreign authors, and no ‘work for hire’ provisions to benefit publishers, with the expectation that any monopoly would tend to diminish the diffusion of information and potentially increase the cost of books (Carey, 1868; Matthews, 1887).

Due to the quantitative nature of the simulation model, copyright protection is treated as a continuum from .1 to 1, so that high copyright protection would be ten times stronger for authors and publishers than as low protection. In practical terms, we might view the differences between copyright at a level of .1 and that of 1 as the difference between copyright levels in 1790 (with protection for 14 years, minimal types of information products covered, and stringent requirements for registration that caused many books to immediately enter the public domain) and full copyright protection (perpetual copyright on nearly all information products with no process for public notice or governmental registration in order to establish copyright protection).

Table 2 shows the model scenario results based on the policy stances of the different players in the copyright question.


Table 2: Results of copyright simulations, 1800–1900

High copyright (1 in model), international copyright (200 foreign authors), no work for hire
High copyright, no international copyright (500 foreign authors), work for hire
Low copyright (.1 in model), no international copyright, no work for hire
Number of American authors30018,04316,4824,374
Monthly royalties per authorUS$6.94US$131.04US$115.37US$136.37
Number of major publishers18283033
Size of production run1,03612,65713,36114,738
Retail price per copyUS$1.80US$1.17US$1.13US$1.04
Total production capacity per publisher2,0001,001,0001,063,0001,182,000
Monthly profits per publisherUS$1,036US$564,203US$566,570US$559,318
Volumes published per month352,2352,3782,618
Books per customer sold per month0.006.74.831.00
Books per customer available0.0192.222.503.03


As indicated in the table, the desired policies of authors, publishers and public domain advocates produce very different outcomes in a 100–year simulation, some of which run counter to the protagonists’ stated goals. For instance, while the authors’ position led to the largest number of authors, it also generated the lowest sales figures, and the fewest number of volumes published. This indicates that the demand for new volumes from authors was the lowest among the three options, and points to a greater level of competition among authors seeking publication. Similarly, while the reader position generated the highest level of sales, the greatest number of different volumes, and the lowest cost for books, it also severely constrained the number of authors. This means that while a greater number of volumes would be available, diversity in authorship would be curtailed. This, in turn, would tend to diminish the likelihood of variety in information products.

Only the publishers’ stance fit their objectives, providing the highest relative level of profit, while maintaining the lowest royalty levels for authors. While book sales did not reach the level of the reader option, and the level of overstock was higher than in the author option (indicating a greater level of waste), the publisher’s strategy maximized their overall profit.

Was copyright the best policy option?

The model also provides us with the opportunity to ask the question whether copyright protection was the best scheme for increasing the number of authors, publishers, and improving customer choice in nineteenth century America. Could there have been other policy options Congress might have chosen to support the progress of arts and sciences, schemes that might have incurred less conflict and administrative and legal costs?

Nineteenth century commentary on copyright highlighted several potential policies government could have used to support or discourage the development of a national cultural inventory. Primary among the schemes discussed were:

High copyright control: At its highest, this would have involved giving authors perpetual rights, by granting their demand for treating the products of authorship as common law property, exempt from statutory limitations as to term (Greeley, 1840).

Limit educational opportunities for citizens: This ran counter to the nation’s early republican ideals, and as a result was not implemented. The impact of this policy would have been to decrease the rate of market growth for books.

Control the retail price for books: Publishers attempted to keep retail prices high by controlling the distribution network or utilizing legal mechanisms to restrain trade (Bradsher, 1966; Scribner v. Straus, 1908). This policy, if successful, would have been similar to the policies of the English Stationers’ Company and nineteenth century English publishers, who successfully sought to both limit the number of printers and control the book distribution network.

Balance author control with public information diffusion: This would involve a tradeoff between absolute and perpetual author rights and information diffusion, through the protection of a public domain, and the decrease in book quality, by lowering the variable cost of items like paper, ink, and binding materials. This, in fact, was most similar to the policy options implemented over the course of the century.

The following table shows the effects of these different policy actions over a hundred–year timeframe.


Table 3: Impacts of policy options on publishing development 1800–1900

High copyright control
Limit educational opportunities
Control retail prices
Balance author rights with public domain
Number of American authors9,5235,87114,72511,162
Monthly royalties per authorUS$132.84US$78.46US$152.30US$167.55
Number of major publishers31356028
Size of average production run14,1655,50719,79223,421
Retail price per copyUS$1.08US$1.02US$1.03US$.86
Total monthly production capacity per publisher1,132,000416,7671,247,0001,898,000
Monthly profits per publisherUS$568,253US$181,864US$580,090US$938,576
Volumes published per month2,4932,6883,7562,301
Books per customer sold per month.921.0531.941.41
Books per customer available2.783.178.164.24
Model variable operationalizationCopyright effect = 1Market growth = .02 per yearBase retail price = $2.50Copyright effect =.4, Variable cost ramped down 50% between 1840 and 1860


As shown in the table, the four policy options produced very different results. High copyright control produced the highest book price and the lowest number of books sold in the model. A policy that deemphasizing public education (and hence constrained the growth of the book market) produced the lowest number of authors, the smallest production runs and capacity among publishers, and the lowest profit both for authors and publishers.

Balancing author rights with the public domain constituted an effective balance relative to high copyright and small markets. It produced the second highest number of authors, the highest author income, the lowest retail price, the second highest sales level and volume production, and the highest profit levels for publishers. However this result came at the expense of lowering product quality and constraining the number of major publishers.

The most surprising policy result was the success of artificial price control, which provided employment for the largest number of authors at the second highest income levels. In addition, it produced double the number of publishers as other options. Retail book prices at the end of the period were the second lowest among all the policies, and sales the highest. The one negative effect of this scenario was the tendency toward over–production of books. While this result allowed for more consumer choice among competing products, it also generated more waste from an economic perspective.

Why was the retail price control option effective? It can be explained by the condition of the American publishing industry at the beginning of the nineteenth century, particularly with respect to a shortage of capital for expansion (Carey, 1942). By maintaining artificially high prices, publishers in this scenario would have been able to generate and utilize excess profits for capital development, allowing for more rapid expansion of the publishing industry than otherwise could occur. To illustrate, Figure 3 shows the growth of capital across the four scenarios, and indicates the most rapid development of capital in the "high price" scenario.

The impact of early capitalization can best be seen when compared to other alternatives. The following figures show the dynamics of publishing based on two different policy scenarios, high initial prices (leading to early capitalization) and the second most effective scenario, a balanced approach with moderate copyright. Figure 4 shows that early capitalization caused by the control of retail prices kept the price of books artificially high for only a short period of time (approximately 10 years) until the increase in capacity began to drive prices downward. The numbers of authors, publishers, and volumes published all rose steadily (while authors’ royalties increased rapidly), and then fell as supply caught up to demand.

In contrast, Figure 5, showing the dynamics of the balanced approach with moderate copyright levels, indicates a slower growth of publishers, authors and volumes. In addition, publisher capacity increased more slowly, meaning that retail prices took a longer time to drop.



The future of copyright

The results of the simulations above show that copyright protection, from an economic and policy standpoint, while good for authors, was not the most efficient policy the founders could have implemented to promote the rapid growth in the progress of arts and the sciences. Given the conditions of the publishing industry in the United States at the beginning of the country, a short–term policy of price controls would have had better long–run positive effects.

This of course raises the question whether such a policy could have been implemented. Certainly, Jefferson’s correspondence with Madison lends doubt to the founders’ willingness to establish systems that would have been perceived to disadvantage citizens and help create monopolies (Waltersheid, 2002). However, in light of the history of early tariff policy (Taussig, 1967), we can state that the United States was generally protectionist regarding the nurturing of native industries, and certainly a theoretical case could be made that internal controls were no different than external ones with regard to their impacts on both consumers and producers.

Given these results, does it make sense to implement price controls at the beginning of the age of digital production and distribution? To test this and other policy options, the same model was used to run extended simulations projecting to the year 2100. In doing so, the model was modified slightly to mirror the real increases of copyright protection embodied in the 1831, 1891, 1909, 1976 and 1998 revisions of copyright law. In addition, anticipated changes in distribution (an average decrease from 3 months to 2 weeks) and per unit variable costs (an average 30 percent decrease) due to the development of digital production and distribution were phased in over the years from 2000 to 2020.

Five policy options were tested against the modified model. The results of the option are summarized in Table 4 below.

High copyright protection: This option modeled an increase of copyright protection to perpetual control, with strong protections against piracy and infringement.

Balance copyright control with public information diffusion: This policy would involve a rollback of copyright protection to levels below current conditions, through policies such as a decrease in the years of protection and restraint on technical controls over electronic distribution and copying that negatively impact the first sale doctrine and fair use.

Limit public education: A shift to a more modest market growth (one percent) was tested in this simulation. This assumes that American market growth would start to slow as population growth slows and the level of literacy reaches its upper limits.

Improve access: This envisions an aggressive policy of market expansion (modeled as an increase of one percent over the base, to a total of four percent per year), both through strong educational incentives in the United States and aboard, and a strong presence of American publishers in an international market characterized by improved literacy.

Control retail prices: A policy of retail price control that would immediately increase retail prices for information prices by 30 percent, either through policy interventions or technological controls on information products, was modeled in this simulation.


Table 4: Impact of policy options on publishing industry through year 2100

Strong copyright control
Balance author rights with public domain
Limit public education
Improve access
Control retail prices
Number of American authors532,392293,862260,1951,005,000389,323
Monthly royalties per authorUS$1,576US$1,653US$1,052US$3,987US$1,745
Number of major publishers6566706271
Size of average production run (in millions)3.133.411.6815.523.87
Retail price per copyUS$.75US$.71US$.65US$.90US$.73
Total monthly production capacity per publisher (in millions)259.24266.41115.481,293.00243.64
Monthly profits per publisher (in millions)US$111.25US$107.83US$41.2US$701.89US$101.84
Volumes published per month5,3855,1684,8225,1404,441
Books per customer sold per month2.682.803.072.232.73
Books per customer available5.095.857.384.018.34


Table 4 shows that, in contrast to the simulation of early publishing conditions, controlling retail prices today would not be an effective policy. Instead, the rapid development of new markets through improvement of access is of far greater importance. This is because the publishing market is in a mature stage, with enough capacity to produce new products in order to meet increased demand. As such, a policy of improving literacy both in the United States and abroad, while insuring open access to new markets, is the best policy option among the alternatives studied.

A comparison of strong copyright control and one that balances authors rights against the public domain shows only moderate impacts on the market, with the exception of a significant difference in the total number of authors (better with high copyright protection) and author royalties (better with moderate protection).

It is important to note that the model does not include the administrative and legal costs associated with different policy options. For instance, are the costs necessary for policing and enforcement of copyright and other intellectual property claims greater than those of implementing price controls, or promoting improved literacy? These and other potential impacts of copyright protection, digital distribution and production, and other variables were beyond the scope of the current model.



Discussion and conclusion

The results of the policy simulations have important implications for policy makers and copyright scholars. First, they show that the presence or absence of copyright protection is most critical to the number of authors in a society, but has little effect on publisher expansion and profits, or on consumer prices. To the extent that society seeks to promote the growth and diversity of authorial perspectives, copyright can be an effective policy choice. This raises the related question whether recent efforts to expand corporate authorship through expansion of ‘work for hire’ provisions (Rosenthal, 2000) act in concert or contradiction to the policy benefits of copyright protection. Given the fact that a weak ‘work for hire’ copyright regime is most beneficial to individual authors, the answer seems to be that the expansion of ‘work for hire’ runs contrary to the positive policy effects of copyright.

Secondly, the simulations show that the same policy options have different effects depending on the condition of the publishing industry at any stage of development. During early stages of development, capital requirements are far more critical to the overall health of a publishing community than copyright or other policy options. This has implications when viewing American copyright policy from the perspective of developing countries. While some commentators have suggested that strong copyright protection is beneficial to third–world development because it attracts overseas capital investment (Maskus, 2000), this theoretical advantage must be balanced against the internal capacity of societies to support the development of internal publishing capacity, native authorship and an internal market for domestic information goods (Friedman, 1999). While the present model does not simulate the multi–country dynamics of information sale and production, we can see from the simulations of the early American publishing industry that an open market approach with strong international copyright protection might attenuate the ability to develop domestic publishing capacity, thereby retarding the growth and health of publishing and authorship in developing countries. In contrast, the model suggests that a system of developing country trade protectionism, coupled with moderate copyright protection and initial price controls, might be more beneficial to the development of domestic publishing in third world countries. This, of course, runs contrary to trade policies advocated by the United States and the World Intellectual Property Organization, but is generally consistent with early United States trade and copyright history.

Additionally, the impacts of copyright and other policy options on mature publishing industries have important implications for current debates about copyright in the digital age. The simulations indicate is that current efforts to strengthen copy control over information products in order to keep prices at a level similar to pre–digital periods is not necessarily the best policy direction. As indicated in Table 4, artificial efforts to maintain or increase prices in mature markets leads to less choice in books, lower publisher profits, and over–production of information products. In contrast, an aggressive policy of market expansion, both internally and externally, would be far more beneficial to all concerned. To the degree this policy could include the expansion of educational and information access opportunities and support both at home and abroad, it might meet with little resistance. However, if American publishers attempt to expand their markets overseas through overwhelming domestic information markets in smaller and less developed countries, they are likely to meet resistance from countries attempting to protect fledging publishing industries and native cultures.

Finally, returning to the original question regarding the necessity of copyright, we come to different conclusions from commentators who advocate for the absolute necessity of this legal mechanism for the promotion of information. However, our conclusions are also at variance with those who advocate its abolition or at least a significant relaxation of current constraints. Rather, we suggest that copyright should be viewed in relation to other potential policy options and to the goals we wish to achieve in supporting intellectual production and distribution both nationally and internationally. As such, any adjustment to the level and degree of copyright protection must be considered in relation to decisions about trade, education, literacy and technology, to achieve and maintain an effective balance between promotion of new knowledge and a healthy public domain. End of article


About the author

Terry Maxwell is Assistant Professor at the Graduate School of Information Science and Policy, University at Albany.
E–mail: tamaxwell@hvc.rr.dom



1. Sony v. Universal Studios, 464 U.S. 417 (1984).

2. U.S. Constitution, Article 1, Section 1, cl. 8.



D.A. Andrews, 1872. "Brief on behalf of authors and publishers in favor of international copyright," North American Review, volume 114, pp. 432–435.

E.L. Bradsher, 1966. Mathew Carey, editor, author and publisher: A study in American literary development. New York: AMS Press.

S. Breyer, 1970. "The uneasy case for copyright: A study of copyright in books, photocopies, and computer programs," Harvard Law Review, volume 84, pp. 281–351.

E.F. Brylawski and A. Goldman (editors), 1976. Legislative history of the 1909 Copyright Act. South Hackensack, N.J.: F.B. Rothman.

H.C. Carey, 1868. Letters on international copyright. Second edition. New York: Hurd and Houghton.

M. Carey, 1942. Mathew Carey autobiography. Brooklyn, N.Y.: E.L. Schwaab.

W. Charvat, 1968. The profession of authorship in America, 1800-1870: The papers of William Charva. Columbus: Ohio State University Press.

A.J. Clark, 1960. The movement for international copyright in nineteenth century America. Washington, D.C.: Catholic University of America Press.

E. Eggleston, 1882. "The blessings of piracy," The Century, volume 23, pp. 942–945.

E. Exman, 1965. The brothers Harper: A unique publishing partnership and its impact upon the cultural life of America from 1817 to 1853. New York: Harper and Row.

T. Friedman, 1999. The Lexus and the olive tree. New York: Farrar, Straus, Giroux.

A.N. Greco, 1997. The book publishing industry. Boston: Allyn and Bacon.

H. Greeley, 1840. "Rights of authors," Southern Literary Messenger, volume 6, number 1, pp. 69–72.

H. Holt, 1888. "Some practical aspects of the literary life in the United States," New Englander and Yale Review, volume 48, pp. 155–188.

T. Jefferson, 1995. The republic of letters: The correspondence between Thomas Jefferson and James Madison, 1776–1826. Edited by J.M. Smith. New York: W.W. Norton.

H.C. Lea, 1885. One hundred years of publishing — 1785-1885. Philadelphia: Lea Brothers.

H. Lehmann-Haupt, 1939. The book in America: A history of the making, the selling, and the collecting of books in the United States. New York: R.R. Bowker.

L. Lessig, 2001. The future of ideas. New York: Random House.

K.E. Maskus, 2000. Intellectual property rights in the global economy. Washington, D.C.: Institute for International Economics.

B. Matthews, 1887. "Cheap books," The Century, volume 35, p. 328.

G.H. Putnam, 1915. Memories of a publisher, 1865–1915. New York: G.P. Putnam’s Sons.

G.H. Putnam, 1909. "Analysis of new copyright law.," Publishers Weekly, pp. 1100–1105.

J. Rosenthal, 2000. "Work–for–hire bill repealed," Billboard (28 October), p. 5.

M.A. Sastry and J.D. Sterman, 1992. "Desert island dynamics: An annotated survey of essential system dynamics literature,", accessed 4 September 2004.

Scribner v. Straus et al., Supreme Court of the United States, 210 U.S. 352 (1908).

A. Sedgwick, 1882. "Constitutional protection of property rights," North American Review, volume 135, pp. 253–266.

P.M. Senge, 1990. The fifth discipline: The art and practice of the learning organization. New York: Doubleday/Currency.

F.W. Taussig, 1967. The tariff history of the United States. New York: A.M. Kelley.

G.A. Tracy (compiler), 1913. History of the Typographical Union. Indianapolis: International Typographical Union.

Twentieth Century Music Corp. et al. v. Aiken., Supreme Court of the United States, 422 U.S. 151 (1975).

H.G. Unger, 1998. Noah Webster: The life and times of an American patriot. New York: Wiley.

Ventana, 2000. Vensim 4.0 Manual. Harvard, Mass.: Ventana Systems.

University of Virginia Library, 2002. "Historical Census Browser," at

E.C. Walterscheid, 2002. The nature of the intellectual property clause: A study in historical perspective. Buffalo, N.Y.: W.S. Hein.


Appendix: Model variables

Publishing sector

Capital per publisher=capacity per publisher*cost per unit of cap. change
Capacity per publisher=production capacity/publishers
Production capacity= INTEG (capacity adj-depreciation rate,36000)
Books produced= INTEG (production-shipped,36000)
Depreciation rate=production capacity*0.004
Cap. adjust time=6 ~6 months time to adjust capacity
Capacity adj=((capital investment/cost per unit of cap. change)*technology multiplier)/cap. adjust time
Capacity adjustment=production capacity
Knowledge diffusion f([(0,0)(100000,0.02)],(100,0.006),(1000,0.008),(5000,0.01), (10000,0.015), (100000,0.016))
Technology multiplier=1+(knowledge diffusion f (knowledge diffusion capacity))
Knowledge diffusion capacity=adj. volumes published*saturation distribution effect
Net Profit= INTEG (income-outlays, 18648)
Outlays=authors royalties+capital investment
Profits per publisher=Net Profit/publishers
Gross revenues=sold*adj per unit price
Net/gross=Net Profit/gross revenues
Income=adj. revenues-production costs
Funds available for distribution=Net Profit*0.9
Adj. revenues=gross revenues*(1-(distribution discount+admin costs))
Capital investment=funds available for distribution-authors royalties
Publishers= INTEG (new publishers,18)
New publishers=((capacity per publisher*market saturation*per book production costs)/capital investment) /publisher time to adjust
Publisher adjustment=publishers/base publisher number
Publisher time to adjust=3 ~months to adjust number of publishers
Adj. production run=(base production run+adj vol/run ratio)/publisher adjustment
Adj. volumes published=capacity adjustment/adj. production run"
Adj vol/run ratio=capacity adjustment/(base production run*market saturation effect)
Pressure for new volumes=adj. volumes published/volumes available
Variable cost per unit=0.34*(1-(technology multiplier/6))
Production costs=(fixed cost per volume*adj. volumes published)+(variable cost per unit*adj. production run\*adj. volumes published)
Production=adj. production run*adj. volumes published
Per book production costs=production costs/production
Shipped=books produced
Books for distribution= INTEG (shipped-sold,36000)
Sold=MIN(potential unit sales, (books for distribution*saturation distribution effect)/time for distribution adj )

Author sector

Authors= INTEG (new authors,300)
Authors royalties=funds available for distribution*(base royalty rate*literary property effect*work for hire adjustment)
Royalties per author=authors royalties/authors
Volumes available=(authors+foreign authors)*output per author
Author adj.=(copyright vol adjustment*wage adjustment)*authors
New authors="author adj.-authors)/author time to adjust
Wage adjustment=(royalties per author)/living wage

Intellectual property variables

Work for hiref([(0,0)(1000,10)],(0.5,0.7),(0.9,0.75),(1,0.8),(1.2,0.9),(2,1),(30,1.1), (100,1.2), (300,1.3), (500,1.4))
Work for hire adjustment=work for hire f(((pressure for new volumes+publisher adjustment)/2)/work for hire condition)
Trade courtesy=net/gross*market price adjustment ~Trade courtesy is highest when there is scarcity in the market, causing a demand for all books produced
Literary property effect=(copyright effect*0.7)+(trade courtesy*0.3)
Copyright vol adjustment=pressure for new volumes*literary property effect

Reader sector

Market saturation effect=market saturation f(market saturation)
Books per customer sold=sold/market
Mkt. change=market*(market growth rate/12)
Market= INTEG (mkt. change,1.89e+006)
Funds available=market*funds per person
Market saturation=MAX(0.01,(books for distribution/market))
Market saturation f([(0,0.5)-(10000,3)],(0,1),(2,1.5),(10,0.95),(20,0.9),(50,0.85), (100,0.8), (500,0.75), (1000,0.7))
Adj per unit price=base per unit price*market price adjustment
Market price adj f([(0,0)-(10000,2)],(0,1.2),(0.5,1.2),(0.8,1.1),(1,1),(1.5,0.9),(2,0.8), (3,0.7), (4,0.6), (5,0.5),(11,0.33),(25,0.2)) ~As market saturation increases, price decreases. Starting at .5 books per person, the price is 100%, but decreases to 90% at 1 book per person, and 20% at 10 books per person
Market price adjustment=market price adj f(market saturation)
Potential unit sales=funds available/adj per unit price v Sat. distrib. f([(0,0)-(10000,1)],(0,1),(20,0.98),(40,0.95),(60,0.9),(80,0.85),(90,0.8), (100,0.75),(1000,0.6))
Saturation distribution effect=sat. distrib. f(market saturation)

Policy levers (altered to simulate different environmental conditions)

Work for hire condition=0.01 (Range .01-1)
Foreign authors=200 (Range 200-500)
Base royalty rate=.16667 (Range .05-.2)
Base publisher number=18 (Range 10+)
Market growth rate=0.03 (Range .01-.05)
Copyright effect=1 (Range .1-1)
Funds per person=2 (Range $.50-5.00)
Author time to adjust=48 (Range 36+ months)
Base production run=1000 (Range 800+)
Time for distribution adj=3 (Range .5+)
Living wage=100 (Range $50+)
Admin costs=0.1 (Publisher administrative costs. Equilibrium 10% gross revenues. Range .05-.2)
Distribution discount=0.2 (Discount for retail distribution. Equilibrium 20% gross revenues. Range .10-.5)
Fixed cost per volume=192 (Costs for setup, editing and prepress for each volume. Set in equilibrium at $192, based on Carey numbers of 1800. Range $100+)
0utput per author=0.12 (Output per month. Equilibrium is .12, or one volume every 8 months. Range .05+)
Base per unit price=1.5 (Base price per book sold. Equilibrium is $1.50. Range $.10+)
Cost per unit of cap. change=10 (Range 1+. $10 = equilibrium)

Editorial history

Paper received 31 March 2004; revised 9 May 2004; accepted 24 August 2004.

Contents Index

Copyright ©2004, First Monday

Copyright ©2004, Terrence A. Maxwell

Is copyright necessary? by Terrence A. Maxwell
First Monday, volume 9, number 9 (September 2004),