The Evaluation of Software Investments Regarding Proprietary vs. Free Software A thesis presented by

Andr´as Breuer

to The Department of Finance Faculty of Economics in partial fulfillment of the requirements for the Investment Analysis and Risk Management major

Corvinus University of Budapest Budapest, Hungary

Thesis advisor: G´eza Sebesty´en

April 2006

Summary Free software is something more and more people know about, but only few can tell what it exactly is, and most do not know how it works. There has been little academic research on its economic properties. Thus it is a promising field to get involved with. At the same time it is also challenging – as it is equally easy to get something wrong. This paper will attempt to answer a number of questions. What is free software and why is it important? What may the reasons be for the growing use of free software? How and why does it get created? What are the special characteristics of the software industry, which allow for the appearance and flourishing of a new information-technology paradigm? What is the importance of enterprise distributions – and what are distributions in general? Answers to the above questions will be used to gain some insight into the main question of the paper, that is: What is the total cost of ownership (TCO) of free software compared to its main alternatives? The paper will offer a framework of analysis that should help with making a capital investment decision in the field of IT. Free software represents a vastly different approach to the creation, use and reuse of information, than what the widely accepted paradigm of intellectual property suggests. The term “intellectual property” is often used to reflect or stress the value and importance of such proprietary information. Thus, with regards to economic theory, linking it to the better researched field of tangible assets. In recent decades the weight of the service sector as opposed to the agricultural and industrial sectors has increased greatly. It is generally accepted that there has recently been a move towards a “new information economy/society,” where the value of information and knowledge outweigh that of many tangible assets. Free software plays a peculiar role in this new world. It seems to disregard some basic means of value creation – it is often distributed without a fee incurred –, and at the same time, it seems to be of immense value. From a mainstream microeconomic perspective it seems counter-intuitive that a number of volunteers would collaborate on a project as complex as creating an operating system. It lies deep in the traditions of mainstream microeconomics that monetary incentives, such as a salary or a contract fee are essential to production – as is capital. It seems that other motives can be just as important as financial ones. In addition, in industries that have almost no capital need, time (or work) is the most scarce resource. The archetypical center of most microeconomic theories is the traditional factory. Thus it may not be sur2

prising after all that in a field highly unlike that of any industry or agriculture a different system of production may be successful. This is precisely the case with the production system of volunteer programmers (“hobbyists” if you will) collaborating via the Internet, who have created GNU/Linux (a free operating system) and Apache (the most commonly used web server software). Free software matters, because many of the leading companies of the informationtechnology (IT) sector rely on it for their core business. Google, for example uses a modified version of the free operating system Linux for the storage and processing of its massive amount of data. Apple uses a modified version of the free BSD operating system as the core of its own OS. Even Microsoft uses some code from BSD in its Windows XP and Windows 2003 Server products. Studios, such as Pixar or Disney have also used Linux to create a number of animated movies. Yahoo uses the free FreeBSD operating system on its servers. IBM, Sun, HP, Oracle, Novell and Intel have all invested considerable amounts of money in free software, and have also developed free software solutions of their own. These companies seem to be doing well lately, so there might as well be a sound reason for their decision to use and invest into free software. At the same time, free software is also more and more widely accepted in the public sector – especially so in Europe and Latin-America. Free software is here to stay. There is a great amount of free software code already written, and it is rapidly increasing. Many companies already use it and many are planning to do so in the near future. This paper has looked into four distinct aspects of this phenomenon: (1) the history of free software, (2) some special economic factors that influence the software market in general, (3) open source software development, and finally (4) what factors need to be considered when making an IT investment decision between free and proprietary software. The history chapter showed the roots of a new paradigm of production that is typical to knowledge-based goods. This chapter shows the importance of the scientific method and the ethics transparent in the scientific community of universities. Without these roots, most people who develop free software now, may not have been motivated to share their endeavors with everyone else. They may have tried to find other avenues to express their creativity. The history of free software is also the history of an old-new paradigm of production that somewhat resembles the functioning of NGOs that never before had such

3

a great economic impact. The product of this community competes in the market with the products of for-profit companies with more and more success. The second chapter deals with the special economic characteristics of the software market. It does not try to give a comprehensive economic model of all markets, it just emphasizes the special characteristics of the software market. The economic rules discovered in this chapter about the factors influencing success in this market are as true for free software as they are for proprietary software. They operate in the same market, after all. The third chapter deals with why people develop free software and who the developers of free software are. It is apparent from this chapter that the most important factors driving free software development are intrinsic motivations. The very human desires of creativity, curiosity and to be proud of oneself influence most free software contributions. There are also many for-profit businesses, who are interested in creating free software as it is a complementary good to their offering. In the fourth chapter we look at what factors one should consider when making an IT investment decision, and suggest possible phenomena one may discover during this process. While it does not cost anything to license free software, indirect and intangible costs can amount to a much greater sum than that of licensing. Free software does have some advantages in many areas because of its openness. Free software offerings resemble a perfect market much better than most proprietary software. This may lead to lower costs. As the market for free software is considerably smaller, a smaller amount of suppliers may cancel out some of this effect. The lack of compatibility and a great amount of code written in-house that would be unusable in a free software environment may be reasons why a proprietary solution would be chosen. Even in this scenario, however, there may be other parts of one’s software infrastructure that may be replaced cost-efficiently by a free software solution. If no costs of switching are involved, a free software solution is bound to be cheaper in most cases.

4