We will now turn to a discussion of "the economic value of information." This topic was suggested during our open discussion period by Ruth A. Zietlow who graduated in 1990 from the University of Wisconsin-Madison with a master's degree in library science.
Since her graduation, she has taken additional courses in the University of Minnesota's Hubert Humphrey School of Public Affairs and the University of St. Thomas' master's-level program in software design and development. She has worked as a reference librarian at the University of St. Thomas in St. Paul, Minnesota since 1991 and in 1992 became responsible for library services to graduate students taking courses off-campus.
Please welcome Ruth as the guest editor of the Kellogg CRISTAL-ED Mail List and take part in a discussion of our new topic on "The Economic Value of Information."
Perhaps by exploring the pricing philosophy of fee-based services, and looking at economies of scale for different types (school, public, special, academic) of libraries and their services, we may be able to see helpful patterns. I think the descriptive statistics taught in MLS programs is a good start to the measurement problem, but it certainly needs to be expanded when it comes to budgetary justifications. What work is being done in this field at the master's and Ph.D. levels? What's possible? As a profession, should we be hiring economists or creating our own?
>One area I would appreciate hearing about is the economic value of information.
I am not an LS professional -- I make my living in knowledge management but I am subject to the same economics as face libraries -- the user must see value in our product or we are out of business!
As I see it, the economic value of information is in its contribution to the knowledge that enables us to transform resources into wealth. This value can be either a potential value or a realized value. To stay in business we need to meet one of two conditions: Our customers must believe that we can cost effectively retain information in a state of high potential value -- or -- our customers must believe that wealth has come from information we have supplied and will continue to do so in the future. If we can accomplish both of these conditions then we increases our ability to prosper. (I understand that "wealth" can take many forms, I use it in a generic form.)
To have realizable potential value, information must be associated with an agent that, when triggered, can transform the information into the knowledge that will enable the realization of wealth for the user. Most often this transformation takes place as "learning." Timing is important. The learning needs happen at a time when the user is in a position make use of the knowledge. Learn too late and the user will miss the boat, learn too early and the changing conditions may render the knowledge useless.
>Perhaps by exploring the pricing philosophy of fee-based services
Based on what I said above, I feel the best path is for libraries to become "knowledge libraries" that provide focused, timely learning opportunities. So, you may ask, isn't that what we are doing now? Maybe it is and maybe it isn't. But the important question is, "do our customers think that is what we do." If not, they're not going to pay to keep us in business.
>I think the descriptive statistics taught in MLS programs is a good start to the measurement problem, but it certainly needs to be expanded when it comes to budgetary justifications.
Not being a product of an MLS program, I don't know what is being taught. But I don't see the answer coming in the form of conventional metrics. The key is becoming integrated into the day-to-day processes and activities of the community we serve. we must be conceived as a key factor in the creation of wealth. Corporate libraries, by their vary nature, only survive if they do just that, and in doing so meet both the conditions I stated in the beginning. Now it is time for all of us, libraries and the rest of the knowledge management professions, to do the same.
>As a profession, should we be hiring economists or creating our own?
Is the question one for economists, who seemingly have only been able to confuse the general population, or one for those with more pragmatic and real-world outlooks.
The views expressed are mine, influenced by many and taught by some. I just ask that you don't blame the expression of them on anyone but me.
Essentially, we have some different issues under this umbrella term, which are partly covered at least by the division of "value" into "exchange value" and "value-in-use." "Exchange value" is what economics is about -- unfortunately, we only know the exchange value of information for that which is sold -- it is the price that people are prepared to pay. However, we could characterize the "exchange value" of library and information services as being the total value of the support costs (including people, materials, space, etc.) that organizations or communities are prepared to pay to keep the service in being -- in that sense, any librarian or information worker should be able to calculate their "economic" contribution to the firm, other organization or community.
What most people want from the "economics of information," however, is some means of measuring the economic value of specific services -- such as the acquisition of a document for a user, or the discovery of the answer to a reference enquiry. Unfortunately, this is where "value-in-use" comes in, since the value of such service is only known by its contribution to the value of whatever the user of the information does, i.e., the value it adds to economic activity. There is no "objective" value-in-use for any given piece of service or any given document -- what is useful to one person may actually add negative value to another if the latter person fails to understand the information and has to spend more time seeking clarification.
Discovering the value-in-use of information services, then, requires the librarian/information worker to collect data from users about the economic impact of service. King and Co. did this for the U.S. Department of Energy database, but there are many circumstances in which their methodology does not work because not everyone thinks in terms of project deadlines, deliverables, unit costs of activity, etc., which was the case with the project engineers, etc., covered in the Energy study. My personal belief is that, for many cases, value-in-use is an intractable problem -- not likely to have an easy solution for most libraries/information services.
As to its place in the curriculum -- I'd have thought it was dealt with to some extent in all LIS schools/departments. How much attention it deserves is debateable, but I would not have thought it worth an entire semester-long course - a couple of lectures and an assignment, perhaps. In the UK, of course, all master's degrees require a dissertation and it is often the case that economic issues feature in these.
I do feel that it is essential that information science graduate students who intend to be employed in a management position take a class in economics/cost-benefit analysis.
In the U.S., most MBA students must take a class in economics, and several law schools and nonprofit management graduate programs require a class in economics. A few weeks ago, a close friend of mine, who is a research physician, lamented about not having taken a class in economics during his medical training. It seems that many of the problems in drug research are not about discovering new drugs to handle new diseases, but rather about finding more cost-effective alternative drugs and medical treatments. Last year a former student from a nonprofit management graduate program told me that she felt the economics class she took with me was the most useful class in the program. During a management strategy meeting at the American Red Cross, the department managers considered hiring an economist to do a cost-benefit analysis of a new program they were considering implementing. Instead, my former student mentioned that she could do the analysis, having learned it in my class. A good class in economics provides the tools for quantitative analysis of management and policy problems that are not always taught in management and administration classes.
In my class in the School of Information Science and Policy at the university at Albany I focus on teaching cost-benefit analysis, or how to use data and economic analysis to make more economically efficient management and policy decisions. Students examine several economic problems in library management, public policy, and information resource management; including interlibrary loan management, copyright policy, estimating demand for photocopying and other library services, user fees, estimating "opportunity" cost, etc. The students learn the spreadsheet program Microsoft Excel in order to set up cost-benefit matrixes, calculate cost functions, and do basic regression and correlation analyses.
Anyone interested in what else is taught on the economics of information at US L/IS Schools should look at Terry Weech's papers in last year's ASIS proceedings (October 1994) and the ENSSIB Economics of Information conference proceedings (May 1995).
What I found most interesting in the cost-benefit analysis class I had was the theory behind the modeling of the regression equation. We could model an equation on the philosophy that library services have a positive impact on the local economy and cultural life. It is the capturing of data that we can use to prove or disprove the theory that's problematic.
Bruce Kingma's class looks very focused and useful. I know I've run into trouble when I tried to find ways to capture data that would reflect larger, more ambiguous studies.
I'm sorry I don't have my notes with me, but I remember reading about a study that tried to capture the value of our public parks by measuring the average time spent by Americans in the parks, the number of Americans that visited the parks within one year, and the average value of time as captured in labor studies.
I don't think this "time" approach would be advantageous for libraries, especially if the assumption that computerized databases speed up research and time spent in the library. However, I think there are many approaches we need to explore further as a profession.
I know Michel Menou is doing a large study in this area trying to capture the value of information centers in developing countries. One advantage of this study is that Menou and his colleagues can isolate the value of these centers because they did not exist in many rural areas prior to this study.
Comments are welcome.
Some of the best studies are on transportation problems; i.e., whether to take the train, bus, jet, or car. The value of time is a critical component here. There are also interesting studies on the opportunity cost of sleep and volunteer labor at nonprofit organizations.
>I don't think this "time" approach would be advantageous for libraries, especially if the assumption that computerized databases speed up research and time spent in the library. However, I think there are many approaches we need to explore further as a profession.
I think there is a reluctance in the IS community to accept this method of valuing library services. Malcolm Getz wrote two excellent articles (College and Research Libraries, 1980s) on valuing the new electronic catalogue at Vanderbilt. The subsequent letter to the editor of C&RL criticized his results based on valuing patron time.
I have always been a bit confused by the sceptics of this methodology. It is a well accepted technique in economics. In addition, the goal of showing that patrons' time has value is to prove there are significant economic benefits from library services.
As for your above observation, this is the classic diamonds and water problem from economics. Diamonds are more expensive than water, but less valuable. The "price" of library services is the value of time spent consuming these services. When the technology results in a decrease in this "price" or opportunity cost, it does not mean that the services are less valuable. Value is defined by consumer surplus -- the difference between the amount someone is willing to pay and what they actually pay. A study on the value of library services would use this information on the price decrease to estimate the demand for the services and then estimate the consumer surplus from them.
>I know Michel Menou is doing a large study in this area trying to capture the value of information centers in developing countries. One advantage of this study is that Menou and his colleagues can isolate the value of these centers because they did not exist in many rural areas prior to this study.
Sounds interesting. There are other interesting studies by Jaffe (American Economic Review, 198?) on the value of university R&D in a community, Chressanthis (ENSSIB conf. proceedings 1995) on the value of university libraries, and the studies by Griffiths and King on special libraries.
The focus of my class is on cost-benefit analysis of management problems, so I do not cover these studies in my class. Most of my students are MLS students looking to become better managers of information services.
It is generally accepted that libraries contribute to cultural life. Therefore, libraries are considered to be cultural institutions. As such, libraries get financial support.
In a broad sense, libraries contribute also to social, political and economic life. That is why libraries should not be subsumed under the category of cultural institutions only.
Society should become aware of the multifunctionality of libraries. I think that for librarians it might become beneficial to propagate this viewpoint by receiving additional funds.
>>I don't think this "time" approach would be advantageous for libraries, especially if the assumption that computerized databases speed up research and time spent in the library. However, I think there are many approaches we need to explore further as a profession.
>I think there is a reluctance in the IS community to accept this method of valuing library services. Malcolm Getz wrote two excellent articles (College and Research Libraries, 1980s) on valuing the new electronic catalogue at Vanderbilt. The subsequent letter to the editor of C&RL criticized his results based on valuing patron time.
>I have always been a bit confused by the sceptics of this methodology. It is a well accepted technique in economics. In addition, the goal of showing that patrons' time has value is to prove there are significant economic benefits from library services.
I'm an economist, so I have no reluctance about using the value of time as a measurement. But a measurement of what? Time is an input factor to the types of services you are discussing (transportation, library usage). There is a tradeoff between using one's time driving to a national park and working more, or watching TV, etc. Comparing the time cost can be quite helpful in establishing relative value when the time cost is the primary factor that differs across choices, or when it is possible to simultaneously control for other factors. Hence, as you say, it has been helpful in studying choice of transportation mode.
What could we learn from measuring the time spent using information services? We can't learn much about the total value users place on information services. Suppose we compare the time spent at the library with time spent recreating. The theory tells us that (holding other things equal for discussion) the consumer will equate the marginal value of the service (library vs. golf) to the time cost. So we will learn the approximate value of the last hour of library use (say, approximately the foregone wages). But this is a lower bound on the total value of library services to the user: I may value most of my library time much more highly than the incremental hour. (How much is one hour of library time worth to you? Suppose you'd already had 10 hours this month -- how much is 1 additional hour worth? Probably a lot less.) I expect that this method would give very low estimates of the value of library services.
(I know, the above method assumes a lot of rationality and would be subject to substantial measurement uncertainty. I'm just describing the best case, to show that it isn't that good for this purpose.)
What about how the time spent using information services has changed over time? Someone raised the concern that as technology improves time spent decreases, even though value increases. That's right: improved technology is increasing the efficiency of info services usage, and thus lower the time cost. Lowering the cost is a good thing, and measurement must get the sign right! But this highlights one of the most serious problems I see in using time valuation for measurement: the output (the good stuff) is changing, in addition to the time cost of creating or obtaining the output. Suppose I spend an hour using the library today. What I get out will not just be more (because the cost of obtaining info has fallen) than I would have 5 years ago, but it will also be different. Since the good is changing, we can't infer much of anything about how people value it based on how much time they spend obtaining it.
Example:
(a) I go to the library to find all of the articles authored by an obscure Hungarian mathematician. Today it takes me five minutes. Five years ago it took me one hour. The time savings measures the increased value of library services to me.
(b) I go and find th Hungarian's Web page, and learn about his hobbies, see his picture, read his C.V., and find out which other mathematicians are influencing him. I spend the same hour I would have spent five years ago just to get his publication list. Time input hasn't changed: does that mean the value of library services hasn't changed? Of course not!
 Home |
 Discussion |