Interview with SSRN's Founder

Economic growth happens because of technology, we all know that. But technology happens first and foremost in the inventor's brain. And the inventor's brain is operating at the scientific frontier. Or is it?

We can speculate that some of the greatest inventions were invented hundreds of times, and we can document that many were invented more than once (hence the need for patent offices to document the exact timing of applications). The redundancy and overlap can be eliminated, and research energies focused more efficiently, if everyone has access to knowledge at the actual frontier instead of their perceived frontier.

Any lag between research completion and its diffusion is effectively a growth retardant. The policy challenge is to increase the speed of knowledge diffusion. Traditionally, research was published by peer-reviewed printed journals or at infrequent conferences. Printed journals remain a vital sign of quality, and we can imagine they will proliferate in a networked knowledge economy. Regardless, since the advent of the Internet, the ability to review research in rough draft or "working paper" format through an online community is creating a revolution.

One of the leading online research communities is the Social Science Research Network (SSRN), created in 1994 and summarized here on wikipedia. Michael Jensen, SSRN's co-founder and a successful financial economics professor, spoke with me today and I thought you might be interested in some of things we talked about:

TK: Why did you start SSRN?  What motivated you and what did you expect?

MJ: I was an accidental entrepreneur, actually.  Wayne Marr originally suggested we start something he called the Financial Economics Network, and I said yes, and that eventually became SSRN.  When he moved on to other projects, I'm left holding the bag (laughs).  Actually, I was motivated by my own experience as a long-time editor of journals. In fact, I co-founded the Journal of Financial Economics, and was familiar with the pluses and minuses of it. It was one of those experiences that helps you see your own blindness. In the 1970s, I began to receive quite a few papers challenging the efficient market hypothesis. The referees rejected them and I rejected them. Any one of these articles standing alone could be rejected, but I began to feel that as a package they cannot be ignored.  The authors were onto something, even if we didn't see it. Sometime around 1975, over the objection of my referees - many of them close friends - we published a special issue with a collection of those papers.  That was controversial but proved to have great value. Subsequent to that, behavioral financial economics evolved. 

In creating SSRN, I envisioned an alternative distribution vehicle.  No peer-review, of course. But papers must be part of the world scientific discourse. The only way to do this before was through academic working papers, which I had organized at the Simon School in Rochester years before, and they were clumsy, slow, inefficient. The Internet allows us to share working papers without all the cost and time of mailing printed copies. The idea of SSRN was to change the way research gets distributed and to thereby change the way research gets done.

In my own field, I was part of a very small group doing cutting-edge work in the early days of modern finance, and I noticed that elites in all fields were 2-3 years ahead of other scholars just because they knew about research that took so much time to get distributed widely. The Internet allowed everyone to see the frontier.

TK: What has been the reception and impact of SSRN?

MJ:  There are around 650,000 papers downloaded a month from SSRN, which is a conservative count based on 1.4 million actual total downloads per month, some going to multiple downloads by the same person and others to automated bots and so forth. There are half a million registered members at SSRN, and around 96,000 authors. We have had an impact.

SSRN was our baby, and we knew it should be successful. But it took a bloody fortune to get there. There were no revenues for quite a long time. Still, the VCs were interested in making a deal early on, but they would have wanted to do things we were not ready for, and I suspect SSRN would not have survived the dot-com crash if we had gone that route.

TK: Will you be upgrading or adding new features to SSRN?

MJ: We are always investing in improving SSRN. There are many different activities going on  right now, including working through a complete redesign of the site.  This will be the third time.

Citation analysis is the main tool we are working on, which will be like the ISI index.  Measures of downloads are valuable, and SSRN currently shows which papers and authors have the most downloads of all time, per year, and so on.  Measuring downloads is highly imperfect, but does say something about popularity.  Once we engineer a citation analysis system, we can assess impact.  My belief is that we can create reputation systems that help people find important work, and where papers can rise and fall in reputation as they warrant over time.

The citation analysis we are developing also provides a powerful research tool. It provides a way to click backward through a paper's references to find research that it was based on, and a way to click forward in the citations to a paper to find those papers which referenced this paper. To accomplish this we are in the process of scraping millions of references from archived papers, something like 3.7 million references so far with 1.3 million citations to SSRN papers.  I've already discovered how powerful this can be when in debugging the system I found a dozen papers directly relevant to my interests by following the links to papers that cited my paper.  I would have never known about them so easily without this technology.

We are releasing a recommender system this summer, where you will see on most abstract pages the 3 to 6 papers that were downloaded by others who downloaded this paper. This will soon be available on SSRN in beta. We now release most of our new developments to our Beta Labs before launching them in the regular site. Registered users can use these features by clicking on the Beta tab which appears at the top of every page where we have released a Beta version of some feature.

Another feature we are working on is the ability to comment on and rate papers. We are hosting a conference this fall that will use a Beta version of it, and it will probably be ready for beta testing in late 2008 or early 2009.

An important caution to innovating too quickly with a system like this is that people's reputations are at stake, and we have to be wary of people gaming the system. The overall idea is to keep improving this parallel system to peer-reviewed journals. We will get some things wrong, but will also continue to learn and to get better. 

TK:  Thanks very much, Professor Jensen.

MJ:  It's an honor. Thank you, and good luck with growthology.org.

Comments

[Lawrence C. Marsh]

SSRN is an example of why people should not fear economic growth. Growth is no longer a matter of just cutting down more trees and producing more physical products, but, more and more, is concerned with high-tech services that provide real value to our lives. SSRN along with Google, Yahoo! are just a few examples. The younger generation understands this. I recently helped out at an internet startup here in Kansas City where I was the only one printing out from my computer. All the younger people read and stored everything electronically. Objections to growth may be primarily from those who don't really comprehend where developed societies are going with growth. We still need to bring developing societies up-to-date with more physical products, but after that most world growth may well turn out to involve high-tech services and transformations of physical products rather than additions to the pile of physical goods.

[Dale Brown]

What are the characteristics of successful inventors that are most important to innovation and invention success? Is there any data on demographics or psychographics for independent inventors? (All that I have found are at http://www.inventorinsights.com/Mind_of_the_Inventor.html and http://www.inventionstatistics.com/Demographics_of_Independent_Inventors.html)

Are there any countries or States or cities that are considered to have model programs and best practices proven most effective at stimulating economic growth through independent inventors? Or institutional inventors? Is there a list of worst practices? What policies should we promote? And what policies should we discourage?

[Reid Lifset]

The rise of digital information services (such as SSRN) is not evidence that economic growth can occur without commensurate physical growth. Rigorous research on "dematerialization" shows no economy-wide decoupling of the use of physical resources from GDP, despite claims to that effect by (I think) Francis Cairncross and Alan Greenspan, in part, because of the indirect, supply chain effects of the production of digital services. See the work of Robert Ayres. Also, the substitution of "bits for atoms" does not necessarily lessen the use of material resources or of environmental impacts of the specific services that have gone digital. See the special issue of the Journal of Industrial Ecology on "E-commerce, the Internet and the Environment", volume 6, #3-4 http://www.blackwell-synergy.com/toc/jiec/6/3-4

[nayan]

SSRN is a panacea for research especially in developing countries who can't afford highly expensive journals.It is a lesson for those economist who argue that there is no such thing as free lunch.