Discussion: View Thread

Melissa,

I believe that Compustat also uses the GICS (Global Industry Classification Standard) from Morgan Stanley. Code 3520 picks up Pharma, Biotech and Life science. Code 352010 is just biotech defined as "Companies primarily engaged in the research, development, manufacturing and/or marketing of products based on genetic analysis and genetic engineering.  Includes companies specializing in protein-based therapeutics to treat human diseases."  I don't remember if Compustat goes to that level.  I would also be interested in what others have found on this issue.

Hope that helps.

Stuart

Stuart Napshin, PhD Candidate, Drexel Univ

----- Original Message -----

From: "Melissa A. Schilling" <mschilli@STERN.NYU.EDU>

Date: Thursday, July 12, 2007 12:16 pm

Subject: Biotech firms

> Hi all,
>
>
>
> I'm looking for a way to search for the number of firms (private
> and public,
> or just public) that are/were primarily engaged in biotechnology,
> going back
> to 1985. Since biotech doesn't have its own SIC or NAICS code I
> cannot use
> Compustat or an EDGAR search. I have looked at Bioscan, but I
> don't think
> that will work for my purposes because a) it includes a lot of
> diversifiedchemical companies that would be hard to characterize
> as "primarily
> biotech", and b) it drops companies from the database if the
> companies no
> longer exist, so its numbers for earlier years are understated and
> changedepending on when you pull the data. I have also looked at
> Corptech but its
> selection process seems a bit ad hoc. I think a text search of the
> "BusinessDescription" field of the SEC filings might work to
> identify the
> publicly-held companies. Disclosure Online (through Lexis Nexis)
> allows such
> a search, but is only available for the most recent year (at least
> in our
> subscription). I think Compact Disclosure allowed this kind of
> search and
> was available for multiple years, but the product has been
> discontinued. I
> also checked with the BLS and the Census Bureau, but no luck.
>
>
>
> Anyone else already gone through this process and have some
> lessons to
> share? Thanks for your help!
>
>
>
> -Melissa Schilling
>
>

Hi,

I went through this process for my dissertation.  In addition to not having SIC or NAICS codes, the biotech industry has another problem: different surveys pick different companies according to their specialization (i.e. pharmaceutical versus agricultural biotechnology). For example, Bioscan is strongly biased towards pharmaceutical biotechnology companies. There is a good article by Ronald Wirtz in Fedgazette, which you can find online. This article also talks about different surveys about the industry conducted by consulting firms and research organizations.

Other than Bioscan, there is a web based data company called Recap.com (or rDNA.com). They differentiate between large public firms and small biotech firms (they also have a third category called device firms). A difference from Bioscan is that they also include companies that do not exist anymore and they provide merger, acquisition or bankruptcy information for at least some of the companies that dissolved. They also report whether the company is public or private. You can also use the previous versions of Bioscan to see the differences over time as each Bioscan will only tell you the number of active companies in that particular year. But the company does not keep or sell the old versions, therefore you have to find them from other sources.

One other source is paper based biotechnology directories. There are different ones.  "Biotechnology Guide USA" has 1988, 1991, 1995, and 1999 versions. "The Biotechnology Directory" starts from 1985 and is repeated in different years.

I looked at the number of firms that were both in Recap.com and Bioscan and found out that only about 40 percent of the companies are in both databases. The other directories were similar in that sense. The numbers you find will be different depending on which sub-industries you focus or which directories you include. I guess the ideal thing to do would be to join different databases, which will be very time consuming. During my research I come to agree with Ronald Wirtz that there is a strong need to map out the biotechnology industry. I joined Bioscan and Recap.com databases and eliminated the large diversified firms from my sample. I can tell you the number of firms between 1992 and 2003 in my sample if that would be something useful to you.
Good luck,
Remzi Gozubuyuk


Melissa A. Schilling wrote:

Hi all,

 

I'm looking for a way to search for the number of firms (private and public, or just public) that are/were primarily engaged in biotechnology, going back to 1985. Since biotech doesn't have its own SIC or NAICS code I cannot use Compustat or an EDGAR search. I have looked at Bioscan, but I don't think that will work for my purposes because a) it includes a lot of diversified chemical companies that would be hard to characterize as "primarily biotech", and b) it drops companies from the database if the companies no longer exist, so its numbers for earlier years are understated and change depending on when you pull the data. I have also looked at Corptech but its selection process seems a bit ad hoc. I think a text search of the "Business Description" field of the SEC filings might work to identify the publicly-held companies. Disclosure Online (through Lexis Nexis) allows such a search, but is only available for the most recent year (at least in our subscription). I think Compact Disclosure allowed this kind of search and was available for multiple years, but the product has been discontinued. I also checked with the BLS and the Census Bureau, but no luck.

 

Anyone else already gone through this process and have some lessons to share? Thanks for your help!

 

-<st1:personname w:st="on">Melissa Schilling</st1:personname>