[8318] in Commercialization & Privatization of the Internet
Socialism, A-130 and the information poor
daemon@ATHENA.MIT.EDU (James Love)
Sat Nov 13 16:44:22 1993
Date: Sat, 13 Nov 1993 16:46:45 -0500 (EST)
From: James Love <love@essential.org>
To: com-priv@psi.com
Note on socialism, A-130 and the information poor
James Love
November 13, 1993
Individuals on com-priv have raised an issue regarding an "older"
culture of socialism, which holds that access to information
should be free, and a "newer" culture that wants to sell
information. I would like to offer one example of how this
debate can be framed, in terms of a particular data set that I
have worked on in the past, to provide a somewhat different
perspective.
Before going to work for the Taxpayer Assets Project (a "huge"
organization that currently has two full time employees), I
worked as senior economist for a subsidiary of the Frank Russell
Company, the nation's premier pension fund consultant. Our
clients were non-profit entities, the tax exempt pension funds of
IBM, AT&T, DEC, Shell and other large corporations. One of my
early jobs was to see if we could develop a CD-ROM product that
would provide supply and demand analysis of commercial real
estate markets. We wanted to build a product that covered the
entire country. Our motivation for developing the project was
the observation that our clients had lost billions of dollars in
bad real estate investments in the late 1980s, which we
attributed to bad underwriting practices.
It turns out that there is a great deal of economic data which
can be used to model the demand for real estate. Employment data
published by BLS is particularly useful. On the supply side,
however, the most useful statistics were those that showed the
square feet of building permits that had been issued. For the
national market, the only source of this data is McGraw Hill's
F.W. Dodge Company. Dodge actually collects the data directly
from local governments, at their own expense. In many
jurisdictions they pay local permit officials to supply them with
the data.
On the other hand, the Bureau of the Census conducts a monthly
survey of building permits, and publishes the data for a very
reasonable price. The Census data is of limited use, however,
since they only ask for the "value" of permits, rather than the
"square feet." No one cares much about the value statistics, and
everyone wants the square feet statistics.
We approached F.W. Dodge and asked about the cost of obtaining
the square feet statistics at the county level (about 3,100
reporting units). Dodge wanted about $200,000. But, that would
only give us the right to use the data inhouse. In order to
"redisseminate" the data, we would have to have pay royalties.
Dodge wanted to see our business plan, and monitor our revenues.
It was also the case that our product would have been a
competitor to an expensive paper product that Dodge sold to the
same target audience.
I went back to Census, and asked them if they would consider
modifying their largely useless survey of the value of permits,
to include a question about square feet. It turned out that this
would cost Census about $10,000, a cost that we might have been
willing to pay. Census would not do this, however, and the
reason was instructive. They said that it would be improper for
Census "compete" against the Dodge monopoly. Apparently this had
been a conscious policy decision that dated back to the 1960's.
Census officials knew that the square foot statistic was
important, and in fact they were buying some square foot data
from Dodge.
I was incredulous, since it was clear that Census knew that it
was providing a "brain damaged" product, and it was doing it on
purpose, to protect a private company's monopoly on an important
statistic. The consequences of this policy were serious.
Because of the extremely limited access to permit statistics,
academic research on supply and demand in real estate markets was
practically non-existent. Investors, including our clients, were
regularly making investments involving hundreds of millions of
dollars with limited and misleading data provided by brokers and
money managers. Research at Goldman Sachs and by us indicated
that many spectacularly bad investments could have been avoided,
with even the most basic supply and demand analysis.
At one point I brought together more than 30 leading real estate
academics and money managers, including senior officials from
AEW, Salomon Brothers, The Boston Company, The RREEF Funds, T.
Rowe Price, John Hancock, JMB, CIGNA, Aetna, Coldwell Banker,
Grubb & Ellis, Prudential, Jones Lang Wooten, and among others,
to write Congress and Census asking that they modify their
permits survey to include the square foot statistic. Census
replied that it could not, because of the Reagan Administration's
OMB Circular A-130, which discouraged agencies from providing
information products which were provided by the private sector.
The upshot of all this is that we had to abandon our CD-ROM
product, and McGraw Hill's F.W. Dodge subsidiary still has a
monopoly on the square foot statistic. The product that we
wanted to provide would have been expensive, perhaps $20,000 per
disk. It would have been sold to very large investors and money
managers, who had national real estate portfolios. At the time,
the Frank Russell Company (my employer) had annual revenues of
more than $100 million, but we were "too poor" to buy the data.
I should also note that we had no interest in "owning" the permit
statistics. Our product was based upon our analysis of the data,
not any control over the data. We were selling our model of how
real estate markets worked, not our monopoly on facts about
activity in the real estate market.
The F.W. Dodge monopoly was not a state sponsored one. Legally,
anyone could have entered the market. But the high sunk costs of
data collection created insurmountable entry barriers. The only
solution to the problem was for Census to modify its survey, and
provide a public source of the data.
When people talk about the "information industry," they should
not think of data vendors who sell public records as the only
players. Companies that use government information to provide
value added products and services are important members of the
information industry too. Many of these are profit making
companies, whose products and services help firms make better
decisions, and increase national productivity.
The goal of federal information policy should be to decide what
types of data the government needs to collect, and how best to
disseminate the data to the people who will use the information.
Projects like the NSF funded dissemination of EDGAR filings are
designed to benefit the whole society, by giving the broadest
possible access to important information. The issue of how this
affects companies like McGraw Hill, Mead Data Central, Dialog or
West Publishing is not nearly as important as is the issue of how
data users are affected. There will certainly be cases when a
large reliance upon private data vendors is appropriate, but
there are also cases when it is appropriate for the government to
take a more active role. This has little to do with stylized
debates about socialism and capitalism, and lots to do with the
issue of making our profit and non-profit sectors as productive
as possible.
Of course, there are also many non-economic reasons for the
government to ensure that the public has low cost access to
government information, particularly in areas where government
accountability is involved, where access to information is
necessary to participate in democratic decision making, and where
broad dissemination of information is necessary to promote the
arts and sciences.
jamie
----------------------------------------------------------------------
James Love, Taxpayer Assets Project; internet: love@essential.org
P.O. Box 19367, Washington, DC 20036; v. 202/387-8030; f. 202/234-5176
12 Church Road, Ardmore, PA 19003; v. 215/658-0880; f. 215/649-4066
----------------------------------------------------------------------
