Governing Magazine/October 1994

TECHNOLOGY COLUMN

THE HIDDEN DATA PRICING WAR

By Jerry Mechling

Jerry Mechling is director of the Program on Strategic Computing and
Telecommunications in the Public Sector at Harvard University's Ken-
nedy School of Government. To reach him, call 617-495-3036 or send e-
mail to jerrym@ksgrsch.harvard.edu.

Talk in technology circles today focuses on the Information Highway,
client/server computing and reengineering. Meanwhile, almost hidden
from view, a war is being waged that may well have a greater real-
world impact on government and society in the next 15 years. I'm
referring to the conflict over what prices should be charged for
government-held information, such as property records, court schedules
and patent applications.
 On one side are the "marginal cost" forces, largely made up of
information services companies, public-access advocates and federal
agencies guided by OMB regulations. On the other side are the "fair
value" forces, assembled largely from state and local agencies and the
burgeoning GIS movement. At issue are several key questions: Who will
pay for information? How much will they pay? And what balance should
we strike in society between public and private services?
 The view of the marginal cost camp is that prices should not exceed
the total cost of the last unit produced. If they do, the marginal
cost argument goes, society loses because the gap between price and
cost squeezes out some net-beneficial uses. Further, since marginal
costs approach zero for computer-based data (electronic copies being
MUCH cheaper to produce than paper ones), marginal cost pricing
encourages wide access to government information, diminishing the gap
between the haves and the have-nots.
 The fair value camp believes that those who benefit from a service
should be the ones to pay for it. If the general public benefits--as
from the creation of a society where safety, privacy and equality are
protected--then the general public should pay, typically through
taxation. But if only individual users benefit, then those individual
users should pay. In mixed cases, such as education and licensing,
costs should be shared so that private users pay to the extent that
the benefits are private.
 The fair value argument is that if private users "ride free," taxes
will be too high and society will lose access to truly public services
it otherwise could afford. In addition, the fair value camp believes
that public organizations will be managed more efficiently when real
customers--not just appropriations committees--are a major source of
revenues.
 To get an idea of how these arguments might play out in the real
world, let's look at typical department of motor vehicle data, such as
owner histories of automobiles. Assume that the marginal cost for an
owner history report would be less than a dollar, while fair (or
market) value would be more like $12 (the price actually charged by
ServiceOntario, an arm of the Ontario government). Twelve dollars
covers most of the costs of data collection and maintenance, in
addition to the marginal cost of data selection and printing. If we
set the price at $1, just about everyone could get the reports, and it
also would be cheaper for resellers to enter the market, keeping
distribution prices competitive. But if we only collect $1, we must
fund the entire DMV data collection effort through taxes. In today's
political and economic climate, that makes it hard to fund new or
expanded programs, such as more frequent testing or training for
elderly drivers.
 But since most requesters of DMV data want to make money from it--the
high-volume users are insurance and auto companies--the fair value
camp argues for the $12 price. (Even at $12, ServiceOntario has found
a huge demand for car history and other reports, especially driver
histories.) If we need to accommodate poor and/or "freedom of
information act" requesters (such as the media), we could offer
"information vouchers" and/or special FOIA pricing--perhaps marginal
cost or lower. We would call on oversight agencies to review our
prices and budgets to determine whether "public value" subsidies are
justified; if we considered it a "public value" to improve the market
for used cars, for example, we could subsidize car ownership histories
and perhaps even offer them for free.
 On balance, what's the right price for government data? While the
issues are complex and no scheme is without problems, I'm for starting
with fair market value, then allowing a significant subsidy for FOIA
and other "public value" services, such as the information-locator
services that are needed on the Information Highway to make it easy to
find government and nonprofit services. We need user fees to force
government to pay more attention to customers and to serve as a
funding source for essential innovation.
 Unfortunately, while the issues are a long way from being fully
resolved, the marginal cost people seem to have a lead, one that may
prove impossible for the fair value camp to overcome.

----------------------------------------------------------------------

Copyright 1994, Congressional Quarterly, Inc. Reproduction in any form
without the written permission of the publisher is prohibited.
Governing, City & State and Governing.com are registered trademarks of
Congressional Quarterly, Inc.
http://governing.com