Rusty Nail on the Information Highway:
The attempt to make the federal government more cost effective has a long and chequered history. Today, the attempt is being made to discipline the Public Service of Canada by extending user charges to more and more federal government goods and services. According to Treasury Board's experts, user charges are appropriate for only some federal goods and services. They propose six criteria to determine whether or not such charges are appropriate rather than relying on general taxation. Application of these six criteria indicate that, while appropriate for some federal goods and services, user charges are inappropriate in the case of federal information. They may prove to be a rusty nail on the up-ramp of Canada's information superhighway.
The attempt to make the Public Service of Canada more cost effective has a long and chequered history. Milestones along the way include: the Glassco Commission of 1962; Planned Programme Budgeting Systems (PPBS) of the '60s and early '70s; the Lambert Commission of 1979; and, Management By Objectives (MBO) and Zero-Based Budgeting (ZBB) of the 1980s. Why these efforts failed remains, for purposes of this article, an open question. Whatever the reasons, the pressure of deficits and debt in the 1990s has led the Government of Canada to renew its effort. This time the chosen mechanism is extension of "user charges" to a widening range of federal goods and services including information.
The rationale for this new strategy is presented in "User Charging in the Federal Government - A Background Document" published by the Treasury Board of Canada on February 7, 1997. The authors, Richard M. Bird and Thomas Tsiopoulos of the International Center for Tax Studies at the University of Toronto, justify user charges as a means to increase the efficiency of the federal work force. "But while user charges may be advocated--or attacked--as a potential additional source of revenue, their primary economic rationale is not to produce revenue." 3 They argue efficiency can be increased through user charges by making managers (workers are not mentioned) more "cost conscious" and aware of benefits indicated by the "willingness to pay" of citizen consumers. In effect, they argue user charges are a second-best solution to the "bottom-line" of the marketplace.
With respect to second-best solutions, it is important to differentiate between "efficiency" and "effectiveness." In the background study, this distinction is not made. Efficiency in economic terms refers to the ratio of outputs to inputs. To measure efficiency one must therefore be able to calculate both inputs and outputs. This is most easily done in the production of goods rather than services, especially in manufacturing, e.g. cars produced per worker.
In the case of many federal goods and most services, however, neither inputs nor outputs can be readily calculated. Accordingly, a less stringent test--cost effectiveness--is usually more appropriate. Surrogates or proxy indicators of inputs and outputs are used. For example, the "recitivism rate" per parole officer (percentage of repeat offenders) can be used as an imperfect proxy for output rather than the more difficult to measure concept of "rehabilitation" measured in human, social, and/or economic terms. Similarly, average salary per parole officer can be used as a crude surrogate for inputs rather than the more difficult to measure "opportunity cost" of relevant financial, human, information, and physical resources in alternative applications, e.g. early education rather than later incarceration.
Bird and Tsiopoulos focus much of their argument on theoretical questions associated with the "marginal costing" of federal goods and services. 4 They do not directly address the more pragmatic question of distinguishing between efficiency and effectiveness. They observe, however, that properly designed user charges often require "collection of complex and difficult-to-obtain information" 5 and the cost of obtaining such information may be so high as to make user charges inappropriate.
In my opinion, the distinction between efficiency and effectiveness can serve as a general if crude indicator of when and where federal user charges are, or are not, appropriate. Thus, if efficiency in production of any given federal good or service can be meaningfully measured, then a user charge may be appropriately applied. When efficiency can not be calculated, however, one may conclude that a user charge is not appropriate.
In this paper I will argue that federal information user charges are inappropriate and constitute a case of the right tool in the wrong hands. To do so I will first define the criteria proposed by Bird and Tsiopoulos. I will then apply these criteria to federal information user charges including their consideration within the wider context of information democracy and information infrastructure in the emerging global knowledge-based economy .
Bird and Tsiopoulos propose six criteria to determine whether or not user fees are appropriate for any given federal good or service. The criteria are used, in effect, to determine whether a good or service is more a "public" rather than a private good. Public goods are goods that can be consumed by one person without diminishing the consumption of the same good by another consumer and where exclusion of potential consumers is not feasible.
This distinction is important because "Certain important public sector activities cannot and should not be financed in this manner," i.e. through user charges. 6 The authors go on to observe that, "To some extent, the continuum between purely "public" goods and purely "private" goods is matched by a similar one between general fund financing by taxes on the one hand and user charges on the other." 7
In principle, a private good generates strictly private benefits for which a competitive market levies a price capturing all consumer benefits and thereby covering production costs plus a profit. It is the hope of making such a profit that motivates private producers. A public good, however, generates benefits extending beyond individual users, e.g. national defense, immunization against contagious diseases, clean air and water, etc. The benefits of public goods can only partially be captured by market price. Accordingly, private producers will not supply public goods and services, or will not supply them in sufficient quantity or quality to meet society's demand. Public goods must therefore be supplied by government or not at all.
For purposes of this argument I will cluster the six criteria proposed by Bird and Tsiopoulos into three sets of related pairs. The criteria are:
1. rivalness and excludability in consumption;
2. economies of scale and lumpiness in production; and,
3. externalities and socio-political objectives transcending market outcomes.
Rivalness and excludability are primary criteria for determining if a good or service is "public" in consumption. Purely public goods and services are "non-rivals" in consumption, i.e. consumption by one user does not reduce the amount available to another. If I watch a fireworks display it does not reduce the amount available to you. On the other hand, a purely private good such as an apple can be consumed only once and is then not available to another consumer. "Broadly, the more 'rival' an activity, the more desirable... it is to charge for it." 8
Purely public goods are also non-excludable, i.e. a user cannot be easily prevented from using a public good or service without paying for it. Extending the fireworks example, while I may not be willing to pay to enter the stadium, I can still watch the display from the balcony of my apartment building at no charge. Private goods, on the other hand, exhibit a high degree of excludability. Thus if I own a car I can lock the door excluding others from using it. "Excludability determines whether pricing is feasible." 9
Economies of scale and lumpiness are weaker criteria suggestive of whether or not a good or service may be more or less "public" in production. Economies of scale refers to decreasing cost per unit output, i.e. it is cheaper--per unit--to produce more than less. A major factor contributing to economies of scale is division and specialization of labour. Adam Smith discovered long ago that more pins can be produced per worker if each worker specializes in one part of production, e.g. the shaft or the head. In most industries there exists some "minimum optimum scale" of production. Below this level, cost per unit output is too high to be competitive. "With unit costs decreasing as scale increases, efficient private sector provision...can be difficult to achieve." 10
Lumpiness refers to the sheer size of initial investment required to produce a given good or service. While a pin factory can begin as a one-person firm with a minimum investment, a long distance telephone network requires an enormous initial investment before the service can be provided. Put another way, an enormous amount of initial cost must be "sunk" before production of long distance services is possible. Hence lumpiness is also called "sunkeness of cost." 11
A "lumpy" private good like a telephone network exhibits rivalness and excludability, i.e. a limited number of lines are available to subscribers or pay-per-use callers who compete for access, especially during peak hours. Public goods, on the other hand, like an immunization campaign against a contagious disease requires an enormous initial investment, but users are additive or collaborative in consumption, i.e. the more who consume, the greater the benefit to all.
In the case of both economies of scale and lumpiness, the authors of the Treasury Board background study note that: "A traditional argument for public provision of certain services in Canada has been the sheer size of the initial investment required." 12
Externalities and socio-political objectives are criteria that transcend the marketplace and are critical in determining whether or not user charges are appropriate for a given federal good or service. Externalities are the unintended benefits or costs accruing to persons other than direct consumers or producers. The classic cost example is downstream pollution from a mill or factory. In the absence of anti-pollution laws, a mill may discharge effluent into a river at no cost to its owner. Downstream, however, citizens and communities drawing water from the river must either pay to remove the pollution or suffer the health effects of it. The mill need not "internalize" these "external" costs of production into the market price of its output. A case of positive externalities is an immunization campaign against an infectious disease. Thus someone fearful of needles need not be immunized to enjoy the benefits, at low risk to him or herself, if everyone else in the population is immunized.
Many public goods are produced by government because of external effects. For example, while the private benefits of municipal roads, sewers, and clean water supplies are positive, no single individual or company can afford to provide them and/or the political implications of monopoly are unacceptable in a democratic society. Such "social overhead capital" improves the health and economic well-being of established citizens and also serves to attract new private investment. While some social overhead capital is utilitarian or functional in nature, for example, roads, there are also "amenities" like parks, recreation and cultural facilities. The role of such amenities is captured in "the amenities theory of industrial location." In this regard, a number of years ago the Arts Council of Great Britain ran with the advertising slogan "What sunshine is to Florida, theatre is to London!"
Social and political objectives refer to non-economic benefits and costs that are considered sufficiently important to justify collective public action. In the case of benefits, such goods and services are called "merit goods." In the case of costs, they are called "demerit" goods and services. There are thus times and situations in which a democratic government decides that the free market is not producing socially or politically acceptable outcomes. In such cases, government may choose to override the marketplace.
A traditional cost example is the criminal law system, which
applies the coercive powers of the State to stop activities that,
at any point in time, are viewed as harmful to society, e.g.
Prohibition. A classic benefit example is regional development.
Market outcomes may leave a given region poor and underdeveloped.
The federal government may use tax dollars to supplement local
income or services or offer incentives--favourable loans, grants,
or tax relief--to private enterprise to locate in such regions
even though the market clearly indicates this is not an economic
decision. In such cases the goods and services provided constitute
"merit goods," the most extreme case of public goods.
Such goods or services are deemed by a democratic government to be
good for society even though the market, for economic or other
reasons, is unable or unwilling to provide them.
Having defined the six criteria proposed by Treasury Board's experts, I will now apply them to federal information user charges to determine whether or not such charges are appropriate. I will not consider federal user charges in general. In many cases they are, in my opinion, an appropriate mechanism to increase the cost effectiveness of government.
Information is non-rival in consumption. In fact, it appears to have the opposite characteristic, i.e. it is collaborative or additive in consumption. In the case of research, for example, consumption of information by one researcher leads to production of new information by other researchers leading to more consumption and yet more production.
Information is, by nature, non-excludable as is evident by the need for special legislative action to create intellectual property rights, e.g. copyright. Even copyright, however, does not protect information as such but rather its fixation in material form, i.e. ideas are not protected but rather the specific form of their expression. Such laws, which vary between countries and cultures, recognize exemptions from copyright infringement. For example, rights are limited in time and are subject to specific or general exemptions from infringement. The general case is to be found in the American copyright concept of "fair use." In essence, this means a nonprofit use does not constitute an infringement. Furthermore, in the United States there is no "Crown copyright" in government information.
In Canada, however, the operative concept is "fair dealing," which limits exemptions from copyright infringement only to specified instances, e.g. it is not an infringement for an agricultural or industrial fair receiving public funding to play recorded music. Furthermore, Crown copyright in works produced by the Government of Canada and the provinces exists in Canada. More restrictive infringement provisions in the use of Canadian information together with Crown copyright results, in many cases, in Canadian researchers and citizens having a cost incentive to access relatively inexpensive U.S. information in preference to Canadian information.
To the degree that information is additive in nature then economies of scale can be expected in the production of information. For example, having constructed enormous data bases such as taxation and health statistics, the Census, Estimates, the Public Accounts, and environmental data, new information can be generated by cross-referencing existing information at a much lower cost than that of original collection. For example, socio-demographic data from the Census can be electronically correlated with environmental data to determine if there are meaningful relationships between life expectancy and selected environmental quality indicators. This, in turn, may direct public policy towards limiting the health cost impact of negative environmental factors, e.g. automobile exhaust.
In the case of many federal data sets, lumpiness is also apparent. An enormous initial investment has to be made by the federal government to develop its data systems. Furthermore, this investment must be maintained by continually updating data sets and training new generations of federal experts about the "ins and outs" of such information. For example, in responding to the reporting requirements of the Corporation and Labour Union Returning Act, company A may use one set of accounting principles while company B uses another. It takes experience and understanding on the part of federal statisticians to "standardize" these differing responses.
An example of lumpiness and the public good nature of federal information is the Census. It is unlikely that any private sector actor could afford to conduct a census of all the Canadian people. Typically private sector surveys involve one to two thousand respondents. The Census involves millions of respondents. Furthermore, without the coercive power of the Statistics Act, respondents would be unlikely to reply in meaningful numbers.
The powerful external effects of information are most evident in economic theory where it is assumed that for a perfectly competitive marketplace to exist there must be "perfect information." If buyers and sellers operate in ignorance or with unequal access to relevant information then market outcomes will be distorted. In practical terms, this is recognized by security exchange laws around the world that make "insider trading" in stocks and bonds illegal.
In the case of federal information, external benefits include, among others, "serendipity" in scientific and social research, enhanced business efficiency, and education. Serendipity or "happy unexpected discovery" is a characteristic of scientific and social research. For example, a veterinary scientist "playing" with disciplinary data in conjunction with federal environmental data may discover correlations between the rise in livestock-specific diseases and climate change. If federal information is available at little or no cost, the researcher can afford to follow a "hunch" and/or by accident come across such a correlation. If federal data is costly, the researcher may not be able to afford to play. The livestock industry may thereby suffer growing and unnecessary losses.
Business runs on information--information about markets, about competitive products and services, about new scientific breakthroughs, about demand and changing demographics of customers. All things being equal, the more information available, the better the decisions made by business. Federal information is no different. If Canada is to be competitive with other nations its information costs must be competitive. As noted above under (a) Rivalness and Excludability, Crown copyright and more limited exemptions from copyright infringement appear to make Canada less competitive relative to the United States. Federal information user charges would tend to increase this burden.
Education is another external effect of readily available information. An example of government action to increase the external benefits of information is the public library system. Society as a whole benefits the more citizens read. Public libraries are therefore a "merit good" promoting literacy and informed citizenship. To the degree federal information user charges limit access to information, they limit the educational benefits of federal information restricting them to those with the ability to pay. In welfare economic terms, this represents a "regressive" outcome, i.e. those needing the educational benefits of federal information most, that is the poor, are least able to afford access to such information.
With respect to socio-political objectives, at least two would be advanced if federal information charges were not imposed or were minimal. The most politically significant is "information democracy." If citizens are to make informed electoral choices and government is to be transparent and held accountable then the information upon which government makes its decisions needs to be available and accessible to citizens. Federal information user charges inhibit the ability to make government action transparent to public scrutiny and limits the ability of citizens to hold government accountable.
Attainment of a second socio-political objective is impeded by federal information charges: development of Canadian information infrastructure. The federal government has set the objective of linking all rural communities to the Internet by the year 2000. This constitutes the hardware half of the emerging Canadian information superhighway or Canada's post-modern information infrastructure.
The other half of the equation, however, involves "Canadian content." Canada is a net importer of information, particularly from the United States. In the case of broadcasting, the federal government, through the Canadian Radio Television and Telecommunications Commission, imposes Canadian content requirements. These requirements are considered by many to constitute a major weapon in defending Canadian cultural sovereignty.
If Canada's presence on the Net is to be significant and its
citizens are to have ready access to information about their own
country then federal information user charges represent a
potentially significant barrier to "Canadian information
sovereignty." In this regard some observes have noted that
intellectual property and information policy represent the new
"CPR" for Canadian nation-building in the emerging
global knowledge-based economy. 13
The information superhighway is not yet completed. Federal
information user charges may represent a rusty spike or nail on
the up-ramp to the Canadian section of the highway. Eliminating or
minimizing federal information user charges could be the golden
spike that allows Canada and Canadians to drive smoothly onto the
up-ramp of the emerging information future.
Accordingly to Treasury Board's experts, user charges are
appropriate for only some federal goods and services. They propose
six criteria to determine whether or not such charges are
appropriate rather than relying on general taxation. Application
of these six criteria indicate that, while appropriate for some
federal goods and services, user charges are inappropriate in the
case of federal information and may prove to be a rusty nail on
the up-ramp to Canada's information superhighway.
Bird, R.M and T. Tsiopoulos. "User Charging in the Federal Government - A Background Document." Ottawa: Treasury Board of Canada, 1997.
Drury, C.M. "Planning Programming Budgeting Guide" Ottawa: Treasury Board, 1969.
Lambert Commission. "Royal Commission on Financial Management and Accountability." Ottawa: Minister of Supply and Services, 1979.
Paquet, G. "Science and Technology Policy Under Free Trade," Technology in Society. Volume II. Pergammon Press, 1990: 221-234.
R.M. Bird and T. Tsiopoulos, "User
Charging in the Federal Government - A Background Document"
(Ottawa: Treasury Board, 1969), 7.