II Key economic characteristics of information

Section Contents

• Introduction
• Economic characteristics of information
• Public good issues
• Externalities
• Natural monopoly
• National interest arguments
• Conclusions

Introduction

Information is not only a commodity which can be bought and sold; it plays a critical role in all modern economies in facilitating the smooth operation of markets. Markets simply can not operate efficiently and competitively without good, reliable information. Information availability and provision is also important to much of the business of government.

Across the world, governments have a central role in information markets. Usually they collect and disseminate the following types of information:

• economic, social, and environmental statistics;
• geologic and topographic surveys;
• business registration data and accounting information;
• land registration data;
• meteorological and hydrographic data.

In addition, governments also collect and provide information on the business of government itself, on policies and on the institutions of government.

In this section, therefore, by way of background, we:

• summarise the economic characteristics of information and highlight some additional special characteristics of geospatial data;
• define and explain what economists call "public goods" and "externalities";
• comment on the relationship between the provision of "public goods" and natural monopolies;
• illustrate how these concepts are related to "national interest" arguments for information provision; and
• comment on the implications for the role of government in information provision.

Economic characteristics of information

Key economic characteristics of information and data are summarised below and then developed in the remainder of this section.

Information-based products and services have some special features. They are often associated with relatively high collection costs and relatively low dissemination costs. Consumption of information by one user does not reduce the availability of the same information to others. Information can often be used for a wide variety of applications and may have very different values in its alternative uses. Hence, the structure of demand can be complex as users' needs are similar but distinct. Furthermore, the value of information to a user may depend on the skills of that user.

Users may be able to derive additional value from large datasets which are available on a consistent basis. Different datasets may, in addition, be brought together to create additional value for users. The value of different types of information will change over time and, hence, there is a need to invest in maintaining "current" datasets.

These economic characteristics of information-based products mean that some types of information provision are associated with what economists call "public goods" and also with "externalities". These concepts are explained further below.

Geospatial data share these economic characteristics but additional issues are also important. It is estimated that between 60% and 80% of all data held by government departments in the UK can be classified as geospatial (footnote 1), where this is defined as any data that has associated with it some geographical referencing, including referencing to the National Grid, postcodes, latitude/longitude or defined areas such as parliamentary constituencies.

It is perhaps most helpful to consider geospatial information as a spectrum ranging from:

• highly detailed topographical information where the geospatial element is fundamental to the nature and value of the information, such as maps; through to
• data such as national and regional macro-economic statistics, company annual reports or information on parliamentary business where the extent of geographical disaggregation is usually minimal.

Geographic Information Systems (GIS) are the computer systems that handle complex geospatial information. They allow users to collect, store, manipulate, link together, analyse, update, and present geospatial data. These systems are beginning to influence the business economics underpinning geospatial information markets as well as opening up new market opportunities.

Public good issues

The textbook definition of what economists term a "public good" is as follows:

"The defining characteristic of a public good is that consumption of it by one individual does not actually or potentially reduce the amount available to be consumed by another individual." (H Gravelle and R Rees, Microeconomics.)

Several types of public good can be distinguished. These distinctions are important because they may influence implicitly policy discussion. A "pure" public good has very specific characteristics:

• the marginal cost of providing an additional unit is zero;
• use by one individual does not reduce availability to others ("non-rivalry"); and
• individuals cannot be excluded from using the good or service ("non-excludibility").

There are also "weaker" (or optional) forms of public goods where:

• the marginal cost of supplying an additional unit is low rather than zero; and
• individuals can be excluded from using the good or service.

Common examples of public goods include television broadcasts and national defence arrangements. Broadcasts are a "weak", or optional, public good in that an individual can choose to consume any amount, or none, of the output produced and that he or she can be excluded from consumption (e.g. by encoding a TV signal). Defence, on the other hand, is a more pure public good in that all inhabitants "consume" what is provided.

A geospatial dataset is an optional public good because individuals can be excluded from using it. Nevertheless, in practical terms:

• new electronic technology is making it more difficult to control access; and
• some individuals are able to access the information that they need free of charge through "browsing" in libraries, indexes or view screens.

Some privately provided "information" goods and services also share these properties of an optional public good. Examples include the programmes underlying software products and operating systems, the content of newspapers, and hotel and airline reservation systems.

Public goods are often associated with "market failures" in a competitive economy. Problems arise from difficulties in making goods non-excludable (eg stopping illegal copying of intellectual property). In addition, users have no incentive to reveal their true marginal valuation of the good; they realise it is worth suppliers reducing their price to cover short run marginal costs or dissemination costs once major investment in the public good has been made. These "market failures" would typically lead to under-provision of a public good in a free market. They begin to provide the economic justification for some form of government involvement in the market either in terms of economic regulation or ownership of information providers.

There are also related reasons why in a free market information providers may find it difficult to extract revenues from users. Key reasons are:

• highly dispersed user benefits; and
• high costs of collecting revenue compared with the income so generated.

So, information products and services are usually "weak" or "optional" public goods in that:

• the marginal cost of supply to an additional person is low, falling close to zero with new technology;
• use does not deplete the availability to other users; and
• there are potential issues relating to free-riding and to the extraction of revenues.

It should be noted, however, that information provided to the Armed Forces, including topographic and weather data, is an input to the delivery of a "pure" public good, in that individuals cannot be excluded from access to public defence.

Arguments about what economists call "externalities" are closely related to this concept of a public good. We turn to "externalities" next.

Externalities

"Externalities" arise where:

• production of a good by one "agent" imposes costs on and/or delivers benefits to other producers or consumers; or
• consumption by one individual imposes costs on and/or delivers benefits to other consumers or users.

It will be seen, therefore, that a "pure" public good is a special form of externality; provision for one group of users will benefit other potential users because it is not possible to exclude them.

Pollution of water, air and by noise are classic examples of "external" costs; refuse collection, education and public health are classic examples of "external" benefits. It is an important feature of markets where "externalities" are present that output levels resulting from free market provision will not be optimal.

In markets with "external" benefits, there is often under-provision and excessive prices because private providers take no account of the wider social benefits when setting prices since no financial benefit will accrue to them. This is illustrated further in Section VI of this report. A range of policies can be used to address the "market failures" which result, including:

• regulation of the markets;
• government provision of the product or service; use of subsidies (or taxes); or
• specific licensing obligations (eg on emission levels).

As we have seen, provision of geospatial information is not a "pure" public good but, nevertheless, there are important "external" benefits associated with:

• ensuring consistency in the collection of data (producer externalities);
• providing users access to the same data (network externalities); and
• promoting efficiency of decision-making (consumer externalities).

The nature of these externalities is explained below.

Inconsistency between different datasets or between data relating to different geographical areas can raise the costs of using data and limit the range of applications. Co-ordination between the emergency services can, for example, be important when major accidents occur and costs can be saved on infrastructure projects when all key users share the same underlying dataset. Collecting data on a consistent basis can, therefore, help to raise the value of the individual datasets. In part, the problem is a practical one associated with the process of agreeing and publishing standards, and defining standards for data specification, especially given the development of historic datasets on different bases. Common standards are necessary, but not sufficient, for consistent datasets. If common standards are not agreed, there is a danger of fragmentation.

In the US, for example, there are Federal Information Processing Standards whose use is mandatory for US federal bodies; geospatial information providers are, however, also allowed to use non-standard formats provided that they can supply the information in the standard form for whoever in government opts to take it in that form.

Many markets are characterised by similar externalities, for example the compatibility of electrical appliances and fittings, which are addressed by industry standards. Whether the geospatial data problem can be addressed satisfactorily in this way will depend on:

• the complexity of the issues involved with agreeing standards;
• the costs associated with complying with standards including the conversion of "legacy" datasets; and
• the distribution of costs and benefits between producers.

the complexity of the issues involved with agreeing standards; the costs associated with complying with standards including the conversion of "legacy" datasets; and the distribution of costs and benefits between producers.

The "external" benefits associated with promoting more informed decision-making and greater accountability of public bodies have long been appreciated and are, indeed, recognised in the Open Government White Paper. The argument is that, by providing citizens with information, debate and decisions will be better-informed to the general benefit of society. Examples include:

• planning inquiries where those making representations might be in a better position to make their case with access to information; and
• pollution control where access to information allows pressure groups to be more effective in influencing government policy and in monitoring activities.

"External" benefits such as these are, of course, difficult to value or more generally to measure. Government policy has been to encourage the use of frameworks to analyse "external" or non-marketed outputs. (footnote 2) There are three broad groups of techniques available for analysing "external" benefits; these are:

• cost-benefit analysis which entails the direct valuation of costs and benefits and is the best approach when feasible;
• a matrix or impact statement approach which involves listing and quantifying effects without aggregating them; and
• weighting and scoring of benefits to combine rankings or quantities into a single indicator.

These techniques can also be used in combination.

Natural monopoly

The provision of public goods is often associated with what economists call natural monopoly. A natural monopoly exists when a single supplier can serve the market most efficiently because unit costs are still falling at the required total output level. The presence of a natural monopoly creates further reasons for government involvement in the market either as owner of the provider or to regulate the provider.

The collection and maintenance of detailed geospatial datasets for a specified region is a natural monopoly; the costs of collecting and maintaining the information required to "map" an area are typically very high in proportion to the additional costs of disseminating this information to extra customers. Issues concerned with natural monopoly are discussed in more detail in Section VI.

National interest arguments

In the context of general debate on information collection and provision by government, arguments are often framed in terms of the national interest. These arguments are usually linked closely to the economic arguments on "public goods" and "externalities". Sometimes additional arguments, less concerned with economics, are also made for collection and provision of information by government. They include:

• protection of life and limb; promotion of democracy;
• protection of the rights of individuals; support for minority groups in a population; equity;
• the need to maintain confidentiality of data collected; and
• a basic need to meet government functions.

These "national interest" arguments can be very important in determining the overall policy stance to information providers and, in particular, the approach taken to cost recovery.

Walter Smith, a former Director General of Ordnance Survey, emphasised the importance of geospatial information to government's primary responsibilities. Following John Stuart Mill, he argued that government has a mandatory responsibility to ensure order and an optional responsibility to encourage material and other progress. (footnote 3) To ensure order, government has to safeguard national security and protect individual rights to life, property and liberty. Smith then argued that the national security objective required the collection of geospatial information that was up-to-date, homogenous and continuous in spatial extent (ie covers the national territory). He went on to note that, whilst much of the information could in principle be collected by the private sector, the specification of the product and quality control were an irreducible state responsibility. This philosophical approach has obvious logic but the practical conclusions Smith drew may be out-of-date. The successful exploitation of major geospatial databases, where multiple products are spun off from a core database whose components are up-dated at different moments in time from multiple input sources each with different characteristics, strongly indicates the need for a single entity with overall responsibility for and involvement in that maintenance. This is buttressed by the legal liability issues which may arise from use of the database, particularly since these will often relate to 'legacy' information i.e. which has been inherited from previous operations.

Conclusions

Information, and in particular geospatial information, has special economic characteristics which mean that governments are often involved with its collection and provision. Information is a form of "public good" and is often associated with "external" benefits. Because public goods and externalities are, in turn, often associated with "market failure" there will be a role for government in these markets.

Furthermore, provision of public goods often includes elements of a natural monopoly which governments usually seek to own or regulate. National interest arguments, which are not purely concerned with economics, often provide an additional stimulus for government involvement.

The role that government takes in these markets is likely to depend on a number of factors including:

• the underlying cost structure;
• the size of any external benefits, assuming that they can be valued or quantified;
• the reasons for which the data is being collected; and
• the risks associated with leaving provision to the free market.

In practice, as the next two sections illustrate, the characteristics of information provided by the government and public bodies, and the extent of government funding of this information provision, vary significantly. A key question to address is whether the observed differences in levels of public funding logically follow the characteristics of the information.

Footnotes

1. What is the British National Geospatial Database?", Bryan Nanson, Neil Smith and Alison Davey. Paper presented to the Association for Geographic Information Conference November 1995 and available on the Ordnance Survey web site

2. Appraisal in Central Government, HMSO April 1991.

3. P Smith, "National Mapping: a case for government responsibility", Conference of Commonwealth Survey Officers Conference, 1979.

Ordnance Survey web site