IV The UK market for information collected by government

Section Contents

• Introduction
• Market segmentation
• Providers of Government information
• Agents and resellers
• End users
• Current practices of the key information providers
• Primary reasons for data collection
• Underlying cost structure
• Funding
• Pricing
• Relationships with the private sector
• Integrated databases
• Market trends
• Conclusions

Introduction

This section reviews the role of the UK government and its agencies in information collection and provision. It includes:

• a segmentation of the markets;
• a summary of current practice in areas including pricing, funding and relationships with the private sector of key government information providers; and
• current market trends.

Market segmentation

Much information is collected by government departments for their own purposes. The market for information collected by government is, however, developing. This market can be broadly divided into:

• providers of information including government departments and agencies;
• agents, hosts and resellers who distribute and package data commercially, and provide products and value added services;
• end users including businesses, pressure groups, local government and individuals who are using data for a wide range of commercial, statutory and non-commercial reasons. These end users can be sub-divided into 'professionals', for whom the information is a critical and integral part of their job, and 'occasional' users whose requirements are more intermittent. The two groups have very different purchasing behaviours. In addition, there are other bodies, such as libraries and the media, who distribute information either face-to-face or for a nominal charge.

Each segment is considered briefly below with special reference to geospatial information markets.

Providers of Government information

As indicated earlier, governments collect, generate, use and disseminate many types of information. Again as we saw in Section II, much of this information falls into the following broad categories:

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

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

In the UK, government information is mainly provided by central government departments and Executive Agencies. Information supplied directly by central government departments usually concerns the activities of these departments and these cover a wide spectrum. The Department of the Environment, for example, collects land use change statistics (via a contract with Ordnance Survey). Our focus in this section is, nonetheless, primarily on Executive Agencies which collect information. Many of these Agencies are defined as government departments in their own right. We have not considered in depth the resource costs associated with collecting some types of information for government which are sometimes borne by the private sector such as, for example, Companies House returns.

We distinguish four primary reasons for information collection by these Agencies:

• it is central to government's role in policy-making and resource allocation;
• it is essential for reasons of "national interest" to support the activities of other public bodies;
• it is required for regulatory purposes to support the smooth running of a market economy or to impose standards; and
• it helps to address "market failures" and, hence, to provide comprehensive datasets for commercial users.

These reasons for information collection are used to segment and group information providers later in this section.

Agents and resellers

Many government agencies supply data both directly to end-users and to resellers including:

• agents who are simply selling products of the original data provider without providing any value added products or services; Ordnance Survey, for example, has many agents in this category and much population census information is also supplied into the market place by this route;
• "publishers" that package data and provide a range of value-added products and services; CACI and other similar commercial organisations, for example, generate and distribute a wide range of "lifestyle" indicators based on census data and Nextbase plc provides a PC based package called Autoroute for plotting journeys based on Ordnance Survey data;
• consultants who, on demand, will purchase data and provide products and value added services (for example, The Data Consultancy).

Publishers range from operations that focus on a single product to those that provide a wide range of products and services. They often operate in specialist fields and take significant commercial risks. In the market for geospatial data, most resellers are agents or consultants providing services on demand.

Government information providers and their resellers often compete with other organisations that have developed products based on their own data, or on out-of-copyright data that has been up-dated (for example, EDX).

In geospatial data markets, many of these organisations are UK-based and provide one or a small range of products which compete with only part of Ordnance Survey's portfolio of geospatial data products and services. There are, however, some signs that suppliers from other countries are now beginning to enter the UK market with products based, for example, on satellite imagery, but they are not competing directly with Ordnance Survey data.

End users

Government-produced information is used very widely. To illustrate the scope of applications, consider just three examples:

• city analysts and brokers use macroeconomic information produced by CSO to inform decisions in the financial markets;
• all local authorities, all utilities, all land registries and many professionals use digital and paper maps based on Ordnance Survey data; and
• meteorological information is available to TV watchers and radio listeners regularly each day.

There are many different users of geospatial information including:

• central government in order to inform policy decisions and for day-to-day activities, for example, the Ministry of Defence uses a wide range of meteorological, hydrographic and geographic data. We do not consider this use of geospatial information further but it is clearly a significant one;
• utilities, which have been the largest investors in Geographic Information Systems in the UK to date, in order to maintain asset registers and keep reliable records;
• local authorities for planning purposes and to manage public services including housing, education, and transport;
• emergency services including the police and fire services;
• the health sector which is increasingly using Geographic Information Systems data to plan services and operations as the new structure develops;
• government agencies such as, for example, HM Land Registry which uses Ordnance Survey mapping information to form the basis of a national map of title plans;
• academic institutions and schools for research and educational reasons;
• commercial users, for a variety of reasons including market research, assessing insurance risks, locating shops and depots, civil engineering, property development and management, and mineral exploration; and private individuals, for example, to support leisure activities.

Historically, the largest British users of geospatial data have been local authorities and the utilities In the local government sector, prices for Ordnance Survey mapping are governed by a service level agreement established between the Ordnance Survey and the Local Government Management Board. This has stimulated the local authority market. In 1993, only 20% of LAs used Ordnance Survey digital data but this had risen to 80% by 1995..

Utilities are also large users of basic scale mapping; for example, it is common to find in excess of 70000 map sheets in a Regional Electricity Company. Ordnance Survey has recently agreed a service level agreement with the utilities, but there is some evidence of price sensitivity. Some of the RECs have decided, for example, to base their Geographic Information Systems on scanned images of old maps.

The remainder of the geospatial information market:

• is highly fragmented with users having very different needs;
• is price-sensitive;
• lacks awareness of both the availability and application of geospatial data particularly at larger scales, although marketing initiatives by primary suppliers and third parties are changing this; and
• often lacks the skills and tools required to handle the data.

This mix and profile of users of geospatial data is a relevant consideration when segmenting and grouping key government information providers.

Current practices of the key information providers

In practice, there are significant differences in the pricing policy, funding and commercial activities of the government information providers. In the area of commercial activities, several agencies would seem at first sight to be operating in a way not entirely consistent with the government guidelines highlighted in Section III. Before summarising this situation, however, we review the rationale for data collection by Executive Agencies.

Primary reasons for data collection

Earlier in this section, we highlighted four primary reasons for Executive Agencies to collect data on behalf of government. These were:

• it is central to government's role in policy-making and resource allocation;
• it is essential for reasons of "national interest" to support the activities of other publicly-owned bodies;
• it is required for regulatory purposes to support the smooth running of a market economy or to impose standards; and
• it helps to address "market failures" and provide data for commercial users.

In addition, government may, on occasions, be more effective as a data collector because businesses will provide data to government which they would not provide to a commercial data collector.

In our view, the key test for the first reason is whether central government would collect and maintain the same datasets to the same quality level even if these datasets had no value and no applications outside government. Where this is true, the agencies concerned may understandably attach a relatively lower priority to dissemination of data beyond government.

Key issues underlying the second primary reason are:

• what would be the implications if the data were not collected?
• which public bodies would be affected?
• how would these public bodies be affected by the lack of information?

The third primary reason is concerned with government's regulatory role in supporting the operation of efficient and competitive markets and with standard setting, licensing and testing roles. Often, in this area, the burden of collection is placed on independent economic agents in the private sector because they have to comply with regulations (eg annual returns to Companies House).

The fourth primary reason is where there are many commercial users and applications, but where government takes a role in collection and dissemination because of perceived "external" benefits.

Classifying information providers against these primary reasons inevitably involves some judgement. Nevertheless, we think it is important to try and distinguish primary and secondary reasons for data collection because this should be a key determinant of the policy stance towards the Agency, especially on pricing and the related cost recovery issues.

On the basis of this segmentation, we have grouped the main Agencies involved currently with data collection as follows:

• the CSO as central to government's economic policy-making role;
• Ordnance Survey, the Meteorological Office, and the Hydrographic Office as essential for reasons of "national interest";
• HM Land Registry, Companies House, and Registers of Scotland which exist for regulatory purposes.

We note that this will change when the CSO merges with the Office of Population Censuses and Surveys to form the Office of National Statistics in April 1996. OPCS gets a significant proportion of its revenue from running surveys for government departments and other customers and from sales, in addition to its vote. We have excluded HMSO from our analysis; this body currently plays an important role in publishing government information but is much less involved in data collection.

On this basis, none of these organisations exists primarily to provide data for commercial users. This is not surprising given that policy has only recently changed to allow additional data collection on the back of primary collection (see Section III). As noted above, this is a matter of judgement though and, in practice, policy-makers may be influenced by factors such as the value of the information to potential commercial users.

This issue is particularly complex for Ordnance Survey and similar bodies which do not make policy-related use of their information internally and, in addition to meeting "national interest" requirements, serve a variety of end-users. These factors influence what information should be collected and, hence, the total level of resources allocated to national mapping and comparable activities.

Underlying cost structure

In addition to the primary reason for data collection, the underlying cost structure of the information providers is a key determinant of cost recovery levels.

Figure 4.1 compares, on a highly illustrative basis, the cost structures of a sample of government information providers. We distinguish between the costs of:

• data collection;
• data maintenance;
• data dissemination;
• data packaging which includes the costs of presenting data for alternative uses or to make the data more accessible; and
• value added services.

The figure presents a schematic view of the relative importance of each process in the overall cost structure of the Agency. We note that, on occasions, information providers have to bear major data conversion costs, especially when technologies move forward. The now-completed digitisation process at Ordnance Survey provides an excellent example.

Figure 4.1 shows that economic and population statistics produced by, for example, CSO and OPCS are characterised by high costs of collection and low costs of maintenance and dissemination. This is because the data are always collected anew each time, albeit often on a sample basis, and are relatively simple compared with other geospatial data.

Figure 4.1 Indicative cost characteristics of different information providers in government

Organisation	Initial data collection	Maintenance of databases	Data dissemination	Data packaging	Value added services
OPCS	XXX	X	X	X	X
CSO	XXX	X	X	X	X
OS	X	XXX	X	X	X
Met. Office	XX	XX	X	X	X
Companies House	X	X	X	X	X

In contrast, data required for regulatory purposes will tend to have low collection costs as much of the burden of collection tends to be placed on the individuals or companies supplying the information.

Once an underlying dataset exists, topographic, hydrographic and geologic data tend to have low costs of initial collection but high costs of maintenance because of the need to keep the database up-to-date. The costs of dissemination and packaging data will tend to be higher given the complexity of the data, although these costs have fallen over time with improvements in technology.

Against this background, we now consider the funding of these information providers.

Funding

Figure 4.2 summaries the financial position of the main government information providers and shows significant variation in the extent of public funding. Some are largely government financed, such as the CSO which recovers less than 4% of costs from commercial sales. In other cases, however:

• fees make a significant contribution to costs, for example OPCS, Ordnance Survey, the Meteorological Office and the Hydrographic Office; or
• costs are totally funded by revenues from the private sector including individuals and no call is made upon government, for example Companies House, Registers of Scotland and HM Land Registry.

Figure 4.2 Government information providers' expenditure, revenue, net cash costs and cost recovery

NB Not all revenues are from sales of information

£ million	Expenditure	Revenue	Net cash Costs	% Cost recovery
CSO*	49.5	1.9	47.6	4%
OPCS* (includes research revenue)	70.0	38.0	32.0	54%
Meteorological Office (includes research revenue)	141.1	57.0	84.1	40%
British Geological Survey (includes research revenue)	40.0	24.0	16.0	60%
Hydrographic Office	37.9	22.0	15.9	70%
Ordnance Survey	74.8	58.6	16.2	78%
Central Office of Information (mainly advertising expenditure for departments)	113.2	112.9	(0.3)	100%
Registers of Scotland	29.7	31.5	(1.8)	106%
HM Land Registry	197.4	235.6	(38.2)	119%
HMSO†	351.0	351.4	(0.4)	101%
Companies House	33.5	36.8	(3.1)	110%

*CSO and OPCS merged on 1 April 1996

† The total figure covers four divisions: Business Supplies, Print, Publications and Corporate Services. Most of the first and part of the second division is not concerned with information provision, reducing the relevant income and expenditure figures by about half.

Figure 4.3 shows cost recovery levels against our four primary reasons for collecting information. This shows that:

• CSO, which collects data central to government economic policy-making, is characterised by low cost recovery;
• agencies collecting data for regulatory purposes typically achieve close to full cost recovery; and
• agencies collecting data for "national interest" reasons achieve cost recoveries in a range between these extremes.

For CSO, this perhaps reflects a previous policy stance that the data collected are essential to and driven very largely by the needs of central government. More recently, there has been an objective of making data as widely available as possible to all users. Charging for data is seen to be consistent with this latter objective, given funding constraints.

The Registers of Scotland, HM Land Registry and Companies House all have the advantage that their data are by-products of regulatory functions where much of the burden of collecting the information is placed on the private individuals or organisations providing the information. This contrasts sharply with other information providers where the cost of creating the information has to be funded either directly by government or by revenue from sales. Furthermore, often those submitting information to the Agency can be charged a cost-related collection and processing fee as well as users being charged for access to the information.

Revenues reported by the information providers include both sales to the private sector and to the public sector. They also include research and consultancy fees as well as information sales. Services to government departments may also be vote funded. The cost recovery recorded by an agency will, therefore, depend on the relationship it has with its government customers. For example, if CSO was to have a contract with its main customer, HM Treasury, it would show a very high rate of cost recovery. On the same rationale, the move of the Meteorological Office on to a Trading Fund could be considered to classify it as 'high cost recovery'.

Figure 4.3 Reasons for collecting information compared with cost recovery

	Low cost recovery	Medium cost recovery	High cost recovery
Business of government	Central Statistical Office Government departments	Ordnance Survey Met. Office Hydrographic Office British Geological Survey
'National interest' and public benefit		Ordnance Survey Met. Office Hydrographic Office British Geological Survey
Commercial services		Ordnance Survey Met. Office Hydrographic Office British Geological Survey	HMSO Central Office of Information
Regulatory requirement			Companies House HM Land Registry Registers of Scotland

Pricing

The pricing policies of any Agency are obviously strongly influenced by its strategic objectives, its financial targets and, hence, its agreed funding.

For information providers, pricing policies range between dissemination cost, marginal cost, full cost recovery and market-based prices. Some interesting examples are as follows:

• ONS now sets prices for surveys to recover full costs and for other products simply to recover the additional costs associated with delivering that product, including the costs of dissemination (printing, publishing and marketing) and any services that are provided (ie there is no contribution to core collection costs).
• The British Geological Survey prices are reportedly based on willingness to pay, but it is unable to charge the fully allocated cost (£20) for its standard paper maps because of a low willingness to pay (actual prices are about £10). Even for its digital data, prices might not always cover marginal cost but British Geological Survey argue that this is justified on promotional grounds.
• Companies House and Registers of Scotland fees for statutory services are set to reflect the full cost of providing the service. Prices for non-statutory services take market values into account and may to some extent cross-subsidise statutory services.
• The Meteorological Office pricing policy is to charge other commercial information providers for the costs of dissemination, which are relatively small (£20000 in 1994/95) plus a "fair" contribution to the collection costs based on turnover or on a new charging policy agreed Europe-wide. Much Meteorological Office data is, however, available free of charge from non-UK sources because of international exchange agreements which place a heavy constraint on prices (though the recent international agreement may limit this access). Value-added services are priced on a commercial basis and make a further contribution to core costs.

For most central government departments, the general policy is to recovery additional costs associated with dealing with requests for information and not to exploit commercial opportunities. The pricing scheme usually involves some time dealing with requests at no charge, but thereafter a charge based on the marginal cost for time and copying. There is, however, significant variation in the charging schemes and a lack of transparency in how charges are set.

Price and product differentiation is a widespread practice. It is often not clear to what extent differences in prices reflect true differences in the costs of supplying the different products and users. There is, for example, a general practice through the National Environmental Research Council, which part-funds the British Geological Survey, of providing discounts to academics and discounts are also provided to local authorities. Other examples include the following:

CSO often discounts products for schools, libraries and academic institutions provided the data is not used commercially. Where possible it also seeks sponsorship of publications from businesses to enhance dissemination; and Ordnance Survey has a very sophisticated pricing structure that involves a combination of volume discounts and two-part tariffs.

The resellers we interviewed generally accepted that government information providers are seeking to exploit the commercial value of data and recognised that prices provide a useful signal to providers on the value of their data. In particular, some interviewees said that reasonable prices for data give an incentive to providers to meet the needs of users and give the users an opportunity to influence what and how information is collected. But interviewees were critical of the lack of transparency in pricing and said government providers should not "interfere" with the market place. By this they mean that providers should not:

• price differentiate between customers;
• restrict access to data; or,
• enter markets in competition with the private sector.

Relationships with the private sector

As explained in Section III, central government information providers also work closely with the private sector:

• in the production of base datasets; and
• to realise the commercial value of their data.

Examples of partnerships with the private sector in production activities include contracting out:

• local price collection for the RPI by CSO and the distribution of information through resellers;
• around 75% of HMSO turnover, including the services of graphic designers, mailing houses and printers; and
• the operation of Companies House regional offices.

In addition, Ordnance Survey has worked closely with the private sector in the digitisation of its data, through the franchising of Superplan and in most aspects of its sale of maps in paper and computer form other than bulk supply under service level agreements.

There are also examples of partnerships with the private sector focused on the development and provision of value-added services and products. Current UK government policy, however, is that it is the role of the private sector to develop value added services and that the public sector should not seek to compete with the private sector, (Note 1) where value added services are seen to include (Note 2):

• the application of technology that increases the ease and speed of access to data;
• combining datasets which can increase applications and provide benefits of one-stop shopping; and re-packaging data for different applications.

Our interviews suggest that government information providers see the skills in marketing and packaging that the private sector can bring as very important to realising the commercial value of data, particularly in immature markets. But these providers also believed that in immature markets they cannot rely solely on the private sector to develop products.

Some examples of partnerships are as follows:

• there is a CSO joint venture to develop, package and sell commercial products which includes a collaborative project with Taylor Nelson;
• Ordnance Survey is working with multiple software houses to develop new products and applications for its geospatial data;
• the British Geological Survey believe that raising awareness is essential and are therefore developing a number of joint venture arrangements with experts in marketing, packaging, publishing and distribution.

Practices do, however, vary and there is some debate about the extent to which providers should be involved in the supply of value-added services and the extent to which they use intellectual property rights to protect their position in this market. Underpinning this are the conflicting pressures on information providers, highlighted in Section III, both to raise cost recovery levels and not to compete with the value added resellers.

Companies House and the Registers of Scotland do not consider the provision of value-added services to be their responsibility.

• Companies House has stated that it will maximise access to basic data (providing only minimal added value services), while leaving the business information industry to add further value through interpretation and combination with other information sources.
• Registers of Scotland does not see the need for exploiting value added opportunities itself in order to achieve its financial objectives. It has an objective of examining the feasibility of providing marketable geographic information services not currently available, but is developing these opportunities with joint ventures. In these partnerships the Registers of Scotland see its role as data provider and not as provider of value added services.

Other government information providers are involved in value added services, but recognise the potential competition issues and have taken steps not to distort competition. Some examples are:

• CSO, which has stated it would not attempt to restrict access to data for its own commercial gain and would not enter into exclusive arrangements with private sector partners, but argues a need for its involvement in value added services because of the immaturity of the market and so the need to promote the availability and potential uses of its data;
• HMSO, which has separated its Copyright Licensing Division from the Trading Division to minimise the risks of restricting licensing for its own commercial gain;

HMSO believes that there might be incentives for information providers to use intellectual property rights to restrict access to data for commercial reasons. It accepts that this could have adverse implications for competition in downstream markets, but believes the problem is easily addressed by the objectives that the providers are set.

These examples contrast with:

• British Geological Survey's position not to sell data that would allow third parties to develop competing databases and products, and to keep control of the market; it believes that this policy is essential if it is to meet its financial objectives and has established joint venture arrangements on an exclusive basis;
• OPCS's position of providing products in competition with its own agents (which CSO may also operate in order to avoid data being exclusively in the hands of third parties operating on a commercial basis).

Integrated databases

Improvements in technology have made it much easier to add value by integrating datasets. This requires consistency of the underlying databases. The Central Statistical Office, which has a central position in the government statistical service, has an aim to "maximise the use and value of statistics by prompting the possibilities for integration of data by improving the co-ordination of government statistics and helping to develop common standards, classifications and definitions." CSO see this co-ordinating role to be essential as individual departments will be resistant to changes in standards that interfere with time series data.

One reason for the merger of CSO with the Office of Population Censuses and Surveys and with the statistics branch of the former Department of Employment to create the Office of National Statistics (ONS) has been to encourage consistency and to create a larger single database.

Ordnance Survey has the task of fostering a National Geospatial Database to be achieved through the integration of national topographic data and other geospatial data held by government departments and agencies. This will certainly require collaboration between Ordnance Survey and ONS since the UK is unusual because of the very wide disparity of geographical units used for the collection and aggregation of official statistics: this problem will have to be addressed if the production of integrated, country-wide geospatial data is to be achieved. This organisational challenge is probably greater than the technical challenge.

We note there are various existing initiatives between agencies and government departments to promote data integration. For example:

• Ordnance Survey have completed, with the Department of Environment, a trial on collecting information on land use (NLUSS).
• There is a pilot scheme for the National Land Information System (NLIS) which provides online access to property-based information from sources such as HM Land Registry, the Valuation Office, Local Authorities and Ordnance Survey. A similar scheme in Scotland, led by Registers of Scotland, is developing rapidly.
• HM Land Registry and Registers of Scotland both hold title plan information in map form, based on Ordnance Survey mapping information.

An issue raised frequently during our interview programme was the meaning of the term "integrated database". The conclusion interviewees reached was that this should not imply one large physical national geospatial database, in one place, owned by one organisation, but:

• at a minimum, would involve co-ordination between government information providers to ensure that databases could be integrated (this would include setting and monitoring adherence to standards); and
• might involve a central management function that would provide one point to deal with user requests, co-ordinate with data providers and handle charges.

A key finding from the interviews was that there is expected to be increasing demand for data integration, but that the efforts of government information providers should be focused on meeting demand for specific applications rather than on schemes to draw together all possible providers to support one large, totally integrated database.

Market trends

Developments in computer technology and the digitisation of data have already had a major impact on the information market. Some examples of improvements in services include:

• Ordnance Survey has now completed digitisation of its 230000 different basic scale maps and has launched some 'spin off' products;
• Companies House has an online service;
• HM Land Registry plan to complete computerisation of the totality of the existing land register by 1998. An online access to the land register has been available since January 1995. The Registers of Scotland plans for an integrated online service involving both text and maps showing property boundaries are well-advanced;
• the Hydrographic Office is increasingly providing products and services in digital form and aims to have all its charts (in excess of 3000) available by 1997; some CSO data are available through online services; and
• the British Geological Survey has a digital catalogue containing over 700 datasets of which about 200 are digital.

Digitisation has led to:

• a reduction in the cost of handling information;
• increases in quality, and new applications that were previously not possible or too costly;
• an increase in the range of "off the shelf" products;
• the development of easy-to-use products; and globalisation of the market for information and value-added services.

The result has been an increase in demand for data and associated products, particularly in the market for geospatial data, where the complexity of the data has in the past been a major constraint. Despite this, however, the market for geospatial data is generally regarded as immature. The reasons for this are said to include:

• an exaggeration in the past of the benefits from Geographical Information Systems (GIS) which technology could not then deliver;
• a lack of awareness of the availability and possible applications of geospatial data, and the associated benefits;
• the cost of specialist staff needed to handle geospatial data; and
• the initial cost of investing in a Geographic Information System and perceived complexities associated with implementation.

There is, however, considered to be significant scope for growth given:

• more realistic views on the benefits of GIS; initiatives by providers of data and products to develop the market;
• the greater availability of trained staff; a reduction in the cost of GIS in some sectors where use can be made of PC-based systems; and
• a better understanding of the complexities of GIS and management of the associated risks.

From our interview programme, key trends in geospatial information in the future are expected to be:

• a downsizing of Geographical Information System technology onto the desktop with a corresponding increase in ease of use;
• increasing demand for a wider range of geospatial data and novel data applications as new users in "non-traditional" sectors are attracted into the market;
• an increase in competition in the supply of geospatial information itself as new technology reduces the cost of data collection and, thus, the barriers to entry; examples include the increased use of raster imagery in various forms, such as scanning aerial photographs to up-date old information;
• the development of standards for the collection and storing of data; some standards have already been established but there is a debate about whether providers will adhere to these given their complexity;
• greater interest in legal liability issues as geospatial data is used more in business decision-making; as yet, however, the size and implications of these issues have not been tested.

Interviewees also noted:

• competition and regulatory issues will be raised as the information market globalises, partly as a result of the Internet; and
• there are already attempts by the private sector to monopolise segments of the information market by purchasing exclusive rights.

These trends indicate that the underlying business economics of geospatial information provision are changing rapidly which, in turn, will have implications for the most effective method of ensuring and financing base data provision in future.

Conclusions

Government-produced information, especially geospatial information, is widely consumed by many different groups of users. The market is diverse and includes some segments which are not price-sensitive, for example 'professional' users of economic data. Key existing users of geospatial information include the local authorities, the military and the utilities. Much of the remainder of this market is highly fragmented and price sensitive. This user profile is important when grouping key government information providers.

There are four primary reasons for Executive Agencies to collect information on behalf of government. These are:

• it is central to government's role in policy-making and resource allocation (e.g. CSO);
• it is essential for reasons of "national interest" and to support the activities of other public bodies; (e.g. Ordnance Survey, the Meteorological Office, the Hydrographic Office);
• it is required for regulatory purposes to support the smooth running of a market economy or to impose standards (e.g. HM Land Registry, Companies House and Registers of Scotland); and
• it helps to address "market failures" and provide data for commercial users.

Cost recovery levels are lowest for the CSO and highest for the information providers required for regulatory purposes. The key conclusions we note on the practices of government information providers are that:

• significant differences exist in the basis for establishing prices and funding levels, but some logic exists in each case once the reasons for data collection and the underlying cost structure are understood;
• there is some lack of transparency in the pricing of data and value added services; and
• there are questions of consistency between policy and practice with regard to participation in value added service and product markets;
• it is genuinely difficult to measure the extent of externalities though these certainly exist

Key issues identified by both government information providers and users in our interviews include:

• the conflict between arguments for more open access to information and the benefits that the market place can provide;
• how to control access to data in future given current technological developments; and
• the increasing importance of integrating geospatial datasets, a development which should, as far as possible, be market- and application-led.

The market for geospatial data is generally regarded as immature and there is considered to be significant scope for future growth. Technological trends are changing the underlying business economics of information provision which will have major implications for future provision of data.

Notes

1. Treasury, The Fees and Charges Guide.
2. Held Tradeable Information, Guidelines for Government Departments in Dealing with the Private Sector, Department of Trade and Industry.