Statistics Division Home
Fundamental Principles
Country Practices
Reference Materials and Links
Sign in

Restructuring the Dutch Central Bureau of Statistics (CBS)


Paper for FSU-seminar, Luxemburg, March 1993

by W.F.M. de Vries, deputy Director-General of Statistics Netherlands


1. The present organisational structure of the Dutch CBS goes back to the mid-seventies. Among other things, statistical departments were then regrouped in four directorates:

-Economic statistics (12 departments)
-Social statistics (9 departments)
-Methods and development (4 departments, incl. automation)
-Office services (4 departments)

This type of organisation model, primarily subject-matter oriented, is not uncommon for national statistical offices. In addition, the organisation of the CBS, like that of many other government agencies, is rather pyramidlike and hierarchical, with quite a few layers.

2. In the same period the number of staff was substantially increased (from about 1500 to well over 2000), while the Bureau was also split up between Voorburg (near The Hague) and Heerlen (a mining town in the south of the Netherlands, about 220 kms from The Hague). The reason for this split was that the government had decided to close down all Dutch coal mines and had to provide alternative employment. Apart from the CBS some other government agencies (e.g. the civil servants' pension fund) were (partially) transferred to the same area.

3. The increase in staff was used to expand statistical work in some areas and to move into some new areas. Part of the increase was obviously lost on inefficiencies, due to having two locations.

4. During the seventies and early eighties new increases in staff were allowed. In 1982 the nominal number of posts in the CBS reached the level of about 3500. In fact, the expansion "on paper" went so fast that we actually did not succeed in recruiting people to the extent we were financially able to. Then came the turning point.

5. From 1982 to 1992 the CBS (and most other government bodies) were forced to cut down their budgets and staff-numbers. The latest round of cuts (1992-1993) brings down the number of staff to just under 2500, about the same level we had in the late seventies. Altogether, cuts thus added up to 30% in ten years time. Protests to the effect that statistics should be spared did not help. The pressure on government spending was too strong.

6. Over the years we had been able, to a large extent, to maintain our statistical output, despite decreasing resources. The most important factor in this respect has been automation. We have been able to invest massively in computers, and now have one of the largest and most advanced computer networks in the country: about 2500 microcomputers are linked up in local area networks which are connected -through what we call a backbone-system- with each other in one "supernetwork", which also includes the mainframes and several minicomputers, printers etc. High-capacity datalines provide
links between the two offices. Moreover, we have developed efficient software (such as Blaise and CBS-View) for several forms of data collection, data editing and data dissemination.

7. Regardless of budgetary decreases, demand for statistics increased, the European Community being an important factor in this respect. The point was reached where the increase in efficiency could no longer compensate budgetary cuts. The latest round of cuts actually forced us to reduce our statistical programme. This created tension. The Central Commission for Statistics, which decides on our programme of work, had great difficulties to establish priorities. Several specific users (e.g. ministries) were dissatisfied.

8. This situation made us realise that something had to be done to remain in business. Therefore, we launched Operation TEMPO in May 1992. TEMPO stands for:

- INDEPENDENT (in Dutch: Onafhankelijk)

Operation TEMPO

9. The preparation and execution of TEMPO was undertaken in close cooperation with Moret, Ernst & Young, Management Consultants. Their contribution has been most valuable and constructive. To analyse our situation we used a kind of simple model, in order to look at ourselves from different perspectives. The centre of the model is our programme (what we produce), because the ultimate purpose of the whole exercise is, of course, to maximise our output, in terms of client-satisfaction. The perspectives we used were internal, external, formal and informal.

11. Basically, the analysis made clear that our problems were the following:

External factors:
- increasing demand for statistics, decreasing budget
- therefore: need to improve efficiency
- tension between European and national requirements
- pressure to reduce the response burden
- need to improve presentation and PR
- need to improve timeliness
- low public profile and not enough self-esteem

Internal factors:
- too much compartmentalisation and specialisation
- therefore: fragmented image
- too many layers and poor internal communication
- unclear responsibilities
- complex internal rules and procedures
- culture is too inward-looking
- bureaucratic management attitudes.

12. In April 1992, two directors of the CBS spent a week at Statistics Canada to be informed about certain management and organisation features and philosophies developed there. Taking this experience into account, as well as information about management and organisation developments in some other national statistical offices, and after careful consideration and soul-searching, we produced a mission statement, which was sent to all CBS-staff. It was introduced with some ceremony in a heads of departments-meeting. In addition, a videofilm was made, in which the operation was explained at large. This film was shown to all staff in meetings of about 100 people each, led by a director. At the same time, a time schedule and action plan were published in the internal personnel bulletin. Annexe 1 to this paper is the text of the mission statement.
Further action, Phase 1

13. Phase 1 of the action plan involved four working parties, each chaired by a director, and composed of six members from various sectors and levels of the organisation. They had to produce a provisional report in the beginning of August 1992.

14. The Phase 1 working parties covered the following areas: timeliness, data collection, presentation and organisation & management. In the following paragraphs the focus for each area is briefly described.

15. The TEMPO working party on timeliness has mainly tried to analyse the (internal and external) causes of delays. One of their main findings was almost trivial, but nevertheless fascinating: timeliness is greatly enhanced if some kind of external deadline (such as an agreed and published release date, or a deadline of an external printer) exists or is introduced. Another conclusion was that commitment is an important factor: if a group of people is made responsible for the entire statistical process, from data collection to data dissemination, timeliness usually improves.

16. The TEMPO working party on data collection has looked into two topics: first how to reduce or alleviate the response burden, and second how to organise data collection internally in the best possible way. As to the first topic the working party was backed up by a study among businesses about questions such as "do we ask the right questions", "can data easily be extracted from business records", "do we ask the same question twice (or more than twice)", "do we ask for data at the most appropriate point in time" etc. As to the second topic the working party has mainly considered the internal structuring of data collection and processing, including (certain forms of) centralisation of these operations.

17. The TEMPO working party on dissemination has studied our public relations policies in general, as well as ways and means to improve the informations flows to users, e.g. in the form of electronic dissemination, but also a more modern and uniform "house style" for printed publications.

18. The TEMPO working party on organisation and management, finally, has explored a fairly wide range of topics, such as the introduction of "contract-management", decentralisation of budgets, an improved planning process, management styles, internal communication, personnel policies in general (including possible simplifications of the present system of grades, mobility etc.), aspects of "corporate culture" and last but not
least: alternative organisation charts.

19. A steering committee has been (and still is) taking care of the internal information process about the TEMPO-operation. In the personnel bulletin questions from staff members are answered and some information on the progress made by working parties etc. is given. In principle, everyone who sends in an idea or question is contacted by a member of one of the working parties.

Further action, Phase 2

20. After the Phase 1 reports were considered, four new working parties
were formed, essentially with the following terms of reference:

I: design a structure of the CBS for the year 2000;
II: design a structure of the CBS for 1993, taking into account the perspectives for a longer period;
III: design a system of contract-management for the CBS;
IV: design a system of management-development in order to improve management of the CBS.

21. All things done, Phase 2 should lead to concrete proposals about "restructuring the CBS". These proposals would then be discussed at all levels, including the statutory labour-management consultation process.

22. Part of the TEMPO-operation is also that the personnel should be closely involved in all stages. Therefore, all staff can give their opinions on the topics raised by TEMPO. An "electronic in-tray" was opened for reactions. The number of reactions received is now (end of April 1993) well over 600. They vary from minor complaints (why can't we have our own fax) to complete organisation charts and extensive proposals on a wide range of statistical and organisational topics.

23. On the basis of the reports from the four Phase 2 working parties, the board has made a number of basic decisions (February 1993). They were presented to the CBS in information sessions, involving all personnel. Apart from that, workshops for in-depth discussion were organised for a 10% sample of the personnel. Relevant points of criticism were taken into account in the final reorganisation proposal, which has meanwhile been approved by the personnel committee (a body of elected representatives of the personnel, from different labour unions). The decisions made can be summarized as follows.

24. The promotion of EDI (Electronic Data Interchange) will be one of our main goals for the medium-term future (CBS 2000). The technology is there or will be with us shortly and we must take an active part in the promotion of its use for future data collection.

25. Our attitude towards respondents must change drastically. In 2000 the CBS should bother respondents as little as possible. Before a new survey is developped, the CBS tries to find alternative ways to collect the data, e.g. from registrations. If it is nevertheless necessary to have a survey, its purposes will be set out as clearly as possible. Relevant results are supplied back to respondents. Small enterprises are surveyed as little as possible. To collect small enterprise data the CBS must try to make do with existing data, e.g. tax data.

26. When data are collected, the CBS must speak the language of the respondent. Questionnaires containing CBS-concepts not used by respondents themselves must disappear. Ideally, data should be extracted straight from accounting systems of companies, which tend to be computerised more and more. Consequently, a great deal of basic information will get to the CBS in a clean, electronic format. During data collection, computer-assisted checks are used, as has already been the case for some time in household surveys. In cooperation with software-suppliers and accounting firms, modules are developped to transform normal accounting data into the desired CBS-format. If data cannot be extracted from accounting systems directly, the CBS supplies user-friendly, electronic questionnaires with built-in checks. This has already been done for Intrastat, the new system of collecting intra-EC international trade data. In the near future, a number of other EDI-pilots will be organised. In order to improve the commuincation between the CBS and respondents, account managers will be appointed. In principle, each firm will deal with the account managers as sole CBS-representative.

27. On the way to 2000, the basic organisational structure of the CBS will consist of seven divisions: 4 statistical production divisions and 3 supporting divisions. The first of these supporting divisions is a division for input-coordination. It is responsible for such matters as the business register, interviewing households, developments of EDI, accountmanagement etc. The second supporting division is for output-coordination. It will develop a central database for overall information supply, and is also responsible for integration of information (incl. national accounts), for general publications and for marketing and public relations activities. The third supporting division is responsible for infrastructure (incl. general computer services) and office services.

28. The four divisions for statistical production will be organised on the basis of a combination of principles, of which subject matter coherence and relations with a certain market segment (both in terms of users of information and of suppliers of basic information) are the most important. These four divisions can briefly be characterised as follows:

I. Agriculture, manufacturing, environment, energy and technology
II. Trade, transport and commercial services
III. Budget sector and well-being of the population
IV. Demography, labour, income and consumption.

29. Within each statistical division there will be 8-10 statistical departments, as well as a management bureau. The internal organisation of statistical departments will be such, that teams are responsible for the complete statistical process. Teams must be of such size and flexibility that work-peaks can be dealt with easier than in the present situation. The number of organisation layers will be reduced from 7-8 now to 4-5 in the near future. Obviously, this re-grouping will also result in a reduction of overhead.

30. As far as management is concerned, divisions will be far more autonomous than the present departments (or even the present directorates). The instrument to achieve and control this will be a system of so-called contract management. Management responsibilities will be decentralised to a fairly large extent.

31. Management contracts (to be concluded between the Director-General and division managers) will consist of the following main components:

- agreed output (statistics and publications to be produced, publications to be sold etc.)
- quality aspects of the output, in terms of timeliness, accuracy, etc.
- other statistical targets to be achieved (e.g. reduction of response burden)
- social targets, such as improvement of work conditions, training etc.
- resources, in terms of staff and budgets
- services the manager agrees to buy from supporting divisions
- rules and procedures to be observed (incl. coordination rules)

32. Top management of the CBS will consist of the Director-General, two policy directors and a fairly small concern-staff, which will support top management in corporate matters such as development of statistical programmes, statistical coordination, public and international relations, legal affairs etc., as well as in internal management areas such as control, personnel policies (e.g. management development) and automation. Top management and division managers together will act as "the managing board" of the CBS.


33. In conclusion, operation TEMPO will lead to a drastic restructuring of the CBS. So far, reactions of the staff have been very favourable. The same goes for the political bosses of the CBS: reactions from our parent Ministry of Economic Affairs on this initiative of the CBS-board have been very positive indeed. It is clear that the implementation process of all the envisaged changes will take some time. We plan to finish most of the process before the end of 1993. This puts things under some pressure, but we think it is better to do things rather quickly than to get entangled in a lengthy and tedious process. The momentum is there; we really make TEMPO.





To all CBS-employees

The board of directors has reconsidered the position of the CBS in changing circumstances. Our conclusion is that the CBS must change too, in order to fulfill its functions properly. The general direction of this envisaged change is reflected in this brochure. It is our vision for the future. The implementation of this vision is a challenge for us all.

Voorburg/Heerlen, May 1992




Reconsideration of the position of the CBS is necessary. On the one hand we face budgetary cuts, on the other an increased demand for statistics. To be able to supply what society needs we have to look at our internal structure and corporate culture. There is a need for change. This is only possible by a common effort.

External factors

Government finance will be tight, maybe even tighter in the future, but the "information society" requires more and timelier statistics to monitor existing and new policy areas. The European Community also demands an increasing amount of statistical information. Competition is growing stronger, which may endanger the centralisation of statistics. The response burden gets more attention than in the past. The public image of the CBS is grey and fragmented. We are seen as reliable, but slow and dull.

Internal factors

Our internal organisation is compartmentalized and has many layers. It is not always clear who is responsible. Relations between departments are not very businesslike. The communication process is often erratic. The division of labour between our two offices is not efficient. Not rarely are we are more focussed on our internal problems than on the interests of our clients.

Mission of the CBS

This confronts us with challenges. What do we want to be in a changing world? The board thinks that we must remain the national, impartial and reliable institute for statistics, but that we have to be more timely and more client-oriented.


To accomplish this mission we need to implement operation TEMPO: timeliness, efficiency, a higher public profile, professionalism and independence must be its pillars.


TEMPO has consequences for the structure and culture of the CBS, for its presentation vis--vis respondents and users, and for its management philosophy.


It is necessary that all these topics are thoroughly discussed by everyone in the organization. In the very near future we will present a time-table and a plan for the structuring of the process of change we have to face. Phase 1 is the discussion of this manifesto. Phase 2 is making TEMPO operational; this will involve working parties etc. Phase 3 is the implementation, which must take place in 1993.






The world changes rapidly and we are experiencing this very clearly. Therefore the board has deemed reconsideration of the position of the CBS necessary. This manifesto presents the results of our thoughts on the matter.

In the past ten years we have experienced many changes. First of all in our budget. In the beginning of the eighties we still expanded our staff and statistical programme. Since then there have been many cuts. The last one made a rather drastic reduction of our programme of work necessary. It is probable that government finance will remain tight for some time to come.

At the same time there is an adverse development. The demand for statistics is still increasing, both at a national and an international level. One of the characteristics of the information society we live in is that there is more emphasis ons timeliness. Society appreciates our reliability and thoroughness, but increasingly criticizes our slowness.

The response burden we are contributing to gets increasing exposure. The willingness to respond decreases. Respondents are tired and more critical than in the past.

Internally there is dissatisfaction about how things are organized. We are strongly compartmentalized and have many layers. Moreover we are often more focussed on internal relations and problems than on the needs of our clients. In essence our internal organisation has not been changed for decades, while society changed considerably.

This manifesto gives a medium-term vision on the position and duties of the CBS. It is meant as the starting point of a discussion about these points, and also sets out the framework for such a discussion. First of all, changes in the environment of the CBS are briefly analysed. Then some points are made about our internal structure. Subsequently we look at the challenges this presents in terms of the mission of the CBS. Choices are made as to the policies needed to realise our mission. Finally, the planning and structure for further steps are indicated.

Two interrelated dimensions are at stake. First the notion that the CBS will have to change. Second that all CBS-employees must be aware of this and must be prepared to make an effort.

External factors






In our modern information society there is an increasing demand for statistics. The CBS tries to fulfill this demand in the best possible way. In the seventies and eighties our programme of work was drastically expanded, partly due to automation. This process of expansion was stopped in recent years. In fact, the latest round of budgetary cuts made a reduction of our programme inevitable. Several groups of users are affected by this and will try to find alternative sources of information. There are different new policy areas for which new statistics are needed. Competition of commercial and other providers of statistics is a real threat. This may endanger the great good of a centralised system of official statistics.

It is not realistic to expect an increase of our budget. For some time to come the pressure on government funds and on the number of cicil servants will continue to exist. Therefore, we face decreasing resources in a period of increasing demands.

A special user of statistics is "Europe". The EC request more statistics than ever and it can force Member States to provide them, by means of legislation. So far, EC demands could generally be met in the framework of our current programme, but there is an increasing tension between European and national requirements. This means that the latitude for a strictly national dimension in our statistical programme is decreasing.

The response burden is an issue of increasing importance. Non-response is becoming a serious problem, both in household and business surveys. Society expects less form-filling from a government that pledges to "take a step back". A specific point of concern is that businesses tend to be irritated when they receive questionnaires from different CBS departments, and have contacts with numerous different officials. Often they note that some questions are asked more than once, sometimes in slightly different forms.

Many see the CBS as dull. The way we present our information is not very sexy. Also, it is not always easy for outsiders to get a simple answer to a simple question. We are rather good at making subtle distinctions in definitions, which may be very confusing for many users. Due to the way we have organised our dissemination, users often have to consult different subject matter departments. Both by our way of data capture and our methods of dissemination we have created a fragmented image of the CBS vis--vis the public.

Apart from more and sexier statistical information users expect increased timeliness. This gets more important as developments in society pick up speed. The general idea is that the CBS may be thorough, but also slow. This idea is not unjustified; often there is a considerable time lag between the reporting period and the publication date. Although several short term indicators are produced rather timely, many other CBS data are rather outdated. Moreover, it is a plain matter of fact that the timeliness of our statistics has hardly been improved over the years, despite enormous automation efforts.

Internal factors





The CBS organisation if largely based on subject matter specialisation. This leads to compartmentalization and coordination problems. It also creates a fragmented image. As long as very specific markets are concerned this is not serious, but markets tend to change and many problems do not simply coincide with market segments.

We have a many-layered organisation. Chains of communication are therefore long. The distance between management and the work floor is great. That is one of the reasons information often does not come through. Long chains of initialing lead to lenghty approval processes, to indifference, lack of commitment and lack of sense of responsibility.

The divison of labour between our two offices is not efficient. We have very many small organisation units, often highly specialised. The boundaries between these tiny kingdoms are often remarkably sharp. Sometimes, we talk more about competence matters than about statistical problems. Responsibilities are often unclear. Internal rules and regulations are complex. Communication about policies is blurred and erratic. Relations between departments are not always businesslike. High performance is not properly rewarded, nor is failure punished.

Generally speaking our culture is often more internally directed than client-oriented.

Apart from these aspects the CBS organisation is not geared to modern ideas about management. There is a general feeling that our efficiency can be improved and that work satisfaction could be enhanced.

Because we are a commonwealth of kingdoms our corporate spirit is not well developed. We have a low public profile and do not excel in self- esteem.

The mission of the CBS




We have to rethink our mission, in view of all these challenges. What is our mission? What do we want to be? We operate in a growth market, but have decreasing resources. Because our independence is too valuable, we cannot become "commercial". If we want to create opportunities for growth, we must therefore improve our image within the civil service system. We also need a better image in the eyes of data suppliers and users of statistics. These are two sides of the same coin: when society values our work more, politics will follow suit.

Our mission is and remains as follows:

The CBS is the national, impartial institute that provides reliable and relevant statistical information.

Advantages of a national and therefore central statistical office are efficiency and coordination. The concentration of certain skills and instruments (e.g. a central business register) within one organisation provide economies of scale, enhances professionalism and prevents duplication of effort. It creates comparability of statistics in terms of different subject matter areas and over time. In view of this principle of centralisation the CBS work programme covers all areas and aspects on which statistical information is required.

The CBS is impartial. It is not associated with any specific political, social or economic interest group, but serves society as a whole, on the basis of a work programme that is approved by a "statistical parliament" (the Central Commission for Statistics). There is no political interference with statistical results.

CBS-statistics are not only compiled impartially, but also with great care. Reliability is and must remain our middle name. We use state of the art methodology and contribute to new methodological developments. Analysis is an important part of our work. Also, we want to be modern in the way we conduct our business. Finally, it is part of our professional duties to reduce our burden on society as much as possible.

If we want our statistics to play a role in public debate and decision making, in short if we want to be relevant, we must be as timely as possible. This means that time lags have to be shortened, not only for short term indicators, but also for our annual statistics.







To fulfill our mission in the future, choices have to be made. We must improve our image and sharpen our profile. This is only sensible if we can boast our qualities. To strengthen our qualities we must make TEMPO. In brief: be more timely, more efficient, more modern, more professional and as independent as before.

The timeliness of CBS-statistics must be improved considerably by reducing production and publication time lags.

Our efficiency must be increased by introducing a better management structure.

Our dissemination methods must be modernized. In particular statistics of general interest must be presented in more attractive ways.

Data collection must be organized in a more professional manner by taking better account of the interests of respondents, instead of just our own specializations and structure.

The CBS must remain independent, even in a period of declining resources.





It is obvious that operation TEMPO implies many changes.

We need increased awareness that the CBS serves society, not the other way around. This applies both to suppliers of information and to users. Our internal structure and culture must be geared to our "clients".

It is desirable that suppliers of information deal with just one CBS-department, if at all possible. This means that the data capture process must be restructured. Questionnaires must be easier to complete, duplication of questions must be avoided, sampling must be directed to reducing and spreading the response burden. Better use will have to be made of administrative sources. Better than before, the usefulness of statistics must be made clear to respondents. Summaries of results must systematically be fed back to suppliers of information.

Dissemination must be restructured as well. A similar reasoning as above applies. Users must in principle be helped through a single "information desk". Apart from very specific details all questions must be handled centrally and speedily. To improve the image of the CBS, modern dissemination methods must be used more extensively. The development of publications for a general audience needs more effort.

The internal structure has to be made less fragmented. The division of labour between our two offices must be reconsidered. The management structure and corporate culture must be rethought in order to promote clearer responsibility structures, more operational independence at lower levels, shorter procedures, more businesslike relations, fewer levels, improvement of communications and less fragmentation.





The process of necessary changes will be structured in phases. The goals are set out; the structure of the CBS must be geared to achieving these goals. The discussion is now open. First of all consensus about the goals is necessary. After that changes must be formulated in operational terms.

This will require a considerable effort. Several working groups will have to make a contribution. We believe, however, that there is more than enough creativity, energy and knowledge available at the CBS to make operation TEMPO a succes. Meanwhile current production activities must go on as usual.

There will be a timetable and detailed planning. After each phase there will be concertation, followed by decisions. The process will be characterised by fair play. Moreover, systematic information about the progress will be supplied, in order to let everyone know "where we stand".

Phase 1 of the process is discussion at all levels about the vision set out in this manifesto. Simultaneously, the board of directors will draft the above-mentioned time-table.

Phase 2 is making TEMPO operational in working parties. Their reports will be subject of the same types of discussions, including those with personnel committees.

Phase 3 is the implementation phase. Phases 1 and 2 must be finished this calendar year. 1993 will be the implementation year.

Each consecutive step will only be made after the preceding step has been concluded with a clear decision. Introducing the changes we face in a proper fashion will require great efforts, as well as a high degree of discipline; the same discipline we need in the CBS of the future.

Back to top | Statistics Division Home | Contact Us | Search | Site Map
Copyright © United Nations, 2014