Regional Workshop for African Countries on the Compilation of Basic Economic Statistics

1
Regional Workshop for African Countries on the
Compilation of Basic Economic Statistics
United Nations Department of Economic and Social
Affairs/Statistics Division (UNSD)
Strengthening national statistical infrastructure.
Addis Ababa, Ethiopia, 16 19 October 2007 Len
Cook, UNSD Consultant
2
(No Transcript)
3
Vision - Forum on Africa Statistical Development

•   Our vision..
• as core producers of statistics in Africa
• is the installation of a statistical
  infrastructure,
• through enhanced partner-ship among stakeholders
  at all levels,
• capable of providing and promoting the effective
  management and utilization of quality data, and
• information for evidence-based policy and
  decision-making,
• monitoring development programmes and
• democratic accountability.
• Statistical Development in Africa A Strategic
  framework

4
Infrastructure A sound balance

• The national statistical infrastructure contains
  of a mix of institutions, institutional
  relationships, systems, processes practices and
  methods.
• Having a balanced approach to infrastructure
  development may be more important for the long
  run success of the statistical system than
  significant achievement at any single one
  element.
• the way the whole range of infrastructure fits
  together and meets country needs will be unique
  to the country

5
Local context of statistical infrastructure
  Place statistical infrastructure of any country
in the wider context of the infrastructure that
exists for all business and the other activities
of government. - Tangible, (transport and
communications systems) - Less tangible (common
tax numbers, tracking goods, services or people
as they enter the country. )
6
Core capabilities of National Statistics
Fundamental Pillars of Official Statistics
PERFORMANCE
RELEVANCE
I N T E G R A T I O N
T R U S T
EXPERT
STATISTICAL SYSTEM FOCUSED CAPABILITIES
GLOBAL NETWORKS
NSI FOCUSED CAPABILITIES
Statistical Legislation Statistical
Standards Statistical Frameworks Training Knowledg
e of methodologies International
co-operation Statistical Organisation
Privacy Legislation Code of Practice Advisory
Bodies National Information Policy Regional
Collaboration Information Technology Market
Infrastructures Accreditation of quality
UN Principles of Official Statistics UNSNA Beijing
(Women) Rio (Environment) IMF SDDS
7
Implications for transfer of systems and processes
Translating survey design methods adopted by well
endowed countries into countries that are less
well endowed with business or government
infrastructure may bring about suboptimal
solutions which are less effective than those
which take account of what infrastructure works
best in the local or regional context.
8
User driven focus on infrastructure

1. The national statistician is an investor, while
   existing and future users are consumers
2. Leadership in national statistical policy within
   the country must focus on enabling decisions
   based on sound statistical measures to be made in
   a timely and decisive manner.
3. Strong user relationships and interaction are
   essential to give the national leadership a clear
   focus on strategies that meet local needs,
   alongside Millennium Development Goals, the
   national accounts and other international needs.
4. Anticipate emerging needs e.g. expectations of
   integration between social and economic
   statistics
5. Manage obsolescence of systems, methods and
   processes

9
(No Transcript)
10
(No Transcript)
11
(No Transcript)
12
Infrastructure depends on Governance(Organisation
al and constitutional)

• Statistics legislation, in the context of other
  laws (privacy, freedom of information,
  administrative)
• Procedure, custom and practice, in applying law
• Organisational arrangements and roles
• Organisation of the national statistical office
• Ministerial arrangements and status, and
  delineation of roles
• Place within the organisation of government of
  the statistical office and the national
  statistician
• Other decision-making authorities

13
Elements of Governance(practice)

• Capacity for independent action
• Quality of leadership
• Professionalism of statistical staff
• National statistical standards and frameworks
• Budget allocation and management
• Forms of compliance and accountability
• Public sector management environment
• Role of management boards, statistical councils

14
Elements of Governance(practice)

• Users are not investors, and it is the national
  statistician who invests in methodology,
  systems, publishing services, survey capability,
  skills and identity needed to produce official
  statistics
• Quality of leadership
• Professionalism of statistical staff
• National statistical standards and frameworks
• Budget allocation and management
• Forms of compliance and accountability
• Public sector management environment
• Role of management boards, statistical councils

15
Governance Now(institutional preferences)

• Networked organisation form
• Re-enforce standards with common tools and
  cross-office architecture
• Ongoing common training and development process
• High quality, responsive central services
• Statistical system responses to emerging issues
• Loose oversight of influential activity that
  integrates with official statistics

16
Local partnerships may speed change and increase
opportunities

1. Partners in collaboration in systems and methods
   that challenge the transfer of practices from
   unrelated countries will come from country
   organisations that have similar interests in
   information exchange and transfer, although their
   field of interest may be very different.
2. Because the payback period for statistical
   investments is comparatively long compared to
   many other fields of activity, technological
   innovation of local relevance in other agencies
   may have an even greater impact on the
   collection, analysis and release of statistics.

17
Fit development path to long run budget

1. Some elements of infrastructure will be
   determined by the economic position of the
   country.
2. For example, there is now a rich array of
   freeware for information technology, which
   embraces the uses that statistical systems have
   for information technology.
3. This means that technology resources can be
   directed at applications rather than software
   purchase and fees.
4. Donor activity can be required to adopt the NSI
   standard tools

18
Development path for NSI Planning
Anticipate change in direction of statistical
sources and statistical measures
High Continuity of Government Funding to
NSI low
Capacity to change the scope and quality of
statistics
Capacity to extend statistical activities and
maintain regular activity
Capacity to repeat annual and monthly operations
Low ---- User commitment ---- high
19
Contribution of investments to performance
Presentation and extraction tools
Competitiveness, Innovation, Responsiveness
Methods and survey tools
Best practice, Reliability
Sources Information Management
Cycle times, Cost structure, Integration
Meta data
Coherence, Knowledge base, Geographic base
20
Managing obsolescence in statistical process
Development project scale
18 months 15 months One year 9
months 6 months 3 months
Sources Information Management
Meta data
Methods and survey tools
Presentation and extraction tools
Investment Life-span
Cycle times Cost structure Integration
Coherence Knowledge base Geographic base
Competitiveness Innovation Responsiveness
Best practice Reliability
21
Integrating elements
COMMONQUALITYMEASURESFOR SURVEYEVENTS
COMMON ORGANISATIONOF DATAARRAYS
INTEGRATION THROUGH COMMON METHODOLOGY
ANDPRACTICE
REAL-TIME PROCESSES OFSTATISTICAL FRAMEWORKS
SYSTEMS
MATCHING BY COMMON STATISTICAL SURVEY FRAMES,
AND IDENTIFIERS
REAL TIME SURVEY PROCESS INFORMATION
REPLACEABLE COMMON TOOLS, AND SYSTEMS
COMMON DATA MANAGEMENT (ORACLE)
USERS DETERMINE WHOLE SYSTEM CROSS CUTTING
VARIABLES
META DATA
22
Survey design and statistical infrastructure
reflect the local context

1. Simple survey designs and estimation methods
   bring added flexibility but may require larger
   sample sizes
2. Household surveys have an increasing place in
   economic statistics, initially through extending
   the scope of well established sources
3. When available, tax records can be used not only
   to manage the survey frame through a business
   register, but to provide responses for small
   firms.
4. Area samples, sampling transactions, combining
   information may be more efficient with current
   tools, where registers are poor

23
Collaboration among African countries may enable
more rapid development of the most relevant
methodologies, (maintain a core of local
expertise).
Graphical methods of validation can enable
outliers to be managed with greater certainty
than traditional methods that necessitate ongoing
oversight by changing personal. Bring together
the validation and linking of the business
activity of large enterprises at a micro level,
to improve the quality of balance of payments,
finance and banking, pensions statistics.
24
(No Transcript)
25
(No Transcript)
26
(No Transcript)
27
Focusing on an Integrated, enterprise wide
technology architecture

• Common tool box
• Systems and methods interaction (External trade)
• Web driven
• International collaboration
• Technology is a boardroom issue
• Start simply
• Change psychology of organisation

28
Essential challenges Clear authority for the
place of the statistical system

• Have final authority for integrity with
  statisticians
• Fund mix of current activity and investments
• Make quality visible

29
Organisational infrastructure
Clear authority for the place of the statistical system Give final authority to statisticians for consistent actions on matters of integrity in statistical practice Commitment to funding of a mix of current activity and investment to bring later benefits Make visible the quality of key statistics
Capacity to organise a long run program of activity Competence includes corporate management Infrastructure as central focus of investment, and of performance benchmarks Match the resources for statistics with groups of statistics and statistical developments
Ways of Keeping relevant Engage with users and the uses made of the statistics
Shared Commitment Consistent reinforcement of integrity in statistical practice
30
(No Transcript)
31
(No Transcript)
32
Decision-making in economic statistics
Global accumulation of knowledge of decisions and
s/understanding needed in societies
National accumulation of knowledge
Statistical Frameworks national/international
Statistical processes
Statistical practice
Statistical methods
Surveying
Statistical measures
Country decisions/ understanding
33
Investment level and Performance
System Performance
System focused contribution
Institution focused contribution
Investment Level
Pacific Africa
Balkans East Europe
France United Kingdom Italy
Australia Canada Norway
34
Costs of inadequate investment in statistical
leadership

• Development risks are seen as procurement risks
  instead of statistical ones
• Ignore dynamics of statistical processes and
  over-emphasis on one-off static performance
• Synchronous activity and balances across the
  whole system not reflected in design goals
• Absence of a compelling agenda for donor
  countries to participate and contribute
• Low focus on learning and development activities
  that expand capacity from joint development and
  top level interaction
• Infrastructure is not a stable platform for
  necessary dynamic of responses to emerging needs
• The piecemeal evolution of statistical processes
  means that the investment resources have a
  reduced impact on the long term performance of
  the statistical system

35
The management focus on change
No risk access Simplify survey base Link with
oversight body Contact management Explain uses to
respondents
trust and goodwill of respondents

Survey capability
Manage Statistical Processes
Standard frames Single geographic
reference System wide meta data Standard survey
design
Information management
Delivery
Analysis
Reporting on trends Demography capability
Standard products Market priorities Small areas

User management
Understand Value added
Delivery standards Delivery monitoring Audit
reviews National Statistics reviews
Risk management Value analysis Strategic reviews
Competitive advantage of whole statistical system
36
(No Transcript)
37
Results of Improved Methodology and
Technology Description Access to new
administrative data-sets Apply classifications in
more detail, at low cost Rapid digitising of
information New monitoring of business
transactions (SWIFT) Strategic view of frame
management
38
Results of Improved Methodology and Technology
Extraction Transfer information fast Flexible
access to existing data-sets Graphical validation
processes Total quality management
processes Apply complex sampling models Recognise
outliers Apply common tools
39
Results of Improved Methodology and Technology
Integration Increased capacity to adopt common
identifiers Apply statistical frames to all
data-sets Match data-sets with common
identifiers Extend analysis from existing panel
data-sets Apply statistical matching Apply
statistical unit classifications to event
data Use power of data management tools
40
Results of Improved Methodology and Technology
Estimation Apply user specified models to
data-sets Estimate sub-population
characteristics Provide micro-data based measures
(concentration, distribution, transitions) Greater
validation of estimates
41
Results of Improved Methodology and Technology
Selection Expand variety of relevant
comparisons Increase small population
measures Provide micro-macro data
integration Regionally specific regional data
selection Data defined regions