Third International Conference on Establishment Surveys Introductory Overview Lecture

1
Third International Conference on Establishment
SurveysIntroductory Overview Lecture

• Definitions and Units
• Andreas Lindner
• OECD Statistics Directorate

2
Content of the lecture

• Importance of statistical units
• National data collection systems and their
  integration into the international dimension
• Different definitions of statistical units
• OECD survey on SME business frames
• Compilation strategies and links to
  administrative sources
• The impact of globalisation
• Linking systems together
• Multinational enterprises (MNEs) and Enterprise
  Groups (EGs)- a new statistical unit
• Outlook

3
Are statistical units important?

• The definition of units is the backbone of
  registers
• They largely determine whether or not economic
  and social reality is correctly reflected in
  statistics
• Together with the business register design they
  allow (or prohibit) to correctly measure
  production, the flow of goods, services and
  capital, employment, financial data, ownership,
  control etc.
• A coherent system of interrelated standardised
  statistical units is key for effectively
  combining data from different surveys and, hence,
  capturing the full array of economic activities
• Coherence with and integration of the
  international dimension is required since it
  reflects the way businesses operate
• Statistical units are very important!

4
Units constitute systems

• Administrative legal system
• Economic system
• Statistical system
• Global, international system
• Are these systems in tune with each other ?

5
National systems

• Great variety of units and definitions exist,
  because they have to fit the national set of
  conditions
• The statistical unit(s) are the translation of
  the legal and administrative set of conditions
  into an as much as possible standardized
  statistical representation of the society
• Ideally, they should allow consistent linking
  across data collections, representing a data
  warehouse and not isolated stovepipes
• Countries largely differ in their ability to do
  so

6
should allow to bridge to the international
dimension

• Any economy is today more than ever integrated in
  an international world economy with complex
  interrelations
• For statistical analysis, this requires to have
  standardized concepts and definitions across
  countries, allowing to compare the comparable
• This represents a permanent challenge to
  statisticians and International Organisations are
  called upon to facilitate this process by issuing
  international recommendations and measurement
  standards

7
An OECD survey on SBS/SME statistics identified
the following statistical key policy
recommendations

• Promote international convergence of statistical
  concepts and processes
• Foster greater international comparability of
  statistics
• Promote development of an integrated business
  statistical register
• Promote data linking to make better use of
  existing data and reduce respondent burden on
  SMEs
• Carry out policy-relevant empirical analyses to
  underpin evidence-based policy making

8
A mix of definitions of SMEs

• Legal statistical definition(5 countries)
• Most EU countries differentiate legal from
  statistical purposes (see box 1)
• Statistical definition generally based on number
  of employees or mix (e.g. persons and turnover
  etc.)
• Different thresholds and sector-specific
  particularities

9
Legal/administrative vs. statistical definition
of SMEs in the EU
10
Rich variety of statistical units

• General impression the same name does not
  necessarily mean the same thing in another
  country. True? Evidence suggests that this is the
  case.
• Is the enterprise vs. establishment split
  meaningful? Why not calling it economic
  statistical unit?
• Despite Council Regulations, the situation in the
  EU seems less harmonized than one might assume

11
Council Regulation 696/93 8 different
statistical units (legal, geographical, activity
criteria)

• the Enterprise
• the Institutional Unit
• the Enterprise Group
• the Kind-of-activity Unit (KAU)
• the Unit of Homogeneous Production (UHP)
• the Local Unit
• the Local Kind-of-Activity Unit (local KAU)
• the Local Unit of Homogeneous Production (local
  UHP).

12
UN classification ISIC

• the enterprise
• enterprise group
• kind-of-activity unit (KAU)
• local unit
• establishment
• homogeneous unit of production.

13
The Canadian statistical structure

• Enterprise the highest level of the hierarchy
  with complete financial statements and
  information about international links. An
  enterprise may have one or more companies
• Company A somewhat homogeneous production
  structure with balance sheets data. May have one
  or more establishments.
• Establishment The most homogeneous production
  level with economic key data(output, inputs,
  wages). May have one or more location.
• Location A unique physical production unit.

14
A wealth of statistical units

• The EU focus on enterprise can be considered as
  a compromise taken in 1993 to accommodate very
  different national practices
• The system of international standardisation is
  incoherent with respect to statistical units
• EU terms need to be translated into world-wide
  methodology
• The legal dimension needs to be distinguished
  from the economic dimension
• The economic dimension is likely to become more
  important

15
OECD survey business frames used for SME
statistics

• Half of countries reported use of combination of
  different sources, sometimes according to sectors
  and/or size classes
• Agriculture often the exception to the rule (agr.
  from statistical source)
• Very different thresholds across countries

16
Effects of different thresholds on coverage
four EU countries in comparison to the Italian BR
100
17
Identified limitations of Business Frames for SBS

• Different updating intervals limit comprehensive
  coverage
• A specific SME frame is the exception
• Confidentiality issues limit availability of data
  for other users/producers
• General concern about quality and coverage of
  demographic data, in particular for deaths
• Difficulties were reported as to the proper
  allocation of activities to industries
• The quality of the Business Frame was generally
  considered as appropriate, although improvements
  are foreseen in many countries with respect to
  SMEs, change of activity, legal status etc.

18
The traceability issue

• Ability to trace change through a unique
  identifier code for the local unit/establishment
  would allow to better capture unit activity and
  de-registration
• Similarly, the introduction of links between
  legal units in enterprise groups and groups of
  companies would enhance identification of changes
  in size, ownership and location for units
  involved in events such as mergers and splits

19
Collection and compilation strategies

• Germany and the US are the only countries in the
  sample under review where the NSO is not in
  charge of official statistics on SMEs.
• Almost all countries have different treatment for
  core and specific statistics
• Some countries have developed tools to monitor
  the response burden. The strategy developed is
  therefore to collect core variables through the
  integration of census/surveys based data and
  administrative data

20
Collection and compilation strategies (2)

• In the majority of countries the NSO is fully in
  charge of data collection
• In the remaining countries, the NSO plays an
  important coordinating role (Germany is the
  exception where the NSO has outsourced SME data
  collection)
• In the majority of countries, SME core statistics
  are differentiated from specific SME variables.
  The typical pattern is a reduced sample for core
  data as opposed to often voluntary thematic
  surveys

21
Data linkage with administrative sources

• Very different access patterns for NSOs to
  individual records could be identified, ranging
  from full to no access.
• Where access to administrative and other sources
  is partial, differences in the definition of
  variables have been commonly seen as a major
  impediment to the use of administrative sources,
  different observation units, classification and
  the absence of a unique identification number
  have also been mentioned.
• Sometimes technical problems hinder access to
  administrative sources. Countries generally have
  identified the main sources they would like to
  access

22
Access and linkage with administrative sources

• The picture regarding access of NSOs to
  administrative SME data is mixed. Country
  practice differs ranging from full access via
  partial access to no access
• Although about 2/3 of responding countries state
  that NSOs have full access, half of them reported
  problems in uses or little practical experiences
• Five countries reported only partial access
• Japan and Switzerland reported that no access
  was granted to NSOs

23
Access and linkage with administrative sources (2)

• In the case of access, but no usage, the main
  reason were different basic units and absence of
  links between registers and administrative data
• Similarly, the main impediments to a better use
  of available data in the two distinct sources
  were
• Different definitions of variables
• No common identifier
• Different classifications and thresholds

24
The impact of globalisation on statistical
units/definitions

• One of the most visible manifestations of a more
  globalised world is trade. The remarkable growth
  of trade flows is undisputable, in particular for
  emerging major players (see next slide)
• But, as can be seen from the remarks later under
  Linking systems together, even trade and
  business statistics are rather separate entities
  needing reconciliation/standardisation

25
Relative growth of exports of goodsGrowth over
the period 1996-2005, OECD total 1
26
The impact of globalisation on statistical
units/definitions (contd)

• The real problem is that statistical systems have
  been overtaken by the speed of globalisation
  obliging national reporting systems and
  structures to align to this paradigm shift
• In other words, national concepts need to
  incorporate the international dimension to
  adequately reflect the reality of todays
  production process

27
Statistical units/definitions more or less in tune

• Administrative legal system
• Statistical System Economic System

28
went out of tune

• Administrative legal system
• Statistical System Economic System

29
As a consequence, both Administrative legal and
Statistical System

• Need to adapt to (better integrate) the new
  reality
• where the production process is international,
• with more complex structures/processes exist
  (outsourcing, flatter vertical integration)
• and where control and ownership become more
  dominant features than legal forms

30
Linking systems together business statistics and
trade

• Global economies require data on who is trading
  and what are the characteristics of trade
  operators
• This question requires establishing a direct
  relation between foreign trade and industrial
  statistics.
• But these two statistical domains are based upon
  different concepts (products versus economic
  activities) and use different classifications
  (SITC, HS, ISIC, CPC).

31
Linking systems together business statistics and
trade

• Central issue of such an analysis is to try to
  classify trade operators according to enterprise
  characteristics
• But this depends on the possibility of using or
  developing common identifiers between the trade
  register and the business register
• Countries largely differ in their ability to
  perform such a linking.

32
The present situation

• The situation in Non-EU countries of OECD can
  be summarized as follows
• Six out of eight countries reported having a
  trade register, only 1 country reported having a
  formal trade register
• How is the trade register updated? Customs
  declarations provide the basis. In half of the
  responses there is a link to the Business
  Register and data from fiscal authorities is used
  as well

33
The present situation (contd)

• Which is/are the unit(s) of reference in the
  trade register(s)? Enterprises and establishments
  constitute together with legal units the
  reference units used.
• Is there at least one common unit of reference
  between the business register and customs
  forms/register? All countries, except one,
  reported yes.
• Is (or can) the basic statistical unit of the
  business register linked to customs forms/
  register of traders? Almost all countries
  reported that they can link the basic
  statistical unit to customs forms/trade registers

34
The present situation (contd)

• The situation reported was a mix of different
  identification means, such as the tax
  registration number, name of the company and its
  address, the Business Register identification
  code.
• Although linkages do well exist, different
  reporting systems in countries inhibit to some
  extend linking.
• The risk of double-counting was acknowledged and
  the problematic measurement of multi-nationals
  was highlighted.

35
The present situation (contd)

• Have Non-EU countries of OECD carried out any
  statistical matching exercise between the trade
  register and the business register?
• Only three countries reported to have done so
• The more systematic integration of the
  international dimension into business frames
  seems to be rather the exception than the rule
• This could be pursued as a development goal at
  international level together with countries

36
Multinational enterprises (MNEs) and Enterprise
Groups (EGs)

• Beyond the national territory, a statistical
  system needs to take into account more
  systematically the international dimension
• Foreign Direct Investment
• Foreign Affiliates Statistics
• Enterprise Groups, in particular Multinational
  Enterprises as new statistical unit

37
FDI Flows and Stocks

• Foreign direct investment (FDI) is a key element
  in the rapidly evolving process of international
  economic integration. FDI creates direct, stable
  and long-lasting links between economies. FDI
  encourages the transfer of technology and
  know-how between countries, and it allows the
  host economy to promote its products more widely
  in international markets

38
FDI

• Foreign direct investment (FDI) is defined as
  investment by a resident entity in one economy
  with the objective of obtaining a lasting
  interest in an enterprise resident in another
  economy.
• The lasting interest means the existence of a
  long-term relationship between the direct
  investor and the enterprise and a significant
  degree of influence by the direct investor on the
  management of the direct investment enterprise.
  Absolute control by the foreign investor is not
  required, and ownership of 10 of the voting
  power is the criterion used.
• Inward stocks are the direct investments held by
  non-residents outward stocks are the investments
  held in other economies.

39
FDI global trends

• The global environment for FDI improved in 2005
  while at the same time corporate profitability
  was generally strong.
• Direct investment into OECD picked up in 2005
  when inflows reached 622 bn.
• The United States and the United Kingdom were
  the main destinations for FDI in the OECD.
• Investment flows to EU countries also increased
  notably by 75 while half of the increase
  accounts for FDI flows into the United Kingdom,
  which tripled in 2005.
• Investments into China, amongst the foremost
  destinations of FDI, further progressed.
• The OECD area continued to be net outward
  investor at around 95 bn in 2005

40
FDI
41
Why Foreign Affiliates?

• Since the mid-1980s and with the quickening pace
  of globalisation of the economy, foreign direct
  investment has become central to worldwide
  industrial restructuring and one of the most
  dynamic elements of international transactions
• The need to assess the role and impact of direct
  investment, not only in financial terms but also
  in the context of its impact on jobs, sales and
  trade, has underlined the importance of having
  data indicative of the industrial activity of
  multinational firms
• To meet these new needs in analyzing the
  globalisation process, the OECD launched the
  regular collection of data on the performance of
  foreign affiliates in the manufacturing sectors
  of the OECD countries

42
Foreign Affiliates

• OECD Definition An affiliate under foreign
  control is defined as one in which
• A single foreign investor holds more than 50 of
  the shares with voting rights.
• The notion of control allows all of a companys
  activities to be attributed to the controlling
  investor.
• This means that variables such as a companys
  turnover, staff or exports are all attributed to
  the controlling investor and the country from
  which he or she comes. Control may be direct or
  indirect.

43
Foreign Affiliates economic activity is
significant in OECD countries Employment in
manufacturing and services in affiliates under
foreign control (as of total employment, 2004
or latest available year)

44
A new statistical unit is needed the Enterprise
Group (EG)

• EU definition
• An enterprise group is an association of
  enterprises bound together by legal and/or
  financial links. A group of enterprises can have
  more than one decision-making centre, especially
  for policy on production, sales and profit. It
  may centralize certain aspects of financial
  management and taxation.
• It constitutes an economic entity which is
  empowered to make choices, particularly
  concerning the units which it comprises.
• EUROSTAT is actively working on the EG Register

45
EGs and EG-sub-groups a scenario based upon real
data

• The economic picture of a country crucially
  depends on the degree of inclusion or exclusion
  of statistical units
• France has launched in 2005 a major high-level
  project of re-designing its enterprise statistics
  system
• This project was carried out by the CNIS, the
  National Council for Statistical Information
• In April 2007, the final report was presented

46
EGs and EG-sub-groups a scenario based upon real
data

• Some conclusions
• INSEE, the French NSO, will use this report for
  defining procedures to incorporate enterprise
  groups in the collection system of economic
  statistics
• An affiliate is no longer considered as an
  equivalent ot an enterprise unit
• An affiliate can not be considered as enterprise
  (lack of autonomy)
• A MNE has to be considered globally (an MNE and
  its branches think and define themselves
  globally)
• Many economic actors consider the EG (or
  sub-group of it) as the relevant unit of analysis
• Users want to analyse market realities, not
  hierarchical structures within EGs

47
EGs and EG-sub-groups a scenario based upon real
data

• The CNIS recommends the addition of a Global
  Statistical Unit (GSU) for EGs
• It is at this level where the strategic direction
  and decisions on affiliates are taken (mergers,
  acquisitions, investment, relocations etc,)
• This GSU is independent of national frontiers
• This logic of definition leads to a statistical
  representation of the national economy as the sum
  of all French GSUs world-wide

48
EGs and EG-sub-groups a scenario based upon real
data

• This approach is innovative in macro economic
  statistics since it goes systematically beyond
  the national territory (to our knowledge, only
  the BEA in the US has a similar approach)
• Most traditional economic statistics take the
  national approach in terms of national territory

49
Simulation done by the CNIS in France

• France World

(A)Foreign- controlled Companies 2 M
(B)Legally Independant Companies 6,3 M
(C)French Groups 2,1 M
(D) International French Groups Df 4,2 M
Dw 3,6 M
50
Simulation done by the CNIS in France

• Truncated Global Statistical Unit(GSU)
• ABCDf 14,6 million salaried employees
• French truncated GSU
• BCDf 12,6 million salaried employees

51
Simulation done by the CNIS in France

• World-wide French GSU
• B C Df Dw 16,2 million salaried empl.
• A world-wide French measurement, including the
  activities abroad from French EGs would yield
  16,2 million employees, compared to 14,6 million
  in the territorial view a plus of 1,6
  millions

52
Outlook

• This lecture has tried to shed some more light
  on the need to move towards a better integration
  of the international dimension into national
  statistical reporting
• Much need to be done and this includes the
  following

53
Updating the SNA

• to allow a better measure and integration of the
  increasingly global nature of economic
  transactions and arrangements. Particular
  attention should be paid to
• Goods for processing the gross or net treatment
  would fundamentally affect the nature and size of
  trade in goods and services (goods or services)
• Government-owned foreign subsidiaries recognize
  both the parent and the subsidiary as
  statistical unit
• Better measuring cross-border service flows

54
International production processes necessitate a
fresh look at international trading and
contracting

• More an better data needed on re-exports
• More and better data needed on further processing
  and merchanting
• FATS gt inward FATS alongside SBS needed, idem
  for a reduced set of outward FATS data (more
  difficult)
• MNEs gt consistent definition of EGs needed and
  integration into registers
• Profiling of domestic operations needed
• Difficulty consolidation/deconsolidation of
  financial results of EGs

55
A preliminary conclusion

• Along with more consistent national data, greater
  international consistency with respect to
  concepts and definitions of statistical units is
  necessary, namely the addition of
• The enterprise group as unit of analysis
• which is relevant for the analysis of MNEs
• and which should be recognized and recommended
  in SNA as one statistical unit
• ... and should be integrated in national data
  collection systems

56
Thank you for your attention!

• Questions? gt andreas.lindner_at_oecd.org