Working Party on National Accounts

1
Towards a better measure ofnon market output

• The OECD non market project
• Report of the joint OECD/UK/Norway seminar
• (London, October, 3-5, 2006)

2
The context SNA 93

• 16.134. In principle, volume indices may
  always be compiled directly by calculating a
  weighted average of the quantity relatives for
  the various goods or services produced as outputs
  using the values of these goods and services as
  weights.  Exactly the same method may be applied
  even when the output values have been estimated
  on the basis of their costs of production.
• 16.138. There is no mystique about non-market
  health or education services which make changes
  in their volume more difficult to measure than
  volume changes for other types of output, such as
  financial or business services or fixed tangible
  assets.  Moreover, changes in their volume are
  also needed in order to be able to measure volume
  changes for the actual consumption of
  households.  The same principles apply to the
  measurement of consumption as to production.  

3
The context the European handbook
pupil-hours or number of pupils

• Education the quantity of teaching received by
  the students, adjusted to allow for the qualities
  of the services provided, for each type of
  education
• the transfer of knowledge, successfully or not
  
• Health quantity of health care received by
  patients, adjusted to allow for the qualities of
  service provided, for each type of health care.
•  The quantity of health care received by
  patients should be measured in terms of complete
  treatments .

class size, school inspections, scores ?
by ISCED-97
number of treatments, or consultations, or first
visits
?
by providers, then by functional care, then by
DRGs
4
The context the Atkinson review

• Principle A the measurement of government
  non-market output should, as far as possible,
  follow a procedure parallel to that adopted in
  the national accounts for market output.
• Principle B the output of the government
  sector should in principle be measured in a way
  that is adjusted for quality, taking account of
  the attributable incremental contribution of the
  service to the outcome.
• Principle C  Account should be taken of the
  complementarity between public and private
  output, allowing for the increased real value of
  public services in an economy with rising real
  GDP.
• 6 other principles (H on triangulation)
• Note the UK had already output methods, but
  sub-optimal ?

good practice for quality adjustment
5
The context a transatlantic gap ?

• European countries by commission decision
  2002/990/EC, almost all countries will apply
  output methods in education and health according
  to the European handbook recommendations, in 2006
  or following years.
• Australia and New Zealand have applied also
  output methods for years.
• North America (US, Canada) still apply input
  methods
• Output methods often lt Input methods
• There could be a transatlantic gap of 0.25 of
  GDP growth / year.
• OECD role is to promote international
  comparability.

6
The OECD non market project

• This project, focussed on education and health
  especially, both in temporal and spatial
  dimensions, was presented by OECD at the 2005
  meeting of the Committee on Statistics.
• INSEE has assigned Alain GALLAIS to coordinate
  this project in OECD, for 18 months (April 2006
  -gt September 2007).
• workshop in London, 3-5 October 2006, co-hosted
  by ONS and the government of Norway. Attendance
  of more than 160 experts of National Accounts,
  PPP, education, health and public services
  efficiency fields.

7
Objective OECD handbook on best practices

• Review of best practices in output methods, more
  detailed international guidelines for temporal
  analysis
• Frameworks and formulas for output methods in PPP
  limited to education and health

• OECD non-EU countries
• Questionnaire OECD June 2006
• AU, NZ wo quality adjustments, projects in the US
  ?
• gt Promotion of output methods

• EU countries
• Questionnaire Eurostat June 2006
• Almost all countries in the process, but few with
  quality adjustment
• gt Promotion of best quality adjustments

• The originality of the OECD non-market project
  develop a synergy between experts in NA, PPP,
  education, health and public services efficiency
  fields (several networks).

8
The London seminar the UK experience

• Of course, the UK was not chosen  by hazard  to
  host the workshop the ATKINSON review and then
  the creation of the UKCeMGA inside the ONS have
  provided a good experience, still in progress
  (public order and safety, social services).
• Importance of the communication strategy. Key
  stakeholders.
• Need for more timely, detailed and better quality
  data on government expenditure.
• A lot of proposals by DfES (education) and DH
  (health).
• Not always easy for the NA unit.

9
The London seminar nordic experiences

• Norwegian StatRes project transparency and
  benchmarking. Mapping the current information
  jungle (input, activity, output, outcome)
• KOSTRA for municipalities
• SAMDATA for health statistics
• Skoleporten for education
• Each sector / agency has its own specific website
• Swedish NA all individual services, including
  social services and culture.
• Finnish project for NA each public entity is
  asked on its shares of turnover / costs and on
  its outcomes / quantity indicators. No quality
  adjustment. Example

10
The London seminar US projects

• There is no agenda for moving to output methods.
• But there are 3 parallel projects on alternative
  methods for education
• Barbara FRAUMENI et al., relying on human capital
  theory
• Michael CHRISTIAN et al., more classic output
  method
• Use of market prices.
• In health, research is still "in its nascent
  stage. A paper of Michael CHRISTIAN has explored
  several possibilities, mainly relying on DRGs,
  leading to counter intuitive results (output lt
  lt input).

11
The London seminar terminology aspects

• In the view of policy makers

Activities quantities ?
Source Wouter Van Dooren
12
The London seminar terminology aspects
No ! (quality adjustment)

• In the view of health economists
•  The term of outcomes refers to a
  multi-dimensional concept of (positively or
  negatively expressed) characteristics of output,
  such as health changes, responsiveness of
  services to peoples non-health expectations
  (e.g., no waiting time, other convenience of
  access, quality of facilities, and amenities in
  hospitals or nursing homes). There are several
  possibilities for valuation of changes in health
  states measured, such as Quality Adjusted Life
  Years (QALYs).

Source Manfred HUBER
13
The London seminar scares and hopes about
output methods

• National Accountants
• Before, with hypothesis of null productivity,
  countries were comparable.
• Now, with diversity on quality adjustment,
  countries could not be comparable any longer.
• Data on productivity will be scrutinized by
  politicians
• But new productivity is often negative !
• Output is not outcome

• Policy makers
• Before, output in NA was of no interest.
• Interest of harmonization.
• Now, NA could capture all outcomes useful for
  policy makers, with an official and consistent
  label, according to the vocation of NA.
• Interest of explanations.
• NA can only be one-dimensional

14
The London seminar differencesbetween output
and outcome (1)

• Influence of consumers characteristics
• The outcome can be influenced by consumers
  characteristics (level of knowledge of pupils,
  health status of patients before they are
  provided any service). Should the output be as
  close as possible of outcome, or should it
  neutralize consumers characteristics to rely
  only on providers ?
• Externalities
• In general, National accounts do not record
  externalities (pollution). But some expected
  outcomes of an activity can concern other ones
  (road safety for health or health status for
  market economy). Should the output of public
  services record the externalities on other
  activities ?

15
The London seminar differencesbetween output
and outcome (2)

• Inequalities
• Some policies in individual non market services
  focus on reduction of inequalities (of knowledge
  and skills in education). Market activities do
  never take them into account (total F C single
  addition of individual F C, total income).
  Should NA make an exception for non market
  services ? Including individual services ?
• Coverage of policy makers purposes
• Must all the dimensions of outcomes be reflected
  in the one-dimensional output ?

16
The London seminar education

• The European handbook suggests several
  possibilities for quality adjustments
• Pupil-hours (if possible adjusted for occasional
  absence of pupils) are a better  quantity 
  indicator than number of pupils
• The class-size or pupil/teacher ratio after a
  sound analysis on its effect on the quality of
  teaching
• School inspections
• Examination data.
• But are all these quality adjustments consistent
  with each other ?

17
The London seminar education for PPP (not yet
validated)
Translation of the same formula in spatial
dimension
Output in NA (my point of view) OECD Education unit statistics Outcome for policy makers (ex of FR point of view)
PISA national scores corrected on ESCS time-lag PISA national average scores Level of knowledge
Out of purpose of NA PISA national SD Reduction of inequalities
Externality ? Out of purpose of NA Citizenship
Quality adjustment for upper second. vocational Rates of employment Insertion in Labour market
18
Education draft framework for PPP (M NM)
Recommendations of the handbook Finest detail of cycles Act. F.C. Num. pupils Quality index
Pre-school - ISCED 0 Pre-school none
Primary school ISCED 1 Primary school PISA / PIRLS TIMSS ?
Primary school ISCED 1 Special primary PISA / PIRLS TIMSS ?
Lower secondary ISCED 2 Secondary 1st stage PISA
Lower secondary ISCED 2 Special secondary PISA
Upper secondary - ISCED 3 general, pre-voc. PISA
Upper secondary - ISCED 3 vocational PISA employ. rate ?
Post-sec. non tert. - ISCED 4 Post-secondary non tertiary PISA
Higher education - ISCED 5 and 6 Long superior ISCED 5A Number of degrees and duration of studies.
Higher education - ISCED 5 and 6 Short technical superior-ISCED 5B Number of degrees and duration of studies.
Higher education - ISCED 5 and 6 Advanced research qualification ISCED 6 Number of degrees and duration of studies.
Other education
19
The London seminar proposals of DH for health
output (UK)

• Two treatments might have the same costs, but
  one might result in a major increase in health
  outcomes (longer life, or better quality life)
  while the other might have much smaller health
  gain. The aggregate measure of healthcare output
  should reflect the number of treatments with
  major health gain, rather than just taking
  account of current costs.
• Use of value weights (QALY) instead of cost
  weights.
• Use of mortality rates and waiting times.
• Use of  value weight  for statins prevention
  better than curative care.
• Patient experience.
• None of these has been into account in British NA
  yet. Some proposals can be exclusive of each
  other.

20
New Paris Health accounts meeting,6th of
October, 2006

•  Cost of illness  has been compared among a
  dozen of European countries, and Eurostat find
  the results homogeneous and plausible.
•  Cost of illness  is recommended for
  international comparisons, should be integrated
  in SHA, which would harmonize some conventions
  (fixed costs, prevention, collective purposes)
• NL wants to calculate volume and price with this
  framework (no methodology yet).
• But few data on QALY (main idea for quality
  adjustment).

21
Possible content of the final OECD best
practices manual 

• Synthesis of national experiences and results
• Definitions, principles and terminology (input /
  output / outcome)
• Then by activities (education, health, public
  order and safety ?)
• Definitions, principles, terminology
• Review of (best) stratification and quantity
  indicators
• Review of best quality indicators
• Temporal formulas
• Spatial framework and formulas
• Verification of consistency between market and
  non-market methods
• For September 2007

22
Next steps

• 2 meetings in 2007 for the Eurostat-OECD NA-PPP
  task force on Non Market Services
• ? Meetings in 2007 for the (planned) upcoming
  OECD task force on Health specific PPP
• Permanently in contact with countries to study
  their practices and projects
• Submission of proposals to Health and Education
  OECD networks
• Workshop in Paris, 7-8 June 2007, on the model of
  the London workshop.
• Presentation of draft handbook to the October
  2007 WPNA