Non-Market Services: What can we measure?

1
Non-Market Services What can we measure?

• Mary OMahony
• NIESR

2

• Non-market services
• Define as services where all or most production
  is by the public sector (extent of private
  provision varies across countries, e.g. US health
  sector)
• Comprises health social services, education,
  criminal justice, defense, public administration
• 20 employment EU-15 25 US and growing

3
Measurement Developments traditional

• National accounts traditionally outputs were
  measured by inputs
• Sector performance providers need some measures
  to evaluate effectiveness of service delivery
• Frequently used performance indicators, e.g.
  number of pupils reaching some education level,
  decreases in mortality rates from heart disease,
  number of street lights repaired within 48 hours
• Problems how to aggregate, mixing up outputs
  and outcomes

4
Measurement Developments - Research

• Dissatisfaction with state of play - Alternative
  measures being pursued by statistical offices
  (Eurostat targets, 2006)
• Require measures that can be used both for
  national accounts and performance evaluation
• Past decade or so - researchers looking at
  performance measures, drawing from methods
  employed in private sector
• Education Jorgenson Fraumeni
• Health US research, Berndt, Cutler, Griliches,
  Newhouse, Triplett
• Some research input from National Statistical
  Offices, e.g. US Bureau of economic analysis, UK
  Office for National Statistics, Australian Bureau
  of Statistics, Statistics Netherlands.
• There exist plentiful data on quantities main
  issue is to incorporate quality changes

5

• Why are non-market services different
• 1. lack of prices
• 2. information asymmetries
• 3. some collective services
• Is an issue for public provision
• is an issue for both public and private
  information is in the hands of professionals,
  insurance markets, difficult to observe benefit
  to consumer
• Probably will continue to measure outputs by
  inputs difficult to define outputs independent
  of inputs
• Solution for 1 use cost weighted activity
  indexes.
• Solution for 1 2 is to consider marginal impact
  of services on outcomes - life expectancy, QALYs,
  lifetime earnings

6
Illustrations Research on non-market
serviceshttp//www.niesr.ac.uk/aim/aim.htm

• EDUCATION
• BEA (Fraumeni et al.) cost weighted pupil
  numbers (FTE) for schools, quality adjustments
  include teacher quality (labour compensation
  index), class size, high school dropouts, college
  enrollment
• Netherlands pupil numbers, foreign, pupils
  moving up
• ABS third level, adjustment for cost of degree
  courses
• NIESR - US/UK comparison - marginal outcome is
  impact of education on market earnings
  regression analysis to compute returns to
  education increments to earnings from education
  used as weights for student numbers, includes
  university and further education

7
Illustrations Research on non-market services

• HEALTH
• ONS cost weighted activity index new measure,
  about 1300 activities replaced old index
  based on 16 categories treated all in-patient
  activities the same e.g. heart by-pass equals
  varicose vein procedure
• ABS similar index
• Productivity growth in the NHS (NIESR/Centre for
  Health Economics, University of York), activities
  (same as ONS), outcomes - increase in quality of
  life following treatment, waiting times,
  mortality rates report due end August 2005

8
UK CWAI for Health
9
Non-market services in EUKLEMS

• Statistical module
• Probably best to confine attention to cost
  weighted activity indexes
• National statistics offices engaged in research
  for Eurostat target need to follow developments
• Quality adjustments differ between countries
  do we attempt to achieve consistency?
• Analytical module
• Incorporate outcome information