Branding

1
(No Transcript)
2
The use of business micro data in policy advice

• Dr. Ulrike Hotopp
• DTI, Strategic Policy Analysis Unit

3
Content of this talk

• Introduction
• Why evidence matters The DTI governance system
• The use of evaluation in policy making
• Examples
• Building a counterfactual
• Research
• On business micro data
• Policy research using business micro data
• Conclusion and plea

4

• 2. Why evidence matters
• The DTI governance system

5
Two Reviews in DTI

• 1. Review of the DTI and its governance structure
  in 2002
• DTI Strategy
• DTI Business Plan
• 2. Review of Business Support programme.

Focus on evidence based policy
6
DTI Main Committees
7
Investment Committee Structure and Membership

• Industrial Development Advisory Board type
  structure (i.e. no officials as voting members)
• External members

• Officials will also attend, including DTIs Chief
  Economist, Chief Executive of the SBS, Directors
  General/Directors from within DTI and a
  representative from Treasury

Typically the Committee meets 5 6 times a year
8
The Investment Committee - remit

• Strategic allocation of business support funds.
• Funding of all significant business support
  proposals.
• Funding for smaller proposals that are pilots
  that could lead to proposals with significant
  expenditure implications.
• To improve the monitoring and evaluation of
  performance of the different support schemes.
• The collection and management of core evidence to
  inform future policy and investment decisions.
• Monitoring the progress of the investments
  throughout the year, and receive formal
  evaluation reports on the performance of the
  overall portfolio and individual products on a
  regular basis.

9
Performance Monitoring Committee Remit

• To appraise
• emerging objective delivery plans and business
  cases,
• especially proposed measures, targets and risk
  management plans.
• To report quarterly to the Executive Board,
• ensuring that underperformance is investigated,
• corrective action is planned and followed
  through.
• To set the Evaluation Programme agenda

10
The policy making process
11

• 3. The use of evaluation in policy making
• Examples

12
Examples of the use of business micro data in
policy making

• DTI Business support
• Evaluation of RSA (Regional Selective Assistance)
  
• and
• SMART (Small Firm Merit Awards for Research and
  Technology)
• National Minimum Wage

13
Evaluation of RSA and SMART R. Harris and C.
Robinson 2001

• Data Matching
• SAMIS and the ARD and the IDBR
• RSA plants are larger, concentrated in certain
  industries, more likely foreign-owned and/or
  branch plants
• Using the extended ARD it was possible to
  estimate the Haltiwanger-type decomposition
  model

14
Production function approach

• The policy-off/policy-on model
• allows for panel effects
• Note, data need to be
• Plant based
• weighted

15
Empirical Conclusions

• RSA plants had lower TFP - some 4 per cent less
  than the average for all plants but little
  difference when looking at just assisted areas
  (except Scotland, Wales, and North West)
• Following assistance, RSA plants improve TFP by
  on average 3 per cent (compared to all GB plants
  the result for assisted area plants only is not
  statistically significant - except Scotland).
• SMART-assisted plants were better than the
  average (some 6.4 per cent more productive), but
• post-assistance there was little significant
  change in TFP (although the results for just
  those plants located in assisted areas are
  positive and significant at the 8 per cent
  significance level under a 1-sided t-test).

16
Policy issues

• Entry and exit are important in terms of
  increasing productivity
• Government policy needs to ensure it does not
  adversely affect either/both
• The impacts of RSA in terms of TFP suggest the
  scheme has limited ability to improve
  productivity in the Assisted Areas
• SMART does better

17
National Minimum Wage Low Pay Commission
Fernando Galindo-Rueda, and Sonia Pereira

• Introduced in 1999 at 3.60 per hour for adults

18
Data sets and variables

• New Earnings Survey
• Annual Business Inquiry
• Employment
• Wages
• Labour Productivity
• Total Factor Productivity
• Manufacturing and services

19
Methodology

• First approach
• Comparing the evolution of those variables in
  firms more likely to be affected by the1999 MW
  introduction with their evolution in firms less
  likely to be affected.
• Second approach
• Using differences within sectors by taking into
  account how a common minimum wage was likely to
  have a higher impact in regions where pay is
  lower.
• Estimate separate trends by region and industry.

20
Empirical conclusions

• NMW had some positive one-off impact on labour
  productivity but not much of an impact in terms
  of total factor productivity.
• Using the available industry and region level
  data to identify the sensitivity to the NMW. A
  firm may be in a low wage sector but in a high
  wage region and will therefore be less affected
  by the NMW than a firm of the same sector in a
  low wage region.
• Some evidence for lower (but positive) employment
  growth in sectors and regions more exposed to the
  NMW than in others.

21

• 3. The use of evaluation in policy making
• Building a counterfactual

22
Building the counterfactual

• Business micro data help establishing the
  counterfactual
• Policy-on-policy-off
• Treated versus non-treated
• Treated versus failed applicants
• Before and after

23

• 4. Research
• On business micro data

24
Research on data linking Prof. Andrew Chesher and
Lars Nesheim Research objectives

• 1. The impact of contributing survey designs and
  non-response on achieved linked survey design and
  implications for inference.
• 2. Measurement error issues arising because of
  imputation of low level (e.g. plant) values using
  high level (e.g. business) values.
• 3. The impact of excluding unmatched units in
  linked survey datasets.

25
Research on data linking Prof. Andrew Chesher and
Lars Nesheim Research objectives ctd

• 4. The impact of including erroneously matched
  units in linked survey datasets.
• 5. The consequences of linking when matching is
  done on values of salient variables and
  contributing surveys have no common responding
  units.
• Recommendations implemented within ONS New
  guidance for data lab users

26

• 4. Research
• Using business micro data

27
Research using business micro data to explore
policy optionsExample seed funding 2004/05

• Richard Harris Matching Plant-level data from
  the ABI/ARD with the Employee Skills Surveys,
  1999 2001
• Ralf Martin Using the ARD and related datasets
  for research on energy efficiency and pollution
  reduction
• Pari Patel Adoption of e-commerce technology. Do
  network and learning externalities matter?
• Peter Wright Job reallocation and labour market
  adjustment A proposal to analyse linked employer
  employee data for the UK

28
Ralf Martin Using the ARD and related datasets
for research on energy efficiency and pollution
reduction

• Data Annual Respondents Database (ARD) and the
  Quarterly Fuels Inquiry (QFI).
• Main results
• Energy efficiency spreads across plants are large
  and persistent.
• Considerable CO2 reduction potential.
• Entry and exit less important for energy
  efficiency growth than for labour productivity
  growth.

29
Ralf Martin Using the ARD and related datasets
for research on energy efficiency and pollution
reduction ctd

• Electricity price spreads are large although not
  as dramatic as either energy efficiency or labour
  productivity spreads. Spreads are fairly constant
  over time and across sectors although there is
  some weak evidence that the regulatory changes in
  2001 had some impact on spreads.
• Strong relationship between quantity and
  electricity prices.
• Significant differences in electricity prices
  across regions

30
Conclusion and plea

• Evidence matters
• DTI structure allows for its use in policy making
• Many examples in all areas of policy making
• Presentation, presentation, presentation

31
(No Transcript)
32
Evaluation of RSA

• Total factor productivity
• Labour productivity

33
DTI Structure
Permanent Secretary Sir Robin Young
UK Trade Investment
OST Chief Scientific Adviser Sir David King DG
Research Councils Sir Keith ONions
Strategy Unit Geoff Dart Chief Economist Vicky
Pryce
Legal Services Anthony Inglese
Services Group Catherine Bell
Fair Markets Stephen Haddrill European
World Trade policy Jo Durning
Innovation Group David Hughes Facilitating
Innovation John Rhodes Technological Innovation,
Sustainable Development (Vacant) British National
Space Centre Colin Hicks Patent Office Ron
Marchant
Business Group Mark Gibson Business Support David
Saunders Business Relations John Alty David
Hendon SBS
Energy Group Joan MacNaughton
34
Balanced Scorecard for Collaborative RD