Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
Growing the Productivity of Government Services
Leandro CarreraSenior Researcher, Pensions Policy Institute
Professor Patrick DunleavyProfessor of Political Science and Public Policy, LSE
Joe GriceChief Economist, Office for National Statistics
LSE Works: LSE Public Policy Group
Suggested hashtag for Twitter users: #LSEworks
Edwin LauHead, Reform of the Public Sector Division, OECD Public Governance and Territorial Development Directorate
Barry QuirkChief Executive, London Borough of Lewisham
Diane CoyleChair
Growing the productivity of government services
2
Professor Patrick DunleavyLondon School of Economics
Government Department and LSE Public Policy Group
Dr Leandro CarreraPensions Policy Institute
Edward Elgar,February 2013
Government sector productivity• Productivity = Outputs/ Inputs
Yet much neglected in the public sector – as withflat productivity’ assumption, based on unmeasurability of government outputs and sustained for national statistics reasons
• Alternatively, ‘productivity’ is much cited – but the concept is inflated to include effectiveness, or made useless by enlargement to mean ‘everything good’
• Yet major progress has been made on measuring and costing outputs, key steps for achieving overall outputs indices
• So in private sector we weight by prices of sold outputs; after Atkinson (2005) in government we weight by costs (or administrative costs)
Problems in measuring public sector productivity
• Productivity is best measured comparatively with many decentralized providers delivering standard services
• Yet decentralization speaks to physical delivery of services, in professionalized and personalized ways –opening up important issues of services quality variations
• Productivity up-growth blips often reflect either unsustainable staff and organizational ‘cramming’ pressures (e.g demand growth, or staff over-cutting) or quality-shading
• So perverse productivity signals are perfectly feasible here. Hence Atkinson recommended quality weighting
• But this is very hard to do well, or continuously.
1. National government agencies
- Customs: trade regulation- HMRC tax collection- DWP social security
Problems in measuring national departments’ productivity
• Unique (within country) departments and agencies, large or very large organizations with minimal internal policy variations
• So no large N datasets, or domestic comparators, ruling out parametric studies and DEA approaches.
• International public management data are also very poor, so cross-national comparison is mostly infeasible.
• Civil servants, politicians and other commentators often dismissive of outputs/inputs measures at national government level – citing the range of agency outputs, strong levels of change in policy (eg new outputs), IT and technology changes, the unmeasurability of ‘quality’, the importance of ‘public value’ and process elements, important government-wide changes, responsiveness to ministers etc.
• Hence historically high levels of resistance to use of measurement –usually via not collecting relevant data, or constantly changing data specifications to prevent any long runs of data
Solutions for measuring national government productivity
• Focus on the over-time evolution of the same department or agency (usually ‘immortal’)
• Focus on departments with relatively standard operations, where fine-gauge quality variations don’t make much difference at the aggregate levels
• Use a ‘standard quality’ constraint – non-comparabilities arise only if quality dips badly. Internalize most policy effectiveness change or churn, or general civil society advances in IT and ‘point of service’ standards
• Develop a detailed narrative for each agency with fine-grain process-tracing of productivity movements to specific policy shifts, organizational developments, reorganizations, etc. = organizational productivity story
0
20
40
60
80
100
120
140
160
180
97/98 98/99 99/00 00/01 01/02 02/03 03/04 04/05 05/06 06/07 07/08
Financial Years
Volume (2001/02=100)
Volume of Output Volume of Input Productivity
Total Factor Productivity in the Customs regulation of trade, 1997-2008
Figure 3: Labour and Intermediate Inputs Productivi ty
40
60
80
100
120
140
160
1997-98
1998-99
1999-00
2000-01
2001-02
2002-03
2003-04
2004-05
2005-06
2006-07
2007-08
Financial Years
Volume (01/02=100)
Volume ofTax Output
Volume ofLabour andIntermediateInputsProductivity
Labour and intermediate inputs productivity in UK t axation, 1997 to 2008, using tax collection activity data
40
50
60
70
80
90
100
110
120
130
140
1997-98
1998-99
1999-00
2000-01
2001-02
2002-03
2003-04
2004-05
2005-06
2006-07
2007-08
Financial Years
Volume (2001/02=100)
Index of Total TaxCollected
Index of Labourand IntermediateInputsRatio Taxcollected to Inputs
The ratio of the deflated amount of tax collected t o labour and intermediate inputs,
for HMRC and predecessor departments, 1997-2008
50
60
70
80
90
100
110
120
130
140
97/98 98/99 99/00 00/01 01/02 02/03 03/04 04/05 05/06 06/07 07/08Financial Year
Volume (99/00=100)
SS Index of Output SS Total Expenditure Index SS Total Factor Productivity
Total Factor Productivity in UK social security, 1997 to 2008
40
60
80
100
120
140
160
180
200
87/8
888
/89
89/9
090
/91
91/9
292
/93
93/9
494
/95
95/9
696
/97
97/9
898
/99
99/0
000
/01
01/0
202
/03
03/0
404
/05
05/0
606
/07
07/0
8
Financial Year
Volume (1988=100)
Index of Input Index of Output Total Factor Productivity
Longer-term estimates of changes in total factor productivity for UK ‘social protection’ serv ices,
from 1987 to 2008
0
20
40
60
80
100
120
140
160M
illion
s of
cus
tom
er c
onta
cts
Phone Post Face to face Online
2005
2008
The changing pattern of the DWP’s customer contacts , 2005 to 2008
Indicative overview on
2. The role of IT andwider management changes
y = 4.4184x + 69.18
R² = 0.459
60
80
100
120
140
160
180
0 5 10 15 20
ICT (as % of total admin costs)
Pro
du
ctiv
ity
Productivity versus lagged ICT spending across DWP, HMRC (tax), and Customs for 1999-2008
y = 5.2608x + 78.09
R² = 0.3817
60
80
100
120
140
160
180
0 2 4 6 8 10 12
PFI (as % of total admin costs)
Pro
du
ctiv
ity
Productivity versus lagged construction PFI spendin g across DWP, HMRC (tax), and Customs for 1999-2008
y = -1.2638x + 116.11
R2 = 0.0098
60
80
100
120
140
160
180
0 1 2 3 4 5 6 7 8Consulting (as % of total admin costs)
Productivity versus lagged consulting spending across DWP, HMRC (tax), Customs for 1999-2008
Five key steps to sustainable public sector productivity growth
1. Focus hard and continuously on productivity growth, countering adverse relative price effects
2. Recognize that public sector innovation is twice as vital for productivity growth as in the private sector
3. Engage public sector workers in facilitating changes, maximizing information revelation by creating trust in management’s non-opportunism
4. Encourage genuine demand transfers across suppliers (e.g intra-governmental competition and mixed economy models) can play a small role
5. Reduce public and political support for those ‘big state’ routes to reducing social inequality that are no longer working well – difficult to do when social inequality is increasing
Thank you for listening
50
100
150
200
250
300
350
99/00 00/01 01/02 02/03 03/04 04/05 05/06 06/07 07/08 08/09
Volume (99-00=100)
Volume of Output (UC Adj) Volume of Total Input Productivity
Total factor productivity in passport issuing
40
60
80
100
120
140
160
180
02/03 03/04 04/05 05/06 06/07 07/08 08/09 09/10
Volume (02/03=100)
Total Expenditure Productivity Volume of Total Input Volume of Total Output
Total factor productivity in DVLA
3. Decentralized agencies’ productivity - the National Health Service
Inpatients treated and outpatients appointments
Cost and Quality
Medical Staff Headcount Input
Output
Labour Productivity
Adjusted for
Essential model for NHS productivity
Key innovations or differences
• Developed quality adjustment as well as cost weighting of outputs
• New independent variables measured via web censuses of trusts covering– Management practices– Use of IT
Independent VariableModel 1
(dep var = cost adjusted labour productivity)
Model 5(dep var = Full cost and
quality adjustedlabour productivity)
ICT Use 27.15 (24.90) 35.709** (18.414)
Management Practices 11.10 (10.45) 12.867* (7.767)
ICT x Management -1.33 (1.03) -1.58** (0.766)
General Training hospital 6.62 (11.63) 12.187 (8.869)
Specialist hospital -6.57 (36.80) 109.634*** (32.184)
Teaching hospital -53.57 (65.19) 48.229 (50.709)
London -71.86*** (35.30) -80.287*** (26.733)
R2 0.09 0.16
N 160 147
Select OLS regression results for productivity across English NHS acute hospital trusts
The conditional effect of IT use, given management practices
The conditional effect of management practices, given IT use
- knowledge recognition
- knowledge capture, collection, storage
- institutional memory,knowledge re-access
4. Organizational learning motivation
and systems
5. Single-loop learning –
about productivity
and efficiency
11. Humanresource
management- practices
and systems
6. Double-looplearning -
abouteffectiveness
7. Triple-loop learning – Strategic
leadership and Ministers’ values
10. Innovation
8. Organizationalunlearning
9. Policy andOrganizational
crises
1. Organizational culture
3. Organizational learning
12. Political process
2. KnowledgemanagementRe-learning loop
Re-learning loop
External influences External influencesExternal influences
Situating organizational learning in government sector organizations within external influences
Growing the Productivity of Government Services
Leandro CarreraSenior Researcher, Pensions Policy Institute
Professor Patrick DunleavyProfessor of Political Science and Public Policy, LSE
Joe GriceChief Economist, Office for National Statistics
LSE Works: LSE Public Policy Group
Suggested hashtag for Twitter users: #LSEworks
Edwin LauHead, Reform of the Public Sector Division, OECD Public Governance and Territorial Development Directorate
Barry QuirkChief Executive, London Borough of Lewisham
Diane CoyleChair