Understanding present and future public service delivery costs
1. Understanding present
and future public
service delivery costs
Ana I More no Monroy (OE CD), Chris Jacobs -C ris ion i
(E C -JRC) and Me rt K om pil (E C -JRC)
E WR Workshop on Unde rst anding pre se nt and fu t ure
pu bl ic se rv ice de l iv e ry cost s in E u rope an re gions
The cyberspace, 1 3 Oct obe r 2 0 2 0
2. The project aims at improving our understanding of:
1. Present costs of public service provision at different
geographical levels
2. Future changes in costs of service provision due to
population age structure and settlement changes
3. The benefits and viability of integrated and flexible models
of service provision
Project aims
3. Pilar 1: Quantitative (in partnership with EC-JRC)
Estimate present and future costs of health and education
services at different geographical scales
Pilar 2: Policy responses
Provide a better understanding on policies to present and future
provision challenges (focus on education, health, and digital
provision)
+ Case Studies/Parallel projects (Alentejo, Spain, Estonia)
Project structure
5. Public service provision is
challenging for rural areas due
to:
✓high transportation costs
✓low economies of scale and
scope
✓difficulty in attracting and
retaining professionals
Currently no available
monetary estimates and
projections of cost of public
service delivery by type of
human settlement
How much more expensive is it to
provide services in rural areas?
6. Estimation of future school delivery using population projections
at grid-cell (1km2) level
✓Demand: Present and future population by age group
✓Supply: Simulated facilities based on optimal access placement
Model costs at facility level considering:
✓Costs arise in the facilities (e.g. schools), not in the areas (e.g. school
districts)
✓Public services are local, and are provided close to places of residency
✓Additional costs arise as a result of smallness
✓Transport costs are not necessarily born by the government
Preparing for future needs
7. ✓Compare school cost across EU countries driven purely by
geographical and settlement structure differences
✓Estimate the impact of changes in teacher compensation and
class sizes across areas and settlement types
✓Compare costs of current school network with efficiently
placed schools
✓Estimate the increase in school cost by settlement type of
future population changes under different policy scenarios
✓Understand the effect of school reorganisation on student
travel times in rural areas
Decision support system for
demographic change
8. 1. Observe spatial
distribution of
relevant service
users (e.g. children in
primary school age)
2. Step by step,
distribute facilities
based on bounding
conditions
3. Allocate users to
facilities (e.g.
children to schools),
measure travel times
and estimate school
expenditure per user
Modelling facility distribution based on
population distribution
9. Estimating school costs
We take advantage of well-
behaved functional forms
to draw likely values for
schools ordered by their
size
Assumed parameters for school expenditure estimation
Primary Secondary
Mean pupil-to-teacher ratio (s.d.) 13 (1) 12 (1)
Mean expenditure on teaching staff per
head (s.d.)
EUR 43 000 (1 000) EUR 55 000 (1 000)
Fixed full-time staff paid at average
expenditure levels
1 2
Percentage TAs in total teaching staff
(paid at half the average expenditure on
teaching staff)
40% 20%
Proportion of non-teaching staff to
teaching staff
1/4 1/4
Expenditure on non-teaching staff per
head
EUR 34 000 EUR 23 000
Remaining mean expenditure per
student
EUR 9 000 (1 000) EUR 20 000 (1 000)
We set parameters to EU averages
where available or based them on
actual school data for England where
no EU equivalent was available
10. Education expenditure in England
0%
10%
20%
30%
40%
50%
60%
70%
80%
-600
-400
-200
0
200
400
600
800
1000
mostly
uninhabited
rural
dispersed rural villages suburbs semi-dense
towns
dense towns cities
Expenditure per student (difference with country mean, GBP) Share of small schools
School size influences cost
per student, e.g. because
small schools tend to
underuse capacities
Source: UK
Department of
education and OECD-
JRC preliminary
estimations
Expenditure per student (mean-centred), England 2019
Expenditure per student vs school size, Actual and simulated
11. In rural areas, schools are farther
Degree of urbanization BE LT PT EU
Mostly uninhabited 11.4 12.2 13.6 12.4
Dispersed rural 7.8 9.8 9.8 9.5
Villages 5.7 5.1 6.0 5.6
Suburbs 4.5 4.9 4.7 4.7
Towns 3.7 2.8 3.7 3.5
Cities 2.5 2.3 2.2 2.5
13. To retain similar efficiency levels,
schools may be even farther!
Calculating costs and travelled distances according to three model variants:
❑ Facility placement optimized for now, student allocation for now
❑ Facility placement optimized for now, student allocation for 15 years from now
❑ Facility placement optimized and student allocation for 15 years from now
Degree of urbanization Average increase (km)Total increase (km)
Mostly uninhabited 0.46 523,814
Dispersed rural 0.27 135,972
Villages 0.24 20,998
Suburbs 0.12 5,418
Towns 0.09 6,585
Cities -0.10 -3,976
15. Delivering quality education in rural
communities
Restructuring
school networks
Consolidating
schools
School clusters
Digital and
comprehensive
approaches
ICT-based
support
Service co-
location
16. Delivering quality health care in rural
communities
Cost
• ↓ patient volumes
• ↓ scale economies
• ↓ scope economies
• ↑ financial incentives
Access
• ↑ unmet medical needs
• ↑ distance to care
• ↓ access to hospital beds
• ↑ hospital closures
• ↑ health workforce gaps
Quality
• ↓ life expectancy
• ↑ survival rates
• ↑ avoidable admissions
and deaths
• ↓ prevention
Income and education levels + ageing
populations + lifestyle factors
Comprehensive
measures
Reinforcing
primary care
Integrated care
Innovative
approaches
Telemedicine
New forms of
rural hospital
organisation
Multi-strategy for workforce
attraction & retention
17. Providing digital services in rural
communities
• Little consideration of
spatial scope in regulation
• Limited funding
Normative
• Substantial gaps
• Insufficient training
uptake
Digital
skills
• Lower speeds and older
technologies
• Fewer options/less value
• Latency, caps…
Internet
access
• Decoupling of activity and place
→ Careful balance of migration
versus efficiency
• Horizontal (broadband) +
targeted
• Integrate education/health care
on digital models
• Substantial + smart funding for
digital upskilling
• Redesign subsidies
• Exploit different competition
models (cooperatives, publicly-
owned entrants, PPPs)
18. Better governance for better service
provision
CHALLENGES POLICY OPTIONS
Spatial spillovers
and externalities
- Earmarked transfers to subnational governments
- Joint delivery through co-operation agreements
- Upscale responsibility
Lack of scale
economies
- Facilitate inter-jurisdictional co-operation or with mergers
- Outsource to private entities or other subnational
governments
Funding,
responsibilities
and capacity issues
- Rely on local revenues to financing their services
- Make funding consistent with functional responsibilities
- Identify and address overlapping responsibilities
- Implement well-designed capacity development
programmes
Wrong scale of
service provision
- Plan delivery in functional regions/areas
- Promote cross border-region projects by facilitating legal
frameworks to enhance collaboration