Transcript Marianne Fay - Infrastructure and Fiscal Policy Specific

Infrastructure and Fiscal
Policy
Specific Challenges
Marianne Fay
Many thanks to Vivien Foster on whose 2005 PEAM
course presentation this is largely based
Outline




Stylized facts
Funding sources
Budgetary boundaries
Budgetary mechanisms
I. Stylized facts
Infrastructure is “big money”

Stocks




Equivalent to about 100% of developing
country GDP
Dominated by electricity (45-55%), and
transport (30 to 40%)
“cheaper” telecom (5-15% and growing) and
W&S (5 to 15%)
Maintenance:

About 2 to 3% of GDP per year
Infrastructure is “big money”

Output:



About 6% of GDP for electricity (MICs)
The same for telecom & transport?
Much smaller for W&S
Expensive when poorly managed

Electricity:




Total hidden costs (underpricing; technical and
commercial losses) estimated in ECA at 4% of
GDP
In Mexico – untargeted subsidies amount to
1% of GDP
Rail and public transport: historically huge
drains on public coffers
Water and sanitation: typically much
smaller (0.4% of GDP in ECA)
Much of it, public responsibility

Differences across sectors:




Fairly universal trend for privatization
of telecom, air transport, possibly rail
More varied with electricity, public
transport
Limited potential for roads
W&S – complicated…
Distinguishing features

Investment





Highly capital intensive (60%+)
Long term planning horizons (30 yrs)
Infrequent lumpy investments
Long lead times (Up to 5yrs)
Unpredictable investment costs
Distinguishing features

Maintenance




Long asset lives (Up to 30 yrs)
High maintenance (2-3% AV)
Exponential cost of deferred
maintenance
Catch-22

Decision to invest based on estimated
rate of return, itself conditioned by
whether maintenance occurs
The maintenance dilemma
Very
Good
100
90
-Filling Cracks
Good
80
70
-Geotextile and Strengthening
Fair
60
-Reconstruction of the Surface
-Reconstruction of the partial base course
50
Poor
40
30
-Complete Reconstruction
Very
Poor
20
10
0
0
2
4
6
8
10
12
14
16
18
20
22
24
26
If maintenance on a 20 year road is not done by the end of the 12th
year. It starts to deteriorate eight times faster than in the early years
Fiscal consequences


For all of these reasons, ill-suited to
unpredictable annual budgetary cycle
Moreover, particularly vulnerable to budgetary
downturns




Politically soft target for budget cuts
Maintenance less attractive than investment
Long lived assets delay hour of reckoning
Even investments can always be deferred
The infrastructure public finance paradox



Maintenance and investment are prime
candidates for cuts
Subsidies – however poorly targeted - are
difficult to eliminate
Some countries spend more on
consumption subsidies than on either
investment or maintenance…
A complication – no data

No country systematically collects
investment data on infrastructure





“infrastructure” a broad and vague category –
unlike health and education
Poor fit with IMF GFS categories
Public investment data notoriously poor – hard
to distinguish from O&M
A few valiant efforts (Calderon & Serven;
specific country studies)
The implication – no monitoring
II. Funding sources
Only three possible sources


Tax payers (fiscal transfers)
Users:



fees
cross-subsidies
Asset depletion:


quality
non-expansion of service…
Historic under-pricing
Ratio of revenue to costs
1.8
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
Telecoms
Source: WDR 1994
Gas
Power
Water
Cost recovery: water
Median
tariff
Degree of cost recovery
US$/m3
Nil
At least
some O&M
Full O&M and at
least some capital
HIC
0.96
8%
42%
50%
UMIC
0.35
39%
22%
39%
LMIC
0.22
37%
41%
22%
LIC
0.09
88%
9%
3%
Global
0.35
40%
30%
30%
Source: Foster & Yepes, forthcoming
Cost recovery: electricity
Degree of cost recovery
Median tariff
US$/kWh
Nil
At least
some O&M
Full O&M
and at least
some capital
HIC
0.11
0%
17%
83%
UMIC
0.06
0%
71%
29%
LMIC
0.05
27%
50%
23%
LIC
0.05
31%
44%
25%
Global
0.07
15%
44%
41%
Source: Foster & Yepes, forthcoming
Who really gets the subsidy?
government subsidy
inefficiency
UTILITY
tariffs below
cost
inadequate
quality of
service
utility workers
customers
coping
costs
alternative providers
In Hyderabad (India), employees capture 40% of the subsidy, and
consumers 60%, half of which they spend on alternative providers
What distributional incidence?
Omega indicator
Water
subsidies
Electricity
subsidies
Other
welfare
programs
Consumption based
0.56
0.57
1.00
Geographical targeting
1.07
1.10
1.33
Means testing
1.63
1.28
1.55
Self-selection
1.84

1.89
Source: Komives et al., forthcoming
III. Budgetary Boundaries
Budgetary boundaries


Infrastructure has a tendency to
creep off the budget for both good
and bad reasons
There are a number of mechanisms
through which this takes place



Extra-budgetary funds (fuel tax, USL)
State Owned Enterprises
Public Private Partnerships
Earmarked funds


Loved by sectoralistsprovide a stable
source of financing in sectors without
possibility of user fees, isolated from
budgetary and political interference
Loathed by macroeconomistsreduce
budgetary flexibility and optimization of
public resources, often lead to poor
governance, lack of transparency
Tot al energy and t ransport subsidies ($m)
Argentina: exploding funds
2000
Special
Funds
Transport
1800
1600
1400
Special
Funds
Energy
1200
1000
800
National
Budget
Transport
600
400
200
0
1999
2000
Source: Argentina PER, 2003
2001
2002
2003
National
Budget
Energy
Argentina: weak governance
International
best practice
Argentina
transport fund
Legal basis
Established by law
Established and
modified by decree
Governance
Autonomous userrepresented board
Controlled directly by
Min Econ
Accountability Published accounts,
audits
Internal public sector
controls
Resource
allocation
Under continual
modification
Transparent, stable
guidelines
State Owned Enterprises





Often represent a large percentage of public
investment in infrastructure
May or may not be consolidated into fiscal
accounts
May be net contributors or drains on the public
purse
Operate in restricted environments that limit their
autonomy and commercial orientation
Management may be guided by macroeconomic
concerns
Contribution to public surplus/deficit (% GDP)
-0.1
-0.2
-0.3
Source: REDI Colombia, 2004
Ecopetrol
ISAGEN
ISA
EPM
ESSA
Huila
Corelca
Telecom
ETB
Emcali
Colombia: drains vs. cash cows
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
Public Private Partnerships




Potential for PPPs varies substantially across sectors
Key criterion for judging whether extra-budgetary is
extent of risk transfer
However, unless 100% risks can be transferred
contingent liabilities remain
Complex fiscal accounting issues arise regarding the
treatment of



Contingent liabilities
Private investment
Committed future public subsidies
LAC: private relative to total
Ecuador
Colombia
Brazil
Mexico
Bolivia
Chile
Peru
Argentina
Venezuela
0%
20%
40%
60%
Private investment as % total
Source: Calderon, Easterly and Serven, 2003
80%
100%
Private investment as percentage total
LAC: private relative to total
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Telecom
Power
Transport
Source: Calderon, Easterly and Serven, 2003
Water
Colombia: new policy after $4.4 Bn
bailouts

New policy guidelines on risk allocation
between public and private partners

Mandatory estimation of contingent liabilities
using Monte Carlo (continuously updated)

Required payments to cover liability are
‘smoothed out’ into a Deposit Plan

Deposits are made from budget to
Contingency Fund in individual accounts

Aggregate estimates reported annually to
parliament (infrastructure >0.5% GDP)
IV. Budgetary mechanisms
Budgetary challenges



Project selectiondeficiencies in
technical capacity for project evaluation,
plus political attraction of ‘white
elephants’
Multi-year planninglong term projects
required multi-year budget envelope to
assure execution
Implementation bottleneckscomplex
procurement plus unforeseen delays
[cash budgeting!] make it difficult to
execute budget
Peru: project selection

SNIP




Coverage




MinFin unit does (pre-)feasibility studies
CBA methodology with min. IRR 14%
Declares viability without prioritization
2/3 of projects with regulated exceptions
Smaller local projects with domestic financing
Projects supported by Supreme Decree
Too many projects leads to budget
constraints, delays and declining IRRs
Inversion de INVIAS (US$m)
Colombia: low execution
1200
1000
Aprobacion
inicial
800
600
Aprobacion
final
400
200
0
9 5 99 6 99 7 99 8 99 9 00 0 00 1 00 2 00 3
9
1
1
1
1
1
2
2
2
2
Ejecucion
final
Conclusions




Cost structure of infrastructure services
leads to fiscal complications
Wide variety of potential funding sources
for infrastructure
Tendency for infrastructure to be on the
boundaries of the budget
Infrastructure poses challenges from a
budgetary perspective