ministry of finance, timor-leste and presentation structure 3 •trends in infrastructure spending...
TRANSCRIPT
Ministry of Finance and World
Bank
Public Expenditure Review of Infrastructure
(March 2015)
Background and Drafting
2
• MoF decided Terms of Reference for report and employed World Bank to assist with drafting
• MoF has led every step of process including drafting, reviewing and presenting
• Process has followed the new deal principle of government ownership
• This has resulted in a higher quality, more practical and useful report than the World Bank normally writes
• The report proves that Government ownership works and results in a better quality product
Report and Presentation Structure
3
• Trends in infrastructure spending
• Impact on economic growth and fiscal sustainability
1: Macro-fiscal
• Explanation and review of investment management cycle
• Planning, budgeting, procurement etc
2: Investment Management
• Detailed review of electricity, irrigation and road sectors
3: Sectors
4
Part 1: Macro-fiscal Chapter
Infrastructure Spending has Sharply Increased
5
0%
10%
20%
30%
40%
50%
60%
$0
$200
$400
$600
$800
$1,000
$1,200
Budgeted Infrastructure Spending
Actual Infrastructure Spending
Actual Infrastructure Spending % of Total Spending
To One of the Highest Levels in the World
6
0
10
20
30
40
50
60
70
80
Pu
blic
Inve
stm
ent
as a
% o
f N
on-
pet
role
um
GD
P
Past Spending (2008 to 2013) was Heavily Concentrated on Electricity, Roads and Bridges and District
Development
7
896
254
169
265
0
200
400
600
800
1,000
1,200
1,400
1,600
1,800
in $ Million
All Other
District Development
Roads and Bridges
Electricity
Electricity Generation has Increased Compared to Our Peers
8
TL 2013
TL 2008 -
200
400
600
800
1,000
- 1,000 2,000 3,000 4,000 5,000
Wat
ts in
stal
led
/gen
erat
ed p
er
cap
ita
GDP per capita (2011 US$)
9
There Has Also Been An Increase in the Percentage of Households Receiving Electricity
22%
53%
0%
10%
20%
30%
40%
50%
60%
2007 2013
Rec
eivi
ng
Ele
ctri
city
Roads (at least until 2012) Had Not Improved
10
0
20
40
60
80
100
120
2010 2011 2012
% o
f R
oad
s
Fast Two Wheel Drive Slow Two Wheel Drive
Two Wheel Drive with Vehicle Damage Slow 4 Wheel Drive
High Recurrent Costs and Possible Under Budgeting Electricity
11
$0
$50
$100
$150
$200
$250
$300
$350
$400
$450
$500
Actual 2008
Actual 2009
Actual 2010
Actual 2011
Actual 2012
Actual 2013
Budget 2014
Forecast 2015
Forecast 2016
Forecast 2017
Forecast 2018
$ M
illio
n
Operations and Maintenance: Fuel for Generators
Operations and Maintenance: Hera and Betano
Operations and Maintenance: Other
Infrastructure Spending
Economic Growth
12
•Government expenditure sharply increased from 2008 to 2012 • Led to economic growth concentrated in the Government administration and construction sectors •Is taking money out of the petroleum fund (by definition not part of non-petroleum GDP) and increasing Government spending (by definition part of non-petroleum GDP) really creating economic value? •In the long term infrastructure spending increases economic growth when it results in roads, electricity and other outputs which firms use to increase production •This has not yet occurred in Timor-Leste, with manufacturing and agriculture growing slowly between 2008 and 2012 •Unsurprising as there are normally lags between spending, the completion of construction and the availability of the new infrastructure •In addition, there are already indications that some energy intensive industries, such as brewing and cement making, are considering moving to Timor-Leste
Frontloading. But at What Amount?
13
Excess withdrawals
Construct Infrastructure
Economic Growth
Higher revenue and lower spending
Back to the ESI
ESI is Conservative. But not Being Conservative does not Change Things that Much Because Most Oil has
Already Been Mined.
14
Sustainable income with
reference case for oil price and
production
Frontloading Can Work
15
$0
$200
$400
$600
$800
$1,000
$1,200
$1,400
$1,600
$1,800
$2,000
2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
ESI Loans Domestic Revenue Total Expenditure Total Sustainable Revenue
Gap shows excess withdrawals
But Only with Difficult Choices
16
2015
Short Term
2016 to 2018
Medium Term
2019 to 2030
Long Term
Fiscal Envelope with ESI 1.3 1.3 1.5
Fiscal Envelope with FSIIPR 1.3 1.4 1.8
• Total expenditure can only grow very slowly over time • Most of increase in recurrent will be on new operation and maintenance spending • Other recurrent expenditure will be falling in real terms per capita •Domestic revenue has to increase to 15% of non-oil GDP in medium term • Infrastructure fund spending would need to be restricted to less than $370 million a year (this is lower than outer year spending in the 2014 budget)
17
Part 2: Overview of the Investment
Management Cycle
Where are the Bottlenecks? (Under Spending by Cause in 2012 for Infrastructure Fund)
18
$23 $26
$135
$6
$211
$59
$0
$50
$100
$150
$200
$250
Over budgeted or activity paid under
different budget
Design or feasibility study has been
delayed
Procurement of major construction component of the project has been
delayed
Payment Processing
Constraints/Delays
Construction and therefore payment
has proceeded more slowly than
envisaged
Other or not explained
Mill
ion
19
Investment Management Cycle Strengths and Weaknesses
• Timor-Leste has exceeded expectations and its peers in terms of the development of its public investment management system
• Strong preparation of project briefs (MPS and Line Ministries), project appraisals (MPS) and budget preparation process (MPS and DNO)
• Areas of weakness:
– Projects being included in budget despite appraisals showing that they have low rates of economic return (e.g. MPS review of irrigation projects)
– Poor design studies from line ministries being shared with MPS and ADN late in the process and delaying procurement
– Little or no consideration of maintenance costs or activities when selecting projects
– Limited ex-post evaluation of infrastructure projects
20
Part 3: Key Issues from the Electricity, Irrigation and Roads
Sectors
Increased Generation has Led to a Large Yearly Subsidy
21
•In 2012 EDTL generated 237 million KWH electricity. This cost $100.7 million • It revenue was $14.9 million so there was a subsidy of $85.8 million. • This subsidy could be reduced in at least three main ways reducing theft, implementing a new tariff structure and reducing the costs • Reducing theft if EDTL completed the installation of 70,000 pre-paid metres, brought all consumers into its database and stopped all theft then there would be a net saving of $21.7 million •So while metering can reduce the subsidy, unless we reform the tariff structure there is still going to be a significant subsidy
100.7
14.9
85.8
$0
$20
$40
$60
$80
$100
$120
EDTL Costs
EDTL Revenue
Subsidy
$ M
illio
n
Reducing the Subsidy Through Implementing a New Tariff Structure
• The tariff structure for residential users: – Up to 20 KWH per month $0.05 – Above 20 KWH per month $0.12
• This is not progressive enough. Somebody using 21 KWH a month is likely quite poor (only using say four fluorescent lights), but is paying the same tariff as somebody using 2,000 KWH per month who is running an air conditioner all day and is likely fairly rich
• 281 biggest customers consumed more than 16% of electricity* but paid same as all but the smallest consumers
• New tariff structures with high use customers paying more should be considered
22
One Potential New Tariff Structure
23
Tariff
category
Tariff
cents per kilowatt-
hour
% of
customers
Up to 50 kilowatt hours/month 5 21
51 - 1,000 kilowatt hours/month 12 77
1,001-2,000 kilowatt hours/month 15 1.4
2,001-5,000 kilowatt hours/month 30 0.3
Over 5,000 kilowatt hours/month 42 0.2
•The tariff structure below is more progressive •It would lead to additional revenue of $2 million • At low levels of consumption it is more generous
Electricity Subsidy Conclusion
24
• Government has successfully increased generation and access to electricity
• This has led to a large subsidy. However significantly reducing the subsidy is possible by:
– Increasing metering and reducing theft
– Increasing the tariff for households who use a lot of electricity (likely richer households)
– Considering ways to reduce the cost of producing electricity
Ministry of Agriculture has Spent and Plans to Spend Quite a Lot on Large Weir Based Irrigation
Schemes (2013 prices)
25
0
20,000
40,000
60,000
80,000
100,000
120,000 2
00
3
20
04
20
05
20
06
20
07
20
08
20
09
20
10
20
11
20
12
20
13
20
14
20
15
20
16
20
17
20
18
20
19
20
20
S m
illio
n
<--Planned Expend. ($592 million)-->
<---Past Expenditure ($82 million)--->
But to Date this Spending has Not Led to High Economic Rates of Return Or Increased Overall Production
26
Laclo Bebui Maliana I
Economic Internal Rate of Return Negative Negative Negative
Benefit to Cost Ratio 0.18 0.32 0.31
Average 98,500
Average 89,500
0
50,000
100,000
150,000
200,000
2005 2006 2007 2008 2009 2010 2011 2012 2013 Av
Mt
Rice
Maize
Why Haven’t Irrigation Schemes Been Successful
27
• The negative returns on these completed schemes can be explained by the high associated construction costs and their limited impact on rice production levels
• The costs of the schemes are high compared to other countries at $10,000 per hectare.
• The schemes benefit a limited number of farmers and have a limited impact on production because: – There is a lack of financial incentives for farmers to produce
more rice (due to subsidized rice and other sources of income)
– The Ministry of Agriculture does not provide an appropriate package of software. More specifically: • Extension workers are poorly trained • Supply of seed, fertilizer and other chemicals is poor • Limited and/or unreliable market and price for rice
• Going forward spending on large scale weir based irrigation should be decreased, spending on software should be increased and tube well irrigation should be considered
Not all Road Construction/Rehabilitation Outlined in the SDP is Necessary to Become an
Upper-middle Income Country
28
0%
1%
2%
3%
4%
5%
6%
7%
8%
9%
10%
Timor-Leste Current
Timor-Leste 6 Priority Roads Rehabilitated
Timor-Leste All Roads
Rehabilitated
Timor-Leste All Roads
Rehabilitated and Ring Road
Built
Low Income Countries
Lower Middle Income
Countries
Upper Middle Income
Countries
High Income Countries
Ro
ad D
en
sity
29
Average Cost Per KM
$0
$200,000
$400,000
$600,000
$800,000
$1,000,000
$1,200,000
Existing alignment and return to a maintainable
standard
Simple repair and rehabilitation of existing roads to
3.5 metres
Improve alignment where
needed and improve drainage
Rehabilitate to 4.5 metres
Expand to 6 metres and
realign when needed
Expansion to 6 metre width and
upgrade alignment
Expand to 7 metres with major slope adjustment
As Unit Costs Increase Benefit to Costs Ratio Declines
30
0
1
2
3
4
5
6
$150 $250 $350 $450 $550 $600 $700 $800
Be
nef
it/C
ost
Rat
io
Average Unit Cost of Reconstruction
B/C Ratio
Key Recommendations
31
1. Expenditure should be constrained to $1.3 billion a year at most in the short and medium term
2. Budget for operation and maintenance spending for roads and agriculture should be increased. More priority should be given to maintaining and less to constructing
3. Establish a “project bank” of pre-qualified projects with design studies, cost benefit analysis and high economic rates of return. CAFI should choose from bank based on priority and fiscal envelope
4. For all infrastructure projects, line ministries should submit detailed plan and costing of expected operation and maintenance activities. This should be considered when deciding whether to approve new infrastructure spending
5. Ministry of agriculture should maintain existing irrigation schemes and increase farmer support services and market development programs and consider tube-well irrigation
6. Installation of 70,000 additional pre-paid metres should be urgently undertaken and then changes in tariff structure considered