philippines lng: developing new import markets• barriers to entry and possible options to manage...
TRANSCRIPT
Philippines LNG: Developing New Import Markets Sarah Fairhurst 3rd Annual FSRU Conference, Singapore, November 26 2014
Prepared by:
Sarah Fairhurst
Agenda
• Understanding Philippines’s energy needs and government policy for LNG imports
• Outlining possible terminal projects: Who might win the race to build the terminal?
• Barriers to entry and possible options to manage them
1
Understanding the Philippines Energy Needs: Current Situation
• Fuel mix is currently very diversified with a
high proportion of renewable and local
sources
• Gas makes up 25% of the fuel for power
generation in the country – higher if you just
look at Luzon
• Practically all the gas comes from
Malampaya – a single gas field offshore
Palawan via long (504km) undersea pipeline
• The Malampaya concession expires in 2024
and while it may have enough gas for some
further expansion, this is not considered
sufficient for more than about 5 years at
current levels
2
After Malampaya – what then?
Coal 43%
Oil 6%
Natural Gas 25%
Geo-thermal
13%
Hydro 13%
Wind 0.1%
Solar 0.002% Biomass
0.3%
2013 Generation Mix
Source: Philippine DOE (Power Statistics)
And although there is a lot of gas currently burnt, much of it is not economic
• The three gas plants all currently run baseload
• However, they ONLY run baseload because of the
“take or pay” (ToP) constraints of the contract for
the sale of gas
• Without these ToP constraints, economically the
gas-fired plants would run less – acting as peaking
or mid-merit plants
Between July 2007 and the end of 2013,
Meralco alone spent an extra US$300m on
its gas-fired IPPs compared to the cost
saving it could have got from its coal plants
3
3833
27
9
2012
without
ToP
48
2012
actual
42
80
60
40
20
0 Geothermal
Coal
Natural Gas
Oil-based
Hydro
Wind
Biomass
2017
est
without
ToP
74
2017
est
100
57
Luzon generation fuel mix
Percent
Source: TLG analysis
The Philippines has diverted a lot of money from electricity into other things since the
completion of Malampaya…..
The Philippines has an “event-driven” electricity sector – much uncertainty and
variability for both natural and technical reasons
Note: * Buying price (with 100% surplus)
Source: PEMC; TLG analysis
PhP/kWh
Average monthly WESM spot settlement price* (2007-14)
4
0
2
4
6
8
10
12
14
16
18
20
22
24
26
Jan-14 Jul-13 Jan-13 Jul-14 Jul-12 Jan-12 Jul-11 Jan-11 Jul-10 Jan-10 Jul-09 Jan-09 Jul-08 Jan-08 Jul-07 Jan-07 Jan-15
Low
hydro
Gas curtailments
and low hydro
Plant forced maintenance
outages and HVDC link
maintenance
Transformer breakdown and San
Jose substation congestion
Plant outages and
coal limitations
Plant outages
Gas curtailments
and plant outages
Gas scheduled outage
and plant outages
ERC-
regulated
price
reduction**
Weather effects vary widely each year – the system must have the capability to
deal with strong swings between El Nino and La Nina
5
Deg. C
NOAA Oceanic Niño Index (ONI)
Source: NOAA
(2.5)
(2.0)
(1.5)
(1.0)
(0.5)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Jan-5
0
Oct-
51
Jul-5
3
Apr-
55
Jan-5
7
Oct-
58
Jul-6
0
Apr-
62
Jan-6
4
Oct-
65
Jul-6
7
Apr-
69
Jan-7
1
Oct-
72
Jul-7
4
Apr-
76
Jan-7
8
Oct-
79
Jul-8
1
Apr-
83
Jan-8
5
Oct-
86
Jul-8
8
Apr-
90
Jan-9
2
Oct-
93
Jul-9
5
Apr-
97
Jan-9
9
Oct-
00
Jul-0
2
Apr-
04
Jan-0
6
Oct-
07
Jul-0
9
Apr-
11
Jan-1
3
ONI measures average surface sea temperatures and it is used by the US
National Oceanic and Atmospheric Administration (NOAA) for their
operational definition of El Nino and La Nina
El-Nino
La Nina
And of course hydro is not the same each day, week, month, season, or year
6
13
2
1
0
5.2
12
5.3
11
5.5
08
4.4
10
4.0
09
5.5
05
4.3
04
4.3
03
5.4
07
4.6
06
4
3 2.9
2002
3.8
6
5
Source: DOE Power Statistics (2002-13); PEMC; TLG analysis; NOAA
Annual hydro generation in Luzon
(2002-13)
0
5
10
15
20
25
30
0
5
10
15
20
25
30
600
400
200
0
D N O S M F J A J J M A
800
Monthly hydro generation in Luzon and
2011 precipitation (2006-2012)
Hydro generation is
low in Q2 when
demand is at its
peak
Low inflows in severe El-
Nino years reduce ability
of hydro plants to
generate
TWh (bars) GWh (lines) mm (bars)
2008
2007
2006
2011 precipitation
(RHS)
2012
2011
2010
2009
Generation peaks
when there is
plentiful inflows
As a consequence of all these things – timely, flexible, capacity is needed
7
0
100
200
300
400
500
600
700
2/3/14 2/2/14 2/1/14 1/31/14 1/30/14 1/29/14 1/28/14 1/27/14
Sual G1
Mariveles G2
Pagbilao G1
Typical weekly generation schedule (w/c January 27, 2014)
MW
Source: PEMC (ex-post schedules)
COAL PLANTS
And most flexibility in the market is currently provided by coal – which is not ideal
New power stations running LNG would meet this need but building power
stations in the Philippines is more complicated than many places in Asia…
• There is no national Government owned electricity “utility” in the Philippines
• There are no Government backed long term PPA’s on offer and no Government Guarantees
– The electricity legislation (EPIRA) prevents the Government from building or contracting for power stations
(except in emergency situations)
• There are two electricity markets – the WESM (Luzon and Visayas) and the IMEM (Mindanao)
• There is a degree of open access and retail competition – meaning some loads are contestable
• There remains extensive regulation of electricity contracts between generators and retailers
– Not all of which is well designed or well implemented
So an investor in power in Philippines has to manage commercial and regulatory risks
through private sector investment and debt. Add to that the costs of building an LNG
terminal as well…
8
Is it even possible?
So how much LNG could be economic in the power sector?
• This is a question we answered recently as part of our work on the Gas Master Plan for the
Philippines
• To answer the question, we modelled the electricity market using our electricity market simulation
tool (QUAFU) to identify:
– What type of plant is is most economic to build in the future in the Philippines given the current plant mix,
future load projections and costs of different type of new build options
– How would existing gas-fired plant operate after their existing gas contracts expire
– Are there any other ways LNG could enhance the system, without adding to system costs?
9
We showed that there was a case for LNG in the least-cost capacity expansion
plan for the WESM and that this ran mid-merit
10
Least-cost capacity expansion plan for Luzon
under expected assumptions
0
200
400
600
800
1,000
1,200
29 27 25 23 21 19 17 15 2013
Net MW
Economic least-cost entry
Natural Gas
Biofuel
Wind
Geothermal
Hydro
Oil
Coal
Note: EWC plant assumed to not be committed
Source: DOE (committed plants as of Aug 2013); TLG analysis
• Near-term need for more cost
effective mid-merit / peaking capacity
• Conservatively, an LNG-fired CCGT
of about 600-800 MW and a LNG
import terminal appear to be the
least-cost option
• They are able to recover reasonable
returns on their invested capital
(including for terminal)
• The amount of LNG-fired capacity
required is relatively robust to near-
term committed capacity
• However, at expected coal and LNG
prices, coal generation will continue
to be least-cost option for most of
future capacity requirement
Uncertainty over
how further supplies
from Malampaya
could be used?
In the longer term, existing plants would run mid-merit rather than as baseload
plant if take-or-pay constraints are eased
• As we noted earlier, the Philippines already has 2700MW of gas-fired capacity which runs
baseload, out of merit
• Between 2022 and 2024, the gas contracts forcing this plant to run baseload expire
• When that happens, the existing plant will be able to operate economically in a mid-merit role
• This means that no further gas-fired capacity is needed in the WESM
11
There is a current window of opportunity for some additional mid-merit plant, but LNG
fired power does not require investments in large amounts of plant
In addition to building new power stations, there is also an opportunity to import
LNG to back up the existing domestic gas
12
Start date Duration
Estimated additional system cost (mPhP)
Scheduled Malampaya maintenance outage
22 Nov 2006 25 days 2,500
27 Jun 2008 4 days 1,000
10 Feb 2010 30 days 1,300
20 Oct 2011 7 days 900
13 Jul 2012 8 days 600
11 Nov 2013 30 days 4,000
Subtotal: 10,300
Unscheduled Malampaya supply outages
Average curtailment of 1,700MW over about 287
hours since 2006 740
Total: 11,040
• Imported LNG is more cost
effective than liquids, meaning
that having the option to back-
up Malampaya with LNG is a
valuable option
• If LNG is used to replace liquid
fuels when Malampaya is on
outage, an opportunity to save
in the order of USD20-25
million per annum exists
• Reduced availability also
means more expensive
generation is needed to
replace lost capacity
Note: * Used in scheduled maintenance outages
Source: DOE; TLG analysis
This is a conservative value – if one took into account the impact on the market price
as well it may be higher
And in the longer term, there are additional markets for gas
13
LNG would result in considerable
savings over LPG and diesel as a
heating fuel
And also as a transport fuel 0
5
10
15
20
25
30
35
US
D/m
mb
tu
0.0
5.0
10.0
15.0
20.0
25.0
30.0
35.0
US
D/m
mb
tu
Our base case scenario has growing demand from power, industry and
transport
• In the short term, we believe it is more economic to
move LNG by truck to industry than by pipeline
• There are a number of diesel fired power stations
in Visayas and Mindanao that could be converted
to run on LNG – most of these are proximate to the
coast and could be serviced by small barges on a
“milk run” from the main Luzon terminal
• Development of gas pipelines in the Philippines is
problematic due to access to land – over time if
this is solved and after a base industrial demand
has grown, then additional pipelines into Manila
make sense
14
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
2017 2020 2025 2030
mtpa
Power
Industry
Transport
Forecast LNG consumption
So, there IS commercial potential for LNG in the Philippines, but not a huge
amount
• Most of the potential LNG demand comes from the power sector:
– To back up existing power stations now; or fuel existing power stations after 2024
– Fuel for new power stations from 2017 onwards
• Backing up existing power stations is worth about USD 20-25 million per annum until 2024
• There is an economic case for about 600-800 MW of new power stations, running mid-merit
– Highly variable gas demand due to variations in annual rainfall (hydro), outages, and weather (El Nino)
meaning quantities of LNG required are significantly less than 1mtpa in the near term
– No economic case for new baseload gas fired power stations – coal is always a cheaper option for
baseload, even after factoring in possible carbon credits
• Small but potentially growing demand for gas in industry and transport
15
Unfortunately our findings do not sit well with recent “policy” … developed
independently of this work
• The Energy Secretary developed a Fuel Mix policy at the same time we were running this project,
which states that 30% of the electricity in the Philippines will come from gas.
• It appears to roughly take the fuel mix in 2008 (the year the RE law was passed) and expands it
out into the future
• However, appears to be a political desire with no economics nor analytics underpinning in.
• Nor any active policy either (no mandatory actions required)
16
The recommendations of the Gas Master Plan are out of line with the “new” Fuel Mix
policy – which is causing confusion in the DOE
So – what will actually happen?
• The Natural Gas Master Plan proposed a specific transaction structure to bring in a privately
owned FSRU to deliver back up LNG and supply for the economic amount of new power stations
• It required only a small amount of Government assistance to get the process started… but so far,
there are no indications that the DOE will act on any of the recommendations
– DOE is understaffed and much focus is currently on the “emergency” power for next year
– A number of internal movements within DOE mean that staff who were in charge of this project have now
moved elsewhere
• However, all is not lost.
• There are a number of initiatives from the private sector that may deliver very similar outcomes
17
A year ago a number of private sector proponents were active in this space….
• At the start of our Master Plan study, we interviewed all the active proponents looking to build
LNG terminals in Philippines
– Most were looking at land-based terminals
– Most were looking at “base load” options for gas power plants and off-take
– Few really understood the market
• Our study recommended an FSRU instead of a land-based terminal
– The amount of gas required is small, meaning expensive land-based terminals are unnecessary
– Flexibility is important – timing may change – faster options are more appropriate
– Building on land is problematic in Philippines – land acquisition and approvals delay projects so a floating
solution has advantages
18
…however now the credible suspects have been narrowed down
• EWC is under construction at Pagbilao
– Land-based terminal using “marine-based” technology to keep costs down
– Significant progress has been made on the terminal and turbines have arrived on site for the power station
– Construction started before the project was fully funded, making it hard to predict how it would outturn
• Shell has been undertaking a FEED study for LNG import at Batangas
– Their JV SPEX is currently the sole supplier of gas to the Philippines, with a depleting field, giving Shell a
clear economic incentive to find a way to continue supplying gas to the Philippines market
• First Gen has also been looking at LNG options
– First Gen owns two of the three gas-fired power stations currently operating and is building two more – a
peaking plant and another mid-merit plant
– Those plants will initially share gas from the existing gas supply, but this is only a short term solution so
First Gen have a clear economic driver to find alternative supplies
• Meralco was previously looking at a terminal but now appears to be just looking at a gas-fired
power station
– As the largest distribution utility in the Philippines, it makes a very credible and credit-worthy counter party
for any importer of LNG
19
Any of the options could supply the economic need for LNG-fired capacity in
the power sector
• EWC’s project is around 600MW and currently has no power contracts - meaning that it would
operate as required in the WESM exactly as our model predicted
• First Gen’s projects take existing inflexible gas and remove the take-or-pay to allow much more
flexible and economic burn of the existing Malampaya gas – but all these power stations will
need fuel in the future after Malampaya
• Meralco has long argued that flexible, mid-merit power is what is needed in the Philippines
electricity market
20
All these projects are consistent with the economic requirement of the power sector
and all are private-sector driven
But barriers to entry still remain
• There are no environmental incentives for gas-based generation in the Philippines
• Regulation of power supply contracts continues to undermine efficient outcomes
– Contract regulation is currently based on a “cost-plus” basis that does not take account of market prices
nor what alternative contracts might be available.
– There are no incentives for distribution utilities to procure an overall least cost portfolio
– As such, it makes it harder to highlight how mid-merit and peaking generation options fit into the mix
compared to “cheaper” baseload coal.
• Government policy and intervention in the electricity market is undermining private sector
incentives
– Interference in the market to lower market prices “after the fact” has caused some generators to lose
money
– Lowering the cap on the market price and adding a new “secondary” price cap at a level lower than a
peaking plant needs to recover its costs has serious implications for new investment
• There remains no clarity of rules on NG use and infrastructure
21
Lack of regulatory incentives for retailers to contract are possible one of the
biggest barriers to entry of new flexible plant
• Retailers are the companies who buy from the wholesale market and sell to end use customers
• In the Philippines these are distribution utilities and co-operatives servicing franchise consumers
(as well as a small number of retailers servicing the contestable market)
• The purchases of these retailers are regulated by the ERC, but this regulation is unlike that found
in most markets
• The regulation is focussed solely on individual contracts
– It does no analysis of whether the contract is actually NEEDED, only the cost of the contract
– It does not take account overall purchases by the retailer
– It does not take into account alternatives available
– And nobody regulates what is not contracted (that is, purchases from the spot market)
• Once approved, all the costs of these contracts are passed directly through to the consumer,
even if they later change, even if the contract is not needed
22
The impact means that retailers focus on what is easy to get approved, not on what is
actually needed. They have no incentives to contract efficiently
In any market, or to meet any load shape of electricity demand, there is a mix
of plants that is cheapest
23
Oil Gas Coal Oil Gas Coal
0
50
100
150
200
250
300
Fixed costs - $/kW per year Variable costs - $/MWh
0
20
40
60
80
100
120
140
160
180
Capital
Fixed O&M
Fuel
Variable O&M
IT’S BECAUSE
YOU WANT TO RUN IT A LOT
IF YOU SPEND ALL THIS
MONEY TO BUILD IT
No single technology is inherently “better” than the other – it is the way they are mixed
that makes the optimal solution
In a market like the Philippines, we would expect retailers to purchase a mix of
peak, baseload and mid-merit capacity to meet their needs
24
Peaking plant – such as diesel or hydro
Mid-merit (flexible) plant – such as CCGT or recips
Baseload (always on) plant – such as coal, geothermal
This is important because it is the incentives on retailers to contract that drives the
contract underpinning new entry into a merchant market
LO
AD
Hours in the year
If you look at the data, retailers in the Philippines are not contracting according to
an optimal mix
25
Average power cost vs. load factor for Luzon grid ECs (2012)
The evidence is consistent with our hypothesis, that the structure of regulation does not
incentivise contracting for anything other than baseload plant
In addition, normal merchant markets put risks on retailers to encourage them
to contract (and to contract efficiently)
• Since WESM purchases are not subject to regulation, all WESM costs are passed through to
consumers
• In other markets, customers have fixed tariffs, and if spot prices are high, it is RETAILERS, not
the customers, who bear this risk
• In Philippines this is backwards.
• Since customers bear the risk, the regulator and the DOE worry about high WESM prices and
have capped the price
26
This band-aid solution merely tackles the symptom, not the disease and in the long
term will make the problem worse
The market cap price much lower than other markets
27
Comparison of the Market Price Caps
0
100
200
300
400
500
600
Singapore
NEMS
159
US
ERCOT
226
32
Australian
NEM
WESM
(2013)
538
62
PhP/kWh
Due to increase
in 2015
With a higher price cap:
More attention would be paid to effective
contracting
strategies – especially contracts to cover
disruptions
outages and peak demand
Flexible and responsive capacity (such as that
provided by LNG) would become
more commercially attractive, leading to
earlier
replacement of older, expensive, and less
responsive
capacity
However risk to retailers would increase, as
they would be more exposed to uncovered
(spot) price risk
And changes to the regulatory environment
are required to support this outcome
WESM
(2014)
The market woes and regulatory short-comings may yet hinder efficient development of
LNG in the Philippines
In summary
• There is an economic case for LNG to enter the Philippines
• The market is driven by the private sector and the private sector is responding to the needs
• Government policy is lagging private sector needs and recent decisions are actively deterring
new investment
• And regulation continues to undermine the incentives for efficient contracting which would assist
in underpinning new entry
28
We continue to watch, wait and push for improvements
Sarah Fairhurst
The Lantau Group
4602-4606 Tower 1, Metroplaza
223 Hing Fong Road
Kwai Fong, Hong Kong