Professional Documents
Culture Documents
E Q U I T Y
R E S E A R C H :
U N I T E D
S T A T E S
Electrical Equipment
Sameer Rathod
212-816-5250
James Sanders
203-975-6150
Citigroup Research is a division of Citigroup Global Markets Inc. (the "Firm"), which does and seeks to do business with companies covered in its research reports. As a
result, investors should be aware that the Firm may have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a
single factor in making their investment decision. Non-US research analysts who have prepared this report, and who may be associated persons of the member or
member organization, are not registered/qualified as research analysts with the NYSE and/or NASD, but instead have satisfied the registration/qualification requirements or
other research-related standards of a non-US jurisdiction.
United States
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Figure 2. Fuel Cell Stock Performance (2005 Year to Date performance and 2004 Annual Performance, Respectively)
2005 YTD Performance
2004 Performance
4%
30%
-6%
20%
10%
-16%
0%
-26%
-10%
-36%
-20%
-46%
Source: StockVal
S&P 500
S&P 600
FCEL
BLDP
MKTY
HYGS
S&P 500
S&P 600
FCEL
BLDP
MKTY
-50%
HYGS
-66%
PLUG
-40%
PLUG
-30%
-56%
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
High prices for oil, which are generally expected to persist in the medium term,
and a five-year delay in passing an energy bill in the U.S., have stimulated an
acute interest in alternative energy sources. Fuel cell producers and buyers will
benefit from nearly $4 billion in tax incentives from the energy bill.
Specifically, a tax credit of 30% per kilowatt (up to $1,000 per kW), which will
be available to telecom companies, will bring the present value of the life cycle
costs of fuel cells below the cost of the incumbent lead-acid battery backup
power solutions. We see the initial and life cycle costs of fuel cells as the key
hurdle to propelling higher volume orders and real commercial rollout.
Telecom companies view fuel cells as reliable, even as
misperceptions about cost abound.
According to survey results, nearly all telecom companies see fuel cells as
reliable electricity generators, yet perceptions about real costs are misplaced.
Reliability of the primary and backup electricity sources are far and away the
highest priority in telecom networks. Companies that have tested and
companies that are willing to deploy fuel cells all recognize the reliability of
fuel cell-generated electricity. However, responses to costs were not as
uniform as responses to reliability. The diversity of thought here likely reflects
the significant differences in the real assets of network architecture among the
telecom players. Our cost analyses show that the present value of the life cycle
costs of fuel cells are meaningfully lower than lead-acid batteries with and
without the upcoming tax incentives. Fuel cell companies selling into the
telecom backup power end markets need to use leverage they have earned on
reliability to convince power purchasers of the fuel cell cost advantage.
Stocks to own...
Plug Power (rated 1S) and Hydrogenics (rated 1S) are the companies in our sector
most likely to benefit from the accommodating macro environment and favorable
cost comparisons. We see the combined 100-plus unit orders received by Plug
Power in July as a harbinger of more frequent medium-sized orders in the next
12 to 18 months and view the companys 300-unit target in 2005 as achievable.
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Table of Contents
Investment Summary ................................................................................................................................ 3
Tackling Telecom: Calling on Sales to Drive Share Price Gains ......................................................... 5
Crossing the Chasm..................................................................................................................................... 5
Fuel Cells: Front Runner or Back Burner? ........................................................................................... 9
Backup Power Is Vital to the Wireless Industry.......................................................................................... 9
Back to Back: Advantages of PEM Fuel Cells vs. Lead-Acid Batteries .................................................... 9
Costs of Wireless Sites, New and Existing ............................................................................................... 12
Life Cycle Cost Comparison: Batteries vs. Fuel Cells ............................................................................. 16
Tower Sites Are Critical to Growth........................................................................................................... 19
Other End Markets Targeted for the Fuel Cell Backup Solution: Utility Substations and Broadband
Sites ...................................................................................................................................................... 23
Fuel Cell Commercialization Survey ........................................................................................................ 27
Appendix: How a Wireless Network Functions ....................................................................................... 31
Batteries, the Incumbent Backup Technology........................................................................................... 33
Fuel Cell Survey Questions....................................................................................................................... 36
Hydrogen Fuel........................................................................................................................................... 39
How a Fuel Cell Works ............................................................................................................................. 41
FIGURES:
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
The massive blackout that gripped the northeastern U.S. in 2003 exposed the
limitations of the wireline and wireless power backup strategies employed across the
nation and even globally by telecommunications companies. Wireline services were
disrupted for many customers, and most wireless systems failed due to battery outages.
Between the August blackout and the terrorist attacks on September 11, 2001, the
importance of maintaining dial tone has become glaringly evident. Accordingly, in
todays environment in which customers are cutting the wire completely, number
portability is becoming more commonplace, and emergency services are being
considered to be required on wireless networks customer retention through more
reliable service has become regarded as critical across the telecom industry.
Overall, our key takeaways for fuel cells in distributed telecom applications from
this report are:
The initial market opportunity for fuel cells appears to be in the cabinet style of
telecom remote sites, rather than in huts and controlled environment vaults
(CEVs). Batteries tend to have a lower life cycle (three to five years) due to
higher exposure to the elements compared with the conditioned environment of
huts and CEVs. Also, cabinet power requirements tend to be lower than the
other two styles, meeting typical fuel cell output in the 1 kW to 5 kW range.
Plug Power believes that a majority of the 180,000 wireless sites in the U.S. fall
into the 5 kW category. The results of our survey support this supposition.
Thanks to the trend of carriers to co-site, or share tower space, there are three
carriers per site on average, tripling the 180,000 data point, as each carrier
supplies its own backup power.
At the market price of $15,000 per unit plus $1,500 for installation the list
price is $18,000 and about $11,500 with tax credits Plug Powers 5 kW
GenCore system is competitive with a new site backup power installation, which
can range anywhere from $10,000 to $22,000.
With the $1,000 per kilowatt (or 30% of the installed cost) tax credit in the
Energy Policy Act of 2005, Plug Powers GenCore system cost at $11,500
($15,000 unit cost plus $1,500 installation cost less $5,000 tax credit) is closing
the gap with a battery replacement project at an existing remote telecom site.
Costs to replace batteries at existing sites, including disposal of old batteries and
installation of new ones, ranges between $3,500 and $8,000. (Battery disposal
can cost about $1,000 per site, but some companies actually make money on
disposal.) Plug Power is decreasing unit costs by 25%40% per annum.
Our respondents most often cited high initial costs of fuel cell units as the
primary drawback to implementation. However, we believe this is not an issue
for certain segments of the wireless marketplace. In terms of the five- and tenyear life cycle costs, we believe fuel cells can beat batteries by about 30%.
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Based on our detailed financial model, we estimate that Plug Power would need
to sell 6,000 GenCore 5 kW units at $3,000 per kilowatt to begin generating
accounting profit and positive operating cash flow. Annual sales of 6,000 units
imply roughly $90 million in sales and approximately 9% of the annual 5 kW
wireless telecom backup market leaving plenty of room for competition and
future share capture.
Fuel Cells Make Evident a Higher Value Proposition in a Large
Global Marketplace
Fuel cells in telecom applications can make up for the shortfalls and provide many
advantages that are not present in lead-acid batteries (the most common source of
backup power) and diesel-fired gensets today. These include first and foremost a
reduction in maintenance, remote monitoring over the Internet, and lower life cycle
costs. In addition, users note weight and size savings, noise reduction, less
sensitivity to temperature fluctuations, and few (if any) environmental concerns.
The accelerated growth of remote terminals in telecom has necessitated a cleaner
solution with lower costs and increased reliability to handle a changing industry.
The global backup
battery market annual
run-rate is about $2
billion. Our market data
include uninterruptible
power supply (UPS)
equipment and
battery/genset costs.
Recent reductions in cost have made PEM fuel cells viable backup power
alternatives to lead-acid batteries and diesel gensets. Plug Power leads the market
with the first commercially available fuel cell backup power system (GenCore).
Several others, including Hydrogenics (HyPM), Ballard Power (NEXA), ReliOn
(Independence), IdaTech, Nuvera, and United Technologies (PureCell), are also
introducing their own stacks and systems for the telco backup power market. The
industry today is in a pre-commercial or validation stage. Plug Power has shipped
more than 100 GenCore systems that are being tested by carriers today and some
have been deployed in their first commercial applications. We are slowly seeing the
other companies place units in the field as the telecom industry familiarizes itself
with the technology and braces for change.
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Due to cost, reliability, and durability issues, stationary fuel cells are not yet ready
for mass residential markets. Instead, critical backup power has been tapped as an
interim growth opportunity until these hurdles can be tackled. Simply put, backup
power is an ideal entry market since big business already pays a premium price for
power reliability and only expects units to operate intermittently, alleviating issues
seen in other always-on applications. In addition, these units operate most
efficiently on hydrogen directly, thereby lowering system expense as reforming
equipment is not needed. Ultimately, we expect these early approaches could pave
the road for lower cost systems and open new after markets.
We believe this report provides a much more direct response than work we have
seen before as to when the market for stationary backup power fuel cells could
emerge.
Whether market acceptance pulls volumes higher depends on two factors: cost and
reliability. While we show that life cycle costs are less for fuel cells compared with
batteries (including a survey of actual battery costs), the major sticking point
today is that the first cost between fuel cells and batteries are not compelling
enough. On the other hand, users are slowly becoming more convinced of the
durability and reliability of fuel cell backup solutions. Most hurdles here could be
surpassed in the coming 1218 months. Most importantly, the new U.S. energy
policy provides a 30% tax credit (up to $1,000 per kilowatt) to the installed cost of
stationary fuel cells a very positive catalyst we see inciting market demand on the
reduced cost gap.
We formed our conclusions after surveying wireline and wireless telecom providers,
government agencies, and uninterruptible power supply (UPS) providers. We also
spoke to battery suppliers. We gathered their opinions on the alternatives they are
pursuing and asked whether they feel threatened by fuel cells coming to commercial
fruition.
Looking forward to the remainder of 2005 and full year 2006, critical milestones for
the industry will be the introduction of more commercial-stage products from the
companies developing backup power fuel cells. In addition, we expect to see
material improvements in reliability/durability coupled with cost reductions. Last,
several companies have received and will continue to pursue critical certifications
from international and domestic standards bodies, including Underwriters
Laboratories (UL), Canadian Standards Association International (CSA), Network
Equipment Building System (NEBS), and European Conformity (CE). We expect to
start seeing larger and more frequent orders of fuel cells by telco companies for
backup power solutions in the next 12 to 18 months.
Early market acceptance by the telcos is imperative to carve out the initial market
over the next two years. The power utility market also has expressed sizable interest
in the technology, but trails the telecom industry by nine to 12 months in evaluation
testing. Another possible market is state emergency communications services, as
noted in the report issued by the state of New York following the 2003 blackout. As
reliability and cost are improved, we expect that incremental opportunities may
emerge to put fuel cells into baseload power (starting with remote telco sites and
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
homes willing to pay a premium) industrial and residential applications. Plug Power
expects to have its next generation prime power GenSys unit out in late 2005.
Plug Power and Hydrogenics Provide Investors with Exposure to
Nascent Commercialization of Fuel Cells
Plug Power and
Hydrogenics are the fuel
cell companies in our
coverage list catering to
telco backup power
applications.
Companies like Plug Power and Hydrogenics have focused specifically on getting
the system cost low enough so that products are gross margin positive even without
the benefits from volume production. The majority of fuel cells delivered today are
PEM systems, which has been a key driver in developing an increasingly robust
supplier base serving nearly 100 PEM fuel cell developers. A vast supplier base
facilitates fuel cell developers in engineering cost out through volume sales to fuel
cell OEMs. As volumes increase at fuel cell companies, we would expect margins
to increase materially.
A key, exogenous catalyst to get the fuel cell market moving toward mass
commercialization is tax incentives on alternative energy, including fuel cells.
Congress and the Bush administration recently passed the Energy Policy Act of
2005. The Energy Policy Act provides a 30% tax credit (up to $1,000 per kilowatt)
on the capital cost of the fuel cell system. This type of incentive has been
instrumental in re-igniting demand in the wind equipment industry over the past
decade. In our view, a similar initiative is a major win for the nascent fuel cell
industry one that may catapult fuel cell orders ahead quicker than first expected.
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
While wireline sites are mandated by the U.S. government to provide eight hours of
backup power to maintain communications in the event of a blackout and/or natural
disaster, wireless sites have no such requirement. However, as evidenced by the
blackout of 2003, the current backup power sources used by wireless providers
proved to be inadequate. On average, wireless applications worked for two to six
hours before failing because either the generators did not have enough time to
recharge the batteries or there was no backup generator in place to provide the
recharge. Most of the wireless respondents we spoke with indicated they expect six
to eight hours of backup time. Regardless of whether a genset was available or not,
the maximum backup time provided by the batteries during the blackout was at the
low end of the time expected.
In our view, the blackout of 2003 exposed some of the biggest weaknesses of leadacid batteries, which include sensitivity to both temperature and power draw
fluctuations and the tendency of wireless providers to either add more batteries into
a facility to increase reliability or worry about power failures in the event that they
happen. We believe that advancements must be made either from the battery
perspective or new technologies must be implemented to protect against a repeat of
the power failures that occurred during the blackout of 2003. In the latter part of this
report, we provide a description of lead-acid batteries and the two predominant
technologies (flooded and sealed) used in wireless applications.
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
and claims of smaller sizes, lighter weights, and lower maintenance costs were
made, skeptical buyers were wooed to try the new technology with warranties.
It is estimated that the
average lead-acid battery
can be cycled 80,000
times.
Flooded lead-acid batteries, reigning as the battery of choice since the 1950s, gave
way to valve regulated lead-acid (VRLA) batteries in the 1970s. The latter was a
slight improvement on the former, but the same dissatisfactory characteristics passed
into the new generation. High sensitivity to ambient temperature change and power
draw fluctuation decreases the batterys actual life (as opposed to its design life, or
the number of years it should operate) and creates uncertainty about duration of
available energy. That is a knock on reliability, one of the two key components of
choosing a backup solution. In the 2003 blackout in the northeastern United States,
many wireless providers bought batteries with backup times of six to eight hours but
in reality the backup strategy lasted less than six hours.
In addition to temperature sensitivity and power draw fluctuations, the following list
contains other characteristics backup buyers consider: useful life, maintenance
requirements (frequency, duration, and cost), installation costs, disposal costs of
weary batteries, costs associated with managing the ambient temperature, size,
weight, familiarity with the technology, noise, durability, security, zoning, remote
monitoring, and environmental friendliness. The costs of the batteries themselves,
compared with fuel cells, are quite cheap. However, when we look at the cost of the
entire package, fuel cells are essentially coming in line with batteries. Take a 5kilowatt cell site, for example. Assuming new batteries are a nominal cost of $3,000
to $5,000, the cabinet (including wiring) in which they are housed to control the
ambient temperature adds an extra $4,000 or $5,000 onto the cost, at a minimum.
When all other identifiable costs are considered (installation, maintenance, disposal,
and backup site engineering), we found that PEM fuel cells are at or better than the
incumbent technologys cost.
On balance, we believe PEM fuel cells win out over batteries on quantitative and
qualitative characteristics as a best-choice backup power solution by about 10 to 1.
We asked telcos to list advantages and disadvantages of various characteristics fuel
cells and batteries, like cost, size weight, zoning, and remote monitoring. The
following chart, which is unweighted and unranked and assumes that costs and
reliability are identical, summarizes the characteristics culled from telecom
companies.
10
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Figure 3. Contrasting PEM Fuel Cells and Batteries on Most Important Backup Power Characteristics
Lead Acid Battery
Note
Useful life
Advantage
Maintenance requirements
Advantage
Remote monitoring
Advantage
Installation costs
Advantage
~$500 savings
Disposal costs
Advantage
Temperature sensitivity
Advantage
Size
Same
Same
Weight
Advantage
Availability
Advantage
Durability
Advantage
Additional output
Advantage
Noise
Same
Same
Security
Same
Same
Zoning
Same
Same
Environmental friendliness
Advantage tally
Advantage
1
10
Interestingly, one of the more significant features of a product like Plug Powers
GenCore backup module, remote monitoring, is low on the radar screen for backup
power solution buyers. That surprised us, and we believe it points to several things.
First, it means that fuel cell sales efforts have work to do to convince buyers of the
benefits of this feature. This feature is not just another bell or whistle, but a
significant element that could dramatically reduce labor costs associated with
maintenance. The trend for remote monitoring has already whetted the appetites of
utility companies. Now, meter readers can drive by homes and offices while radio
frequency identification (RFID) tags, like those made by Ropers TransCore,
transmit readings right into the vehicle; needless to say, significant labor cost
savings are expected, and we expect customers will look at this option as a payback
approach in time. Also, we think it points to telecom managements current focus
on extending the networks and investing capital in the latest revenue-generating
equipment. Cost-centered focuses appear to overlook the backup energy
component. That responsibility seems compartmentalized to the energy groups
within the telecom companies. In other words, backup power only becomes an issue
when the grid goes out.
Lead-acid batteries
typically last from three
to five years, and less
than two in hotter
climates with wider high
and low temperature
swings.
PEM fuel cells have a much greater range of resilience than batteries to ambient
temperature fluctuations. This represents a significant advantage. Lead-acid
batteries lives are reduced at temperatures excessively high and low. When buying
batteries, the volt and amp ratings and useful life is based on a 75 to 78 degree
Fahrenheit ambient temperature. Remote sites in the southwest and southern United
States have more frequent battery replacement cycles due to the degradation caused
by temperature changes as well as continuous temperatures above or below 70 to 80
degrees. Not only are replacement and maintenance cycles affected by labor costs
and other expenses, but also network reliability is at risk the most critical aspect
11
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
of a network in the campaign to retain subscribers. Fuel cells can operate reliably in
a range from minus 50 degrees to positive 115 degrees Fahrenheit.
The space available for backup power solutions varies widely. In some
configurations, though few we believe, have ample space for myriad batteries and
any additional equipment to condition their environment. Many locations are towers
and rooftops located in residential or commercial zones. Thus, the size and weight
of the backup solution matters. For instance, a battery backup solution for a roofmounted wireless base station can have a battery setup that weighs up to 1,600
pounds. The GenCore unit, on the other hand, weighs about 400 pounds and fuel
storage or an electrolyzer or reformer are not much more. Also, batteries are often
mounted in racks, which require additional labor and materials expense to keep
numerous batteries and considerable weight suspended. A GenCore is
approximately the size of a household dishwasher, hardly a square-footage hog.
Another significant advantage of fuel cells over batteries is meeting the need for
additional output. If a remote site requires more power, additional batteries have to
be installed to increase power output. On the other hand, increasing the hydrogen
directly input into fuel cells solves the increased output requirement. Additional fuel
is far less costly than engineering, labor, installation, and new materials costs
associated with increasing battery output.
Finally, although it was not part of our study, we point out that fuel cells are a
different form of solution than batteries. A battery backup power solution is an
electricity storage solution, whereas fuel cells are electricity-generating solutions.
So long as fuel such as natural gas, methane, hydrogen, or water can be supplied, the
fuel cell can generate electricity and power the load. We believe this is a
noteworthy advantage of fuel cells over batteries and one that is perhaps
underappreciated at this time. In a situation like the blackout in 2003, the wireless
network provided less than the six or eight hours of backup power intended. In a
more serious situation such as September 11, the ability to provide continuous
electricity through fuel cell backup solutions would benefit not only wireless
subscribers but especially government and security responses.
12
After talking with numerous wireless providers and consultants, the one constant we
found regarding the cost of installing a remote station was that there is no average
cost. Variables such as size (i.e., power draw and signal strength), hut versus
cabinet, and amount of battery backup time were some of the many reasons
mentioned for the wide range of responses. We had to really focus more on where
we thought a fuel cell application could compete today. In our view, the appropriate
market would be the wireless cell site that was using a cabinet style system and
required approximately 3 kW to 7 kW of power. Added power output means more
batteries and more cost per site. From there, we were able to talk more effectively
with the wireless consultants and providers. Upside to these numbers exists in
supplying higher powered fuel cell modules, as well as wireline, broadband, and
electric utility substations.
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Some of the companies that we spoke with to get a better understanding of what
goes into current backup power solutions were Enersys, C&D Technologies, Center
Power, Douglas Battery Mfg. Systems, Eagle Picher, East Penn Mfg. (Deka), Exide,
and Power Battery Company. In addition, we spoke with American Power
Conversion (equipment/cabinets) and multiple contacts from the Cellular
Telecommunications Industry Association (CTIA).
New Wireless Tower Installation
New wireless sites cost
$15,000$25,000, on
average, while
replacement of batteries
at an existing site are
typically $3,600$8,000.
On average, we heard that a new wireless installation could range from $10,000 to
$22,000 depending on the type of equipment used. Included in that cost is the
cabinet (i.e., a metal box, sometimes insulated), which can cost from $7,000 to
$15,000, depending on how much equipment is installed in the box by the cabinet
manufacturer. Above and beyond that cost, you typically have site and installation
costs, which can range from $1,500 to $3,500 and battery costs, which can range
from $1,600 to $6,000 (usually three or four strings of four to six batteries each
providing approximately 4 kW7 kW of total power). Using the data from our
survey, we have arranged responses into three cohorts for new wireless sites (low
end, middle, and high end) and two categories for existing wireless sites (low and
high ends). See the figures below.
13
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Figure 4. Cost Comparison for 5-Kilowatt Cell Sites: Battery Backup System vs. Fuel Cell Systems
$25,000
$20,000
$15,000
Tax Credit
$10,000
$5,000
$0
Tax Credit
$4,950
$0
$0
$0
$0
$0
Total
$11,550
$10,200
$15,200
$21,400
$3,600
$8,000
14
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Figure 5. Overview of Costs for Existing and New Wireless Cell Sites
$21,400
$25,000
$15,200
$20,000
$15,000
$10,200
$8,000
$10,000
$3,600
$5,000
$0
$0
$0
$0
$1,000
$1,000
Installation
$600
$1,000
$2,000
$1,000
$1,000
Engineering
$1,000
$1,200
$1,400
$0
$0
Battery Cost
$1,600
$3,000
$6,000
$1,600
$6,000
$3,000
$4,000
$5,000
$0
$0
Radio equipment
$4,000
$6,000
$7,000
$0
$0
Disposal costs
Radio equipment
Battery Cost
Engineering
Installation
Disposal costs
15
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Battery Replacement
On a ten-year life cycle cost analysis, fuel cells are about 24% less expensive than
battery solutions on a three-year replacement schedule (see the figure below). In the
analysis, we choose the mid-case new cell site battery backup solution cost of
$15,200. We then assume that the batteries are replaced every three years. We also
assume that the company will incur about $1,000 in maintenance cost to service the
batteries at the site. We chose an 8% discount rate to achieve the present value; our
choice is based on a current prime rate of 6.5% plus 1 points for conservatism. The
present value of the battery backup solution is $28,000. We then calculated the cost
of a Plug Power 5 kW fuel cell solution. We assume an actual cost of $16,500 (or
$15,000 for the unit, plus $1,500 for installation). We then determined the present
value of the costs to maintain the fuel cell unit and supply it with fuel. We assume
these costs run at about $700 per annum. The present value of the fuel cell for the
unit, installation, fuel, and maintenance is $21,500, 24% less than the battery solution.
A slight variation on this scenario is to assume that the batteries are replaced on a
five-year cycle rather than a three-year cycle. In this analysis, the present value of
the battery solution would be $24,500, a 12% premium to the present value of
$21,500 for the fuel cell backup solution.
16
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Figure 6. Ten-Year Payback Analysis: Batteries vs. Fuel Cells Cost Comparison (Without Tax Credit)
Average Site Installed (power output)
kW
Total
Disc. Rate
8.0%
Batteries
Year
0
1
2
3
4
5
6
7
8
9
10
Total
$15,200
$3,000
$3,000
$3,000
$24,200
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$10,000
$16,200
$926
$857
$3,175
$735
$681
$2,521
$583
$540
$2,001
$28,220
$34,200
$15,200
$3,000
$18,200
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$10,000
Total
$28,200
Disc. Rate
8.0%
$16,200
$926
$857
$794
$735
$2,722
$630
$583
$540
$500
$24,489
$21,573
Next, we took into consideration the benefits from the tax credit available under the
new energy policy. The fuel cell unit ($15,000) and installation ($1,500) are both
qualified for the 30% tax credit, a savings of about $5,000. The credit lowers the
cost of the fuel cell to $11,500 from $16,500. Our other assumptions are identical to
the first analysis. In this analysis, fuel cells are 41% less expensive than the battery
solution on a three-year replacement cycle, and 32% less expensive on a five-year
replacement cycle (see the figure below).
Figure 7. Ten-Year Payback Analysis: Batteries vs. Fuel Cells Cost Comparison (With Tax Credit)
Average Site Installed (power output)
kW
Total
Batteries
Year
0
1
2
3
4
5
6
7
8
9
10
Total
$15,200
$3,000
$3,000
$3,000
$24,200
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$10,000
$34,200
Disc. Rate
8.0%
$16,200
$926
$857
$3,175
$735
$681
$2,521
$583
$540
$2,001
$28,220
$15,200
$3,000
$18,200
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$10,000
Total
$28,200
Disc. Rate
8.0%
$16,200
$926
$857
$794
$735
$2,722
$630
$583
$540
$500
$24,489
$16,623
17
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
We also considered a 15-year payback analysis. Fuel cell units likely have a useful
life of ten to 15 years, according to Plug Power and Hydrogenics. In the figure below,
we assume three- and five-year battery replacement cycles on the mid-case battery
backup solution with an initial cost of $15,000 per year and about $700 per annum
in maintenance and fuel costs. The present value of the three-year cycle scenario is
$31,500; the present value of the battery solution in the five-year cycle is $28,000.
We use the same assumptions to calculate the present value of the life cycle costs for
the fuel cell solution. The assumptions of the fuel and maintenance requirements are
identical. The present value of the fuel cell solution is $23,000 in the 15-year
payback analysis, 27% less expensive than the battery solution on a three-year
replacement basis and 18% less expensive than the battery solution on a five-year
replacement cycle. Once again, the fuel cell solution is less expensive and it is so
before the tax credit (see the figure below).
Figure 8. 15-Year Payback Analysis: Batteries vs. Fuel Cells Cost Comparison (Without Tax Credit)
Average Site Installed (power output)
kW
Total
Batteries
Year
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Total
$15,200
$3,000
$3,000
$3,000
$3,000
$27,200
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$15,000
$42,200
Disc. Rate
8.0%
$16,200
$926
$857
$3,175
$735
$681
$2,521
$583
$540
$2,001
$463
$429
$1,588
$368
$340
$31,408
$15,200
$3,000
$3,000
$21,200
18
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$15,000
Total
$36,200
Disc. Rate
8.0%
$16,200
$926
$857
$794
$735
$2,722
$630
$583
$540
$500
$1,853
$429
$397
$368
$340
$27,876
$22,971
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
With the tax credit, the present value of the fuel cell solution, including fuel and
maintenance costs, is $18,000. This cost is 43% and 35% less expensive than the
battery backup solution on the three- and five-year replacement cycles, respectively
(see the figure below).
Figure 9. 15-Year Payback Analysis: Batteries vs. Fuel Cells Cost Comparison (With Tax Credit)
Average Site Installed (power output)
kW
Total
Batteries
Year
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Total
$15,200
$3,000
$3,000
$3,000
$3,000
$27,200
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$15,000
$42,200
Disc. Rate
8.0%
$16,200
$926
$857
$3,175
$735
$681
$2,521
$583
$540
$2,001
$463
$429
$1,588
$368
$340
$31,408
$15,200
$3,000
$3,000
$21,200
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$1,000
$15,000
Total
$36,200
Disc. Rate
8.0%
$16,200
$926
$857
$794
$735
$2,722
$630
$583
$540
$500
$1,853
$429
$397
$368
$340
$27,876
$18,021
Our conversations with telco companies highlighted two criteria: cost and
reliability. Nearly across the board, we heard no complaints on the reliability of the
fuel cell backup power solution. We heard a considerable number of responses
claiming costs were prohibitive in choosing the fuel cell alternative. However, our
costing analysis shows that when total life cycle costs are considered, fuel cells are
less expensive than the incumbent technology.
The next 12 months are key for the fuel cell manufacturers: they need to convince
telco customers that indeed fuel cells are a better solution compared with batteries.
This should be made easier with the 30% energy tax credit. Doing so is the linchpin
to catalyze the fuel cell market from validation to commercialization.
For each tower, there is a set of radio frequencies that provides service to its
particular geographic range. While increasing the power to a wireless site is one
way to improve the coverage of the signal from a distance perspective, only a
limited number of users can use the associated frequency range. As a wireless
provider, this strategy presents a problem because the amount of users on the tower
is limited. So the solution that evolved was to lower the power of each wireless site
(i.e., reducing the range of the signal while maintaining the same frequency) and
increase the total amount of sites (i.e., keeping the same geographic coverage area).
See below for a graphical depiction.
19
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Source: www.privateline.com
250,000
30,000
200,000
175,000
150,000
20,000
125,000
15,000
100,000
75,000
10,000
225,000
50,000
5,000
25,000
2007E
2006E
2004
2005E
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
1993
1992
1991
1990
1989
1988
1987
1986
1985
Due to the trend of co-siting, we estimate that there are two to three carriers at each
wireless site on average. Some sites have as many as five carriers. Because each
carrier installs and maintains its own equipment, including backup power, in reality
there are really two to three times the number of sales opportunities as wireless
tower sites. We take this into consideration in our market analysis.
20
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
With the impressive growth in wireless towers, the wireless telecom backup power
market is one of the most attractive markets for fuel cell backup power.
Construction of new towers makes entry into this market a more viable option, as
existing sites have to be reconfigured. The new wireless sites that Plug Power and
its competition are pursuing are prime targets for a product that is already
engineered to be dropped into the current configuration.
We performed a study of the wireless and wireline markets. Plug Power has
mentioned that its strategy favors the wireless over the wireline market. In our
wireless backup power study, we looked at the number of wireless sites in use. The
CTIA estimates there are roughly 180,000 sites today. Using this as a basis, we
forecasted 13% annual growth the long-term CAGR. We then assumed a share of
the total sites fit GenCores 1 to 5 kW power specification. Our base case assumes
that 40% of the wireless sites installed today fit this power specification. We hold
that rate constant going forward. Our worst and best cases assume rates of 30% and
50%, respectively. Next, we assumed that Plug Power could steal 1% of the market
for each of the scenarios. The figure below shows the results of our study out to
2011. We estimate that Plug Power can capture 708 GenCore sales in 2006 in the
wireless backup market. Going forward, we assume that Plugs market share
increases by a factor of 1.75. The bottom chart below shows sensitivities around base
case penetration rates in 2006. The analysis shows it could capture as few as 71 sales
or as many as 1,346. We assume the same penetration sensitivities on both existing
and new remote sites, with penetration rates ranging from 0.1% to 1.9%.
Figure 12. Market Model: Wireless Backup Power (5 kW) Market: Plug Power Penetration
Scenario
Base case
Worst case
Best case
2006E
2007E
2008E
2009E
708
531
886
1,373
1,030
1,716
2,727
2,045
3,409
4,668
3,501
5,835
2010E
10,556
7,917
13,195
2011E
20,949
15,712
26,187
Assumptions: % 5 kW*
Share of new sites
Share of exist. sites
Base case
40%
1%
1%
Worst case
30%
1%
1%
Best case
50%
1%
1%
*% 5 kW = the share of total wireless sites that require 1 to 5 kW of electricty
708
0.1%
0.4%
0.7%
1.0%
1.3%
1.6%
1.9%
1.9%
797
889
980
1,072
1,163
1,255
1,346
21
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
We also performed the same scenarios analysis on the wireline backup power
market. Many of the assumption levers are the same. In wireline, however, we use
a 20% rate in our base case for the number of remote sites in the U.S. that are in
Plug Powers GenCore units 1 to 5 kW power specification range. Our worst case
and best case assume rates of 15% and 25%, respectively. Like the wireless
scenario, our wireline study assumes that the market share captured in 2006 is 1%
and that it growth annually by a factor of 1.75.
In 2006, our study shows that Plug Power can place 88 GenCore 5 kW units into the
wireline backup power market. Sales increase to 1,550 by 2011 (see the figure
below).
In the bottom chart, we performed a sensitivity analysis on 2006 base case
penetration. Our base case assumption is 1%; the sensitivity examines rates of
0.1%1.9%. We assume the same penetration sensitivities on both existing and new
remote sites. In 2006, we estimate that Plug Power may sell as few as nine and as
many as 167 GenCore units into the wireline backup power market.
Figure 13. Market Model: Wireline Backup Power (5 kW) Market: Plug Power Penetration
Scenario
Base case
Worst case
Best case
2006E
2007E
2008E
2009E
2010E
2011E
88
66
110
156
117
195
278
209
348
493
370
616
874
656
1,093
1,550
1,162
1,937
Assumptions % 5 kW*
Share of new sites Share of exist. sites
Base case
20%
1%
1%
Worst case
15%
1%
1%
Best case
25%
1%
1%
*% 5 kW = the share of total wireline sites that require 1 to 5 kW of electricty
88
0.1%
0.4%
0.7%
1.0%
1.3%
1.6%
1.9%
1.9%
158
159
161
162
164
165
167
Recall that we estimate that Plug Power becomes cash flow and profitable around
6,000 GenCore units at the price of $15,000 for 5 kW and $1,500 for installation. If
Plug Power only concentrated on the wireless and wireline end markets, our studies
show that they would become profitable between 2009 and 2010. Our model
currently projects profitability in 2008. Other markets are targeted for the GenCore
product, including utility substation and broadband fiber power nodes and amplifiers
and regenerators.
22
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
23
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Figure 14. Market Model: Utility Substation Backup Power (5 kW) Market: Plug Power Penetration
Scenario
Base case
Worst case
Best case
2006E
2007E
2008E
2009E
2010E
2011E
168
148
178
300
264
317
548
484
580
978
863
1,036
1,746
1,541
1,849
3,117
2,750
3,301
Assumptions:
% 5 kW*
Share of new sites Share of exist. sites
Base case
85%
1%
1%
Worst case
75%
1%
1%
Best case
90%
1%
1%
*% 5 kW = the share of total utility substation sites that require 1 to 5 kW of electricty
168
0.4%
0.6%
0.8%
1.0%
1.2%
1.4%
1.6%
1.6%
248
251
255
258
262
265
268
For illustrative purposes we have conducted the same scenario and sensitivity
analysis on the cable backup power market (see the figure below).
Figure 15. Market Model: Cable Backup Power (5 kW) Market: Plug Power Penetration
Scenario
Base case
Worst case
Best case
2006E
2007E
2008E
2009E
2010E
2011E
192
115
308
341
205
546
609
366
975
1,080
648
1,728
1,915
1,149
3,064
3,394
2,037
5,431
Assumptions: % 5 kW*
Share of new sites
Share of exist. sites
Base case
25%
0.3%
0.3%
Worst case
15%
0.3%
0.3%
Best case
40%
0.3%
0.3%
*% 5 kW = the share of total broadband sites that require 1 to 5 kW of electricty
192
0.0%
0.1%
0.2%
0.3%
0.4%
0.5%
0.6%
24
0.5%
302
306
310
313
317
321
324
0.6%
363
366
370
374
378
381
385
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
The figure below summarizes the total opportunity for Plug Power if it entered the
four end markets we described above. Our base case shows in the top chart that by
2009, Plug Power may sell about 7,219, more than the 6,000-unit breakeven volume
our model projects. The bottom charts combines the four sensitivity analyses
discussed above. We should note that these estimates do not include our assumptions
for the GenSys and GenSite products that Plug is in the process of commercializing.
Please see all 5 of our expanded market models in the appendix.
Figure 16. Scenario Analysis on Plug Power Penetration into the 5 kW Backup Power Market
Scenario
Base case
Worst case
Best case
2006E
2007E
2008E
2009E
1,157
861
1,481
2,169
1,615
2,774
4,163
3,103
5,312
7,219
5,382
9,215
2010E
15,092
11,263
19,201
2011E
29,011
21,661
36,855
856
956
1,057
1,157
1,257
1,357
1,457
1,093
1,193
1,293
1,393
1,493
1,593
1,693
1,330
1,430
1,530
1,630
1,730
1,830
1,930
1,566
1,666
1,766
1,866
1,966
2,066
2,166
In short, the market opportunity for fuel cells is very large. Even isolating the 5 kW
segment leads us to conclude that should commercial markets develop as they
appear poised to do, there is plenty of space for multiple players. Using the
conservative, base-case estimates we mentioned in the four market analyses above,
we see a fully developed 160,000-unit (5 kW) opportunity blossoming in 2006 and
recurring annually with modest growth.
Using the same assumptions as the four market analyses above, we reach the
following conclusions:
Assuming that 40% of the 224,000 wireless sites by year-end 2006 (i.e., 13%
growth over 2004 sites of about 176,000 and 2005 sites of about 198,000)
require 5 kW or less of power, and that three carriers co-site at each wireless
site, a five-year replacement cycle yields about 71,000 opportunities in 2006.
Assuming that 20% of the estimated calendar 2006 198,000 wireline sites (based
on 1.3% growth per annum) require 5 kW of power or less, a five-year
replacement cycle yields about 8,800 opportunities.
Assuming that utility substations will growth about 2% per year to 102,000 in
2006, and that 85% of those require 5 kW of power or less, a seven-year
replacement cycle yields about 17,000 opportunities per annum.
25
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
200,000
5kW units
160,569
208,546
182,350
181,733
2008E
2009E
225,009
169,452
150,000
100,000
50,000
2006E
Wireless market
2007E
Wireline market
2010E
2011E
$3.5
Billions of dollars
$3.0
$2.41
$2.54
2006E
2007E
$2.74
$2.73
$3.13
2008E
2009E
2010E
$2.5
$2.0
$1.5
$1.0
$0.5
$0.0
26
2011E
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
It is quite unlikely that any player will capture several percentage points of that
market in the near term, but customer recognition of the value proposition could in
time translate to more significant sales. If fuel cell companies sell their units for
$3,000/kW, we estimate that the dollar value of the 2006 market would be about
$2.4 billion. Assuming even a 20% haircut to our numbers, units could approach
128,000 and sales about $1.9 billion.
We stress that these values are potential, especially because broadband is further
behind than the other three markets. Excluding cable/broadband units, we estimate
between 77,000 and 116,000 total opportunities in 2006. The figures above detail
our market estimates from 2006 to 2011.
When looking at just Plug Powers penetration into these markets according to the
(base case) assumptions we enumerated in the previous section, we estimate that Plug
Power could sell a total of about 1,157 units in 2006. This is one-sixth the number of
units Plug Power needs to sell break even. We show that Plug Power can hit the
6,000-unit mark by 2009. Please see the figure below for estimates through 2011.
Figure 18. Plug Power: Capture of the 5 kW Backup Power Market
35,000
30,000
5kW units
25,000
20,000
15,000
10,000
5,000
2006E
2007E
2008E
2009E
2010E
2011E
Wireless market
Wireline market
Our survey was conducted over the second and third quarters this year and was
concentrated on the telecommunications backup power end market. After compiling
a list of 50-plus national, regional, and local telecommunications providers, we
identified various internal contacts such as chief technology officers, switch
operations managers, chief network architects, power maintenance engineers, and
knowledgeable investor relations directors.
First and foremost, we learned how simple questions, such as How much power is
required at your site, How many batteries do you use at a site, Have you
considered alternative backup power solutions, and How much does your typical
backup power solution cost, generated widely variant answers. Every carrier and
every network is different. Oftentimes, a backup solution for the same company in
27
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
one location will not be the same in another. We spent considerable time with our
contacts hashing out the specifics of their network architecture, backup power
requirements, and costing estimates. Despite what we might think about the ease of
placing a call from New York to Los Angeles, there is little homogeneity in our
national telecommunications infrastructure.
Power requirements for
wireline and wireless
sites ranged anywhere
from 1 kW to more than
60 kWs.
28
While it first appears that fuel cells can meet or beat batteries on total site costs for
backup power, the cost per power generated does not match up. For the major
wireline company above whose total site costs were about $11,000, the dollar-cost
per kilowatt was about $1,500, half of both Plug Power and Hydrogenics publicly
stated cost per kilowatt. A local western wireless provider with fewer than 20 sites
could provide backup power solutions 54% less expensive than the fuel cell
solutions. The regional wireless provider mentioned above, however, generated
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
backup power at a 54% premium per kilowatt compared with the fuel cell
companies solutions, driven primarily by engineering costs rather than power
production.
Plug believes 40% of the
180,000 cell sites
currently in use are 5 kW
or less.
Such a wide array of results confirmed the difficulty of running a quick market
analysis and generating estimates. Because the companies we cover are focused on
the 5-kW sliver of the telecommunications backup power supply market, we believe
conservative estimates are in order. In terms of wireless, Plug believes that 40% of
the 180,000 cell sites in the U.S. fall into their defined 5-kW niche. The success of
fuel cell commercialization depends a great deal on the manufacturers and their
focused sales efforts to discover and deliver reliable, cost-efficient solutions to the 5kW sweet spot. Since batteries are notorious for actual lives less than their stated
design life, the replacement-oriented market provides plenty of turnover and
opportunity for new solutions. Our conclusion is that battery buyers in telecom have
yet to be convinced.
Breaking and Entering: How Fuel Cells Can Gain Share in the
Battery Replacement Market
Conversations with our power contacts did not leave a clean and defined answer to
the question of how fuel cell companies can break into the replacement cycle and
gain share of the backup power market. When asking whether carriers would deploy
fuel cells into their networks if the technology could meet or beat the incumbent
battery systems on reliability and cost, we did not receive clear and distinct answers
from our contacts. We experienced bona fide hesitation, so good reports widely
distributed from current test piloting are critical. Yet, out of that reluctance, we
were able to delineate several characteristics that would help fuel cell companies
make the sale.
Batteries in the fuel cells
are an issue with our
wireless and wireline
contacts.
We want to stress that our survey has generally found fuel cells to be just as reliable
as batteries. Hesitations surfaced when power chiefs spoke about the fact that fuel
cells at this point still have batteries in them. That fact means that company
personnel still has to maintain the remote sites, and the company still has to deal
with batteries. Ultracapacitors are viewed as a superior alternative to batteries given
their reliability and extensive cycling ability. Second, the addition of fuel cells into
network architecture often meant increased complexity in network maintenance, a
raw reality deeply appreciated by those performing the task. So, why bother at all?
Fuel cell sales efforts have an uphill battle in creating the perception that they are
knowledgeable suppliers with considerable industry experience. While the
technology is generally perceived to be highly available and highly reliable, the
scant field testing and lack of actual deployment generate a cry for prove it in a
situation in which it just has to work. Tycos profound industry knowledge lends
sturdy platform to Plug Power and increases credibility. Tyco has been testing Plug
Powers product for the past year and is satisfied with performance characteristics, a
result that will likely lead to higher sales in the near term.
We did not perceive our contacts to be overly excited about the initial costs. While
they did not feel a slight premium to first cost was particularly burdensome, the
reality of firm and large-enterprise decision-making seemed to act against them.
Once again, the backup power folks are completely invisible to upper level
29
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
management, as one carriers power contact described his experience to us, until a
crisis erupts. At that time, they are headliners; but when the crisis is resolved, they
once again fade out of sight, out of mind. It appears the challenge lies on the fuel
cell sales efforts to convince power purchasing authorities that the increased initial
capital expenditure will be offset by the reduction of operating expenditures over the
product life cycle.
Still, our conversations led us to believe that the first cost is the key driver to
converting a prospect into a sale. While industry capex dollars are flat, fuel cell
companies would need to convince telecom companies to redirect this vital
investment capital to as-yet widely untested, network-critical new equipment. That
could prove to be a tough sell, as companies vie for new and existing subscribers
within an industry-wide evolution to the latest technologies.
Figure 19. Wireless and Wireline Industry Capital Expenditures
$60,000
$50,000
$40,000
$30,000
$20,000
$10,000
Wireless
2010E
2009E
2008E
2007E
2006E
2005E
2004
2003
2002
$0
Wireline
Finally, some aspects of fuel cells, which could be quite highly valued, are generally
overlooked. Fuel cell sales efforts point to the benefits of power system intelligence
and network management through offerings such as remote monitoring. However,
responses on our surveys were tepid.
As is widely reported, though, the battle is not just among telecom providers, but
between them and cable companies as well. When it comes to new spending, even in
the event that the power group can get budgets funded, it is unlikely that a new backup
fuel technology would be able to preempt revenue-generating equipment at the site.
One power chief likened the location, location, location mantra in real estate to its
analog in telecom: footprint, footprint, footprint. This point was corroborated by
several of our contacts. In addition to intra- and inter-industry market share battles,
the telecommunications industry as a whole has embarked on a long-term migration
from circuit switching to packet switching. Furthermore, industry consolidation is
taking place. An issue that pops up immediately in corporate telecom marriages is
how to integrate very different network architectures. Adding fuel cells into the
power supply is sometimes perceived as an unnecessary complication.
30
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
When a wireless call is made, the wireless signal travels over a frequency through a
channel to the nearest tower and then is routed to either another tower or a landline
depending on what the final destination is. In the case of long-distance calls, the
signal usually goes to a switching and/or call center and is then sent to the tower or
landline closest to the intended party. Depending on the geographic region and the
services offered by the wireless provider, the way the signal travels on the frequency
can entail a number of different techniques, including analog, TDMA, and CDMA.
Figure 20. Layout of a Telecom Network
31
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Since the inception of wireless, the frequency band that transfers signals has
remained fundamentally the same. However, the way the frequency is used has been
altered significantly. Multiple carriers may use the same spectrum space, but
depending on the technology that is employed (e.g., analog, TDMA, and CDMA),
the amount of channels offered could vary significantly.
Different wireless
broadcast techniques
such as analog, TDMA,
and CDMA do not
materially change the
power draw of a wireless
site.
32
Today, most wireless signals are digital, where information (i.e., voice and data) is
converted to a series of 0s and 1s (i.e., binary code) called data packets and
transported back and forth through the wireless network. While some providers still
utilize the legacy non-digital Frequency Division Multiplexing technique (i.e.,
analog), it is more common to find Time Division Multiplex (TDMA or GSM) or
Code Division Multiplex (CDMA, WCDMA) techniques. While understanding
these differing technologies may be important for a report on the future of the
wireless market, the main thing to note for our purposes is that the power
requirements are essentially the same for all three.
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Source: www.calpoly.edu
Lead-Acid Batteries
Invented in the mid-1800s, lead-acid batteries were the first commercially available
rechargeable battery. Typical end market uses include automobiles, motive
applications (i.e., forklifts, golf carts, etc.), telecom, and uninterruptible power
supply systems (UPSs). For this report, we are focused on the telecom industry,
which has historically used two main styles: flooded and valve regulated lead-acid
batteries (VRLAs).
The Move from Flooded to Valve Regulated Lead-Acid Batteries and
the Importance of the Warranty
Flooded lead-acid batteries got their start in the early 1950s as an alternative to the
traditional Plante and high antimony batteries. Advantages versus the incumbent
technology included smaller size and weight and lower maintenance costs.
However, the biggest drawback at the point of introduction was the lack of operating
33
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
history; users of traditional batteries did not trust the new lead-acid battery
technology.
Figure 22. Example of a Flooded Lead-Acid Battery
Source: www.enersys.com
Flooded lead-acid
batteries initially
displaced the incumbent
lead-acid technology
because of a strong
warranty.
34
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Like flooded lead-acid batteries, VRLA batteries also had long warranty periods
because that was the standard in the battery market. However, a problem arose with
this guarantee. The VRLA batteries themselves did not last as long as anticipated
because end users were exposing the batteries to nonoptimal environmental
conditions, which affected the life of the battery and resulted in several replacements
at no cost to the customer. On the flip side, the end users were becoming frustrated
because the battery life would be shorter than expected and, as a result, their
applications would periodically fail (e.g., a wireless tower would lose power, which
resulted in a dropped call or loss of signal).
While these issues persisted for many years, these battery manufacturers and end
users eventually worked out their differences (mainly because batteries were the
only game in town for backup power) and new warranties were developed that were
more geared to the ultimate end usage and the amount of outside influences that
were affecting the batteries. Today, batteries are sold either with or without
warranties. If the warranty is purchased, it can range anywhere from one to 20
years, with the coverage usually pro-rated over the applicable time frame.
35
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
36
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
37
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
38
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Hydrogen Fuel
Plug Powers GenCore unit is designed to run on 99.95% dry gaseous hydrogen.
This type of hydrogen gas is available from several distributors, but in March 2004,
Plug Power and Airgas, the largest U.S. distributor of industrial, specialty, and
medical gases, entered into a five-year hydrogen fuel service agreement where
Airgas would be the preferred suppliers of hydrogen fuel to GenCore unit owners.
In this agreement, Plug Power and Airgas will jointly market hydrogen fuel services
products for GenCore unit in the United States. Additionally, Airgas will provide a
variety of fuel services, including cylinder drop-off, connecting to Plug Powers
GenCore unit, and providing a complete line of specialty gas equipment to help
manage the fuel supply.
The cost of hydrogen can vary depending on the geographic location, quality/purity,
quantity, and delivery method. For the purposes of the GenCore fuel cells unit, only
purer grades of hydrogen can be used, which can cost anywhere from $0.50 to $1.00
per cubic feet depending on the quantity purchased. The smaller container in which
the hydrogen is stored, or tanks, can be purchased from the distributor for
approximately $160 to $175 where as the larger 200 cubic feet tanks must be rented
or leases at a rate of approximately $55$65 per year. In addition, depending on
location, delivery can cost between $30 to $50 and can be often delivered within 24
to 48 hours of initial request.
Figure 25. Current Hydrogen Pricing
Hydrogen
Tank
Delivery
Total
60
$60
$160
$45
$265
Volume (Ft3)
80
200
$75
$100
$175
$75*
$100
$100
$350
$275
39
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Currently, the primary and most cost effective method of producing large volumes
of hydrogen in industry is to steam reform natural gas.
CH4 (Natural Gas)+ H2O (Water)
40
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
It is important to note that since natural gas is the primary input, the cost of
producing hydrogen is closely tied to that of price of natural gas.
There are several experimental methods of producing hydrogen. A few technologies
that show the most promise include biological water splitting, photoelectrochemical
water splitting, reforming of biomass and wastes, and solar thermal water splitting.
Biological water splitting is the process where certain photosynthetic microbes
produce hydrogen from water in their metabolic activities using light energy.
Photobiological technology holds great promise, but because oxygen is produced
along with the hydrogen, the technology must overcome the limitation of oxygen
sensitivity of the hydrogen-evolving enzyme systems.
Photoelectrochemical water splitting is currently the cleanest known way to
produce hydrogen because it utilizes sunlight to directly split water into hydrogen
and oxygen. Currently, research is under way to identify more efficient, lower cost
materials and systems that are durable and stable against corrosion in an aqueous
environment.
Reforming of Biomass and wastes produces hydrogen via pyrolysis or gasification
of biomass resources such as agricultural residues like peanut shells; consumer
wastes including plastics and waste grease; or biomass specifically grown for energy
uses. One of the by-products of this process is hydrogen.
Finally, solar thermal water splitting much like steam reformation of natural gas, but
instead of an outside energy source, this process uses highly concentrated sunlight,
which can be used to generate the high temperatures needed to split methane into
hydrogen and carbon.
PEM fuel cells work with a thin, permeable sheet, or membrane. Hydrogen enters
the fuel cell via the anode and air (oxygen) enters via the cathode. At the anode, a
platinum coating splits the hydrogen into positively charged protons and negatively
charged electrons. The protons pass through the membrane and the electrons pass
around the membrane in an electrical current. When the protons and electrons are
reunited with oxygen at the cathode, water and heat are produced. Fuels commonly
used to obtain hydrogen can be methane, propane, gasoline, diesel, and water.
Advantages of PEM fuel cell technology are its high power density, high efficiency,
and its ability to work at relatively low temperatures of about 175 degrees
Fahrenheit (80 degrees Celsius). PEMFC are being developed for stationary
distributed generation (e.g., telecom remote terminals backup power), automotive
vehicles, air conditioning, and domestic-sized electronic equipment given their
reliability, small footprint, and decreasing cost.
41
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
A variation on this type of fuel cell is the direct methanol fuel cell (DMFC). Rather
than reforming the fuel, such as in a PEM fuel cell, DMFCs do not require any fuel
reformation. Furthermore, because methane is a heavier fuel than hydrogen, storage
is much easier and the density is a lot higher (i.e., the amount of hydrogen in a given
volume of methanol is higher than that of even highly compressed hydrogen). Also,
methanol is a more poisonous gas than hydrogen. While the fuel is convenient, this
system is not as efficient as PEMFCs. DMFCs generally produce a small amount of
power over a long period of time. Therefore, their application potential is primarily
well suited toward consumer electronics and stationary generation rather than
automotive.
A second variation on the PEM fuel cell type is the regenerative fuel cell (RFC).
This technology is fairly young and is primarily being developed by NASA. RFCs,
like PEM fuel cells, use hydrogen and oxygen to create electricity, and its emissions
are heat and water. In an RFC, solar power or some other source electrolyze the
emitted water back into hydrogen and oxygen.
42
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Figure 29. Market Model: Total Addressable 5-kW Backup Power Market
Wireless
Base Case
Share of Sites - 5 kW (assumes 40% of total)
Existing
New
Total - 5kW Sites
2003
2004
2005E
167,206
28,378
195,583
195,583
15,287
210,870
210,870
27,019
237,889
2006E
2007E
2008E
2009E
2010E
2011E
237,889
30,481
268,370
268,370
34,386
302,756
302,756
39,358
342,115
342,115
44,475
386,590
386,590
50,257
436,846
436,846
56,790
493,636
30,481
40,367
70,848
34,386
44,060
78,447
39,358
49,694
89,053
44,475
42,623
87,098
50,257
62,297
112,553
56,790
70,848
127,638
305
602
1,205
2,384
4,714
9,321
1.0%
1.8%
404
771
1.0%
1.8%
708
1,373
3.1%
1,522
3.1%
2,727
5.4%
2,284
5.4%
4,668
9.4%
5,843
9.4%
10,556
16.4%
11,628
16.4%
20,949
Wireline
Base Case
2003
Share of Wireline Sites - 5 kW (assumes 20% of total)
Existing
New
Total - 5kW Wireline Sites
37,518
488
38,005
2004
38,005
494
38,500
2005E
38,500
500
39,000
2006E
2007E
2010E
2011E
39,507
514
40,021
40,021
520
40,541
40,541
527
41,068
41,068
534
41,602
41,602
541
42,143
507
8,279
8,786
514
8,387
8,901
520
8,562
9,083
527
8,674
9,201
534
8,786
9,320
541
8,901
9,441
5
1.0%
2009E
39,000
507
39,507
2008E
9
1.8%
16
28
50
3.1%
5.4%
9.4%
16.4%
89
1,461
83
147
262
465
824
1.0%
1.8%
3.1%
5.4%
9.4%
16.4%
88
156
278
493
874
1,550
Utility
Base Case
2003
Share of Substations - 5 kW (assumes 85% of total)
Existing
New
Total - 5kW Substations
Potential market size for a 5kW backup power unit
Additions - New Substations
Additions - Replacement Substations
Addressable substation backup power market
Substations Captured by Fuel Cell Sales (5kW)
New Substations Captured
Assumption
80,097
1,602
81,699
2004
81,699
1,634
83,333
2005E
83,333
1,667
85,000
2006E
2007E
2008E
2009E
2010E
2011E
85,000
1,700
86,700
86,700
1,734
88,434
88,434
1,769
90,203
90,203
1,804
92,007
92,007
1,840
93,847
93,847
1,877
95,724
1,700
15,080
16,780
1,734
15,381
17,115
1,769
16,128
17,897
1,804
16,451
18,255
1,840
16,780
18,620
1,877
17,115
18,992
17
30
54
97
173
308
1.0%
1.8%
3.1%
5.4%
9.4%
16.4%
151
269
494
882
1.0%
1.8%
3.1%
5.4%
168
300
548
978
1,574
9.4%
1,746
2,809
16.4%
3,117
43
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Figure 29. Market Model: Total Addressable 5-kW Backup Power Market (continued)
Cable / Broadband
Base Case
Share of Broadband Sites - 5 kW (assumes 25% of total)
Existing
New
Total - 5kW Broadband Sites
2003
2004
2005E
277,500
3,561
281,061
281,108
3,607
284,715
284,762
3,654
288,416
2006E
2007E
2008E
2009E
2010E
2011E
288,464
3,702
292,166
292,214
3,750
295,964
296,013
3,799
299,811
299,861
3,848
303,709
303,759
3,898
307,657
307,708
3,949
311,657
3,702
60,453
64,155
3,750
61,239
64,989
3,799
62,519
66,318
3,848
63,332
67,180
3,898
64,155
68,053
3,949
64,989
68,938
62
110
194
11
20
35
0.3%
0.5%
0.9%
181
322
574
0.3%
0.5%
0.9%
192
341
609
1.6%
1,018
1.6%
1,080
2.8%
1,805
2.8%
1,915
4.9%
3,200
4.9%
3,394
44
2003
2004
2005E
562,321
34,028
596,349
596,395
21,022
617,418
617,465
32,841
650,305
2006E
2007E
2008E
2009E
2010E
2011E
650,353
36,390
686,743
686,791
40,384
727,175
727,224
45,446
772,670
772,719
50,654
823,373
823,423
56,529
879,952
880,003
63,157
943,159
36,390
124,179
160,569
40,384
129,067
169,452
45,446
136,904
182,350
50,654
131,079
181,733
56,529
152,017
208,546
63,157
161,852
225,009
338
661
1,310
2,570
5,046
9,912
0.9%
1.6%
819
0.7%
1,157
1,509
1.2%
2,169
2.9%
2,852
2.1%
4,163
5.1%
4,649
3.5%
7,219
8.9%
10,046
6.6%
15,092
15.7%
19,098
11.8%
29,011
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
2001
2002
2003
2006E
2007E
2008E
2009E
Year
104,288
3
127,540
3
139,338
3
162,986
3
175,725
3
198,241
3
223,642
3
252,297
3
285,096
3
322,158
3
364,039
3
411,364
3
312,864
382,620
69,756
418,014
35,394
488,958
70,944
527,175
38,217
594,722
67,547
670,925
76,202
756,891
85,966
855,287
98,396
966,474
111,187
1,092,116
125,642
1,234,091
141,975
22%
9%
17%
8%
13%
13%
13%
13%
13%
13%
13%
Growth Rate
1994-2004 CAGR: 13%
BASE CASE
2000
2001
Share of Sites - 5 kW (assumes 40% of total)
Existing
125,146 125,146
New
27,902
Total - 5kW Sites
125,146 153,048
2002
2003
153,048
14,158
167,206
167,206
28,378
195,583
2004
2004
2005E
2007E
2008E
2009E
237,889
30,481
268,370
268,370
34,386
302,756
302,756
39,358
342,115
342,115
44,475
386,590
386,590
50,257
436,846
436,846
56,790
493,636
30,481
40,367
70,848
34,386
44,060
78,447
39,358
49,694
89,053
44,475
42,623
87,098
50,257
62,297
112,553
56,790
70,848
127,638
33,639
40,367
36,717
44,060
41,412
49,694
35,519
42,623
51,914
62,297
59,040
70,848
1,205
2,384
4,714
9,321
210,870
27,019
237,889
2010E
2011E
2006E
195,583
15,287
210,870
2005E
2010E
2011E
305
602
1.0%
1.8%
404
771
1.0%
708
1.8%
1,373
3.1%
1,522
3.1%
2,727
5.4%
2,284
5.4%
4,668
9.4%
5,843
9.4%
10,556
16.4%
11,628
16.4%
20,949
WORST CASE
2000
2001
Share of Sites - 5 kW (assumes 30% of total)
Existing
93,859
93,859
New
20,927
Total - 5kW Sites
93,859 114,786
2002
2003
114,786
10,618
125,404
125,404
21,283
146,687
2004
2006E
2007E
2008E
2009E
178,417
22,861
201,277
201,277
25,790
227,067
227,067
29,519
256,586
256,586
33,356
289,942
289,942
37,692
327,635
327,635
42,593
370,227
22,861
30,275
53,136
25,790
33,045
58,835
29,519
37,271
66,790
33,356
31,967
65,323
37,692
46,722
84,415
42,593
53,136
95,728
33,639
30,275
36,717
33,045
41,412
37,271
35,519
31,967
51,914
46,722
59,040
53,136
1,788
3,535
6,991
146,687
11,465
158,153
2005E
158,153
20,264
178,417
2010E
2011E
229
451
904
1.0%
1.8%
3.1%
303
578
1.0%
1.8%
1,141
3.1%
5.4%
1,713
5.4%
9.4%
4,382
9.4%
16.4%
8,721
16.4%
531
1,030
2,045
3,501
7,917
15,712
22,632
29,972
52,604
25,339
32,467
57,806
28,615
36,129
64,744
31,569
30,254
61,822
34,157
42,340
76,498
35,602
44,415
80,016
45
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Figure 30. Market Model: 5-kW Wireless Backup Power Market (continued)
BEST CASE
2000
2001
Share of Sites - 5 kW (assumes 50% of total)
Existing
156,432 156,432
New
34,878
Total - 5kW Sites
156,432 191,310
2002
2003
191,310
17,697
209,007
209,007
35,472
244,479
2004
2006E
2007E
2008E
2009E
297,361
38,101
335,462
335,462
42,983
378,445
378,445
49,198
427,643
427,643
55,594
483,237
483,237
62,821
546,058
546,058
70,988
617,045
38,101
50,459
88,560
42,983
55,076
98,059
49,198
62,118
111,316
55,594
53,279
108,872
62,821
77,871
140,692
70,988
88,560
159,547
33,639
50,459
36,717
55,076
41,412
62,118
35,519
53,279
51,914
77,871
59,040
88,560
1,507
2,979
5,892
11,651
244,479
19,109
263,588
2005E
263,588
33,774
297,361
2010E
2011E
46
381
752
1.0%
1.8%
505
964
1.0%
1.8%
3.1%
1,902
3.1%
5.4%
2,855
5.4%
9.4%
7,303
9.4%
16.4%
14,535
16.4%
886
1,716
3,409
5,835
13,195
26,187
37,720
49,954
87,674
42,231
54,112
96,343
47,691
60,216
107,907
52,614
50,423
103,037
56,929
70,567
127,496
59,336
74,024
133,361
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
2000
2001
2002
2003
2004
182,805
185,181
2,376
187,589
2,407
190,027
2,439
192,498
2,470
2005E
2006E
2007E
2008E
2009E
2010E
2011E
195,000
2,502
197,535
2,535
200,103
2,568
202,704
2,601
205,339
2,635
208,009
2,669
210,713
2,704
Growth Rate
Note that wireline sites are CIR estimates for all years.
Base case
2000
Share of Wireline Sites - 5 kW (assumes 20% of total)
Existing
36,561
New
Total - 5kW Wireline Sites
36,561
1.3%
2001
2002
2003
2004
36,561
475
37,036
37,036
481
37,518
37,518
488
38,005
38,005
494
38,500
2005E
38,500
500
39,000
2006E
1.3%
2007E
1.3%
2008E
1.3%
2009E
1.3%
2010E
1.3%
2011E
39,000
507
39,507
39,507
514
40,021
40,021
520
40,541
40,541
527
41,068
41,068
534
41,602
41,602
541
42,143
507
8,279
8,786
514
8,387
8,901
520
8,562
9,083
527
8,674
9,201
534
8,786
9,320
541
8,901
9,441
41,397
8,279
41,935
8,387
42,812
8,562
43,368
8,674
43,932
8,786
44,503
8,901
Assumption
Worst case
2000
Share of Wireline Sites - 5 kW (assumes 15% of total)
Existing
27,421
New
Total - 5kW Wireline Sites
27,421
2001
2002
2003
2004
27,421
356
27,777
27,777
361
28,138
28,138
366
28,504
28,504
371
28,875
2005E
28,875
375
29,250
16
28
50
1.0%
1.8%
3.1%
5.4%
9.4%
16.4%
83
147
262
465
824
1,461
1.0%
1.8%
3.1%
5.4%
9.4%
16.4%
88
156
278
493
874
1,550
502
8,197
8,699
505
8,240
8,745
504
8,300
8,804
499
8,209
8,708
484
7,962
8,446
452
7,440
7,892
2006E
2007E
2008E
2009E
2010E
89
2011E
29,250
380
29,630
29,630
385
30,015
30,015
390
30,406
30,406
395
30,801
30,801
400
31,201
31,201
406
31,607
380
6,210
6,590
385
6,290
6,675
390
6,422
6,812
395
6,505
6,901
400
6,590
6,990
406
6,675
7,081
41,397
6,210
41,935
6,290
42,812
6,422
43,368
6,505
43,932
6,590
44,503
6,675
Assumption
Best case
2000
Share of Wireline Sites - 5 kW (assumes 25% of total)
Existing
45,701
New
Total - 5kW Wireline Sites
45,701
2001
2002
2003
2004
45,701
594
46,295
46,295
602
46,897
46,897
610
47,507
47,507
618
48,124
2005E
48,124
626
48,750
12
21
38
1.0%
1.8%
3.1%
5.4%
9.4%
16.4%
62
110
197
349
618
1,096
1.0%
1.8%
3.1%
5.4%
9.4%
16.4%
66
117
209
370
656
1,162
376
6,147
6,524
378
6,180
6,559
378
6,225
6,603
374
6,157
6,531
363
5,972
6,335
339
5,580
5,919
2006E
2007E
2008E
2009E
2010E
67
2011E
48,750
634
49,384
49,384
642
50,026
50,026
650
50,676
50,676
659
51,335
51,335
667
52,002
52,002
676
52,678
634
10,349
10,983
642
10,484
11,126
650
10,703
11,353
659
10,842
11,501
667
10,983
11,650
676
11,126
11,802
41,397
10,349
41,935
10,484
42,812
10,703
43,368
10,842
43,932
10,983
44,503
11,126
6
1.0%
11
20
35
63
111
1.8%
3.1%
5.4%
9.4%
16.4%
103
183
328
581
1.0%
1.8%
3.1%
5.4%
110
195
348
616
1,030
1,093
1,937
627
10,246
10,873
631
10,300
10,931
630
10,375
11,006
623
10,261
10,885
605
9,953
10,558
565
9,300
9,865
9.4%
1,826
16.4%
47
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Figure 32. Market Model: 5-kW Utility Substation Backup Power Market
Year
2000
2001
90,573
92,385
1,811
2002
94,232
1,848
2003
96,117
1,885
2004
98,039
1,922
2005E
100,000
1,961
Growth Rate
Note: 2000-2004 are CIR estimates
Base case
2000
2001
Share of Substations - 5 kW (assumes 85% of total)
Existing
76,987
76,987
New
1,540
Total - 5kW Substations
76,987
78,527
2002
78,527
1,571
80,097
2003
80,097
1,602
81,699
2004
81,699
1,634
83,333
2005E
83,333
1,667
85,000
2006E
2008E
2009E
2010E
2011E
102,000
2,000
2007E
104,040
2,040
106,121
2,081
108,243
2,122
110,408
2,165
112,616
2,208
2%
2%
2%
2%
2%
2%
2006E
2007E
2008E
2009E
2010E
2011E
85,000
1,700
86,700
86,700
1,734
88,434
88,434
1,769
90,203
90,203
1,804
92,007
92,007
1,840
93,847
93,847
1,877
95,724
1,700
15,080
16,780
1,734
15,381
17,115
1,769
16,128
17,897
1,804
16,451
18,255
1,840
16,780
18,620
1,877
17,115
18,992
17,741
15,080
18,095
15,381
18,974
16,128
19,354
16,451
19,741
16,780
20,135
17,115
Worst case
2000
2001
Share of Substations - 5 kW (assumes 75% of total)
Existing
67,930
67,930
New
1,359
Total - 5kW Substations
67,930
69,288
2002
97
173
308
5.4%
9.4%
16.4%
151
269
494
882
1.0%
1.8%
3.1%
5.4%
168
300
548
978
1,746
3,117
1,683
14,929
16,612
1,704
15,112
16,816
1,715
15,634
17,349
1,707
15,569
17,276
1,668
15,206
16,873
1,569
14,306
15,875
2007E
9.4%
2,809
16.4%
2009E
2010E
78,030
1,561
79,591
79,591
1,592
81,182
81,182
1,624
82,806
82,806
1,656
84,462
1,500
13,305
14,805
1,530
13,572
15,102
1,561
14,231
15,791
1,592
14,515
16,107
1,624
14,805
16,429
1,656
15,102
16,758
17,741
13,305
18,095
13,572
18,974
14,231
19,354
14,515
19,741
14,805
20,135
15,102
73,529
1,471
75,000
2006E
1,574
2008E
72,088
1,442
73,529
2005E
54
3.1%
76,500
1,530
78,030
70,674
1,413
72,088
2004
30
1.8%
75,000
1,500
76,500
69,288
1,386
70,674
2003
17
1.0%
2011E
Best case
2000
2001
Share of Substations - 5 kW (assumes 90% of total)
Existing
81,516
81,516
New
1,630
Total - 5kW Substations
81,516
83,146
2002
85
152
272
5.4%
9.4%
16.4%
133
238
436
778
1.0%
1.8%
3.1%
5.4%
148
264
484
863
1,541
2,750
1,485
13,172
14,657
1,503
13,334
14,837
1,513
13,795
15,308
1,507
13,737
15,244
1,471
13,417
14,888
1,384
12,623
14,007
2007E
9.4%
2,479
16.4%
2009E
2010E
2011E
93,636
1,873
95,509
95,509
1,910
97,419
97,419
1,948
99,367
99,367
1,987
101,355
1,800
15,967
17,767
1,836
16,286
18,122
1,873
17,077
18,949
1,910
17,418
19,328
1,948
17,767
19,715
1,987
18,122
20,109
17,741
15,967
18,095
16,286
18,974
17,077
19,354
17,418
19,741
17,767
20,135
18,122
88,235
1,765
90,000
2006E
1,389
2008E
86,505
1,730
88,235
2005E
48
3.1%
91,800
1,836
93,636
84,809
1,696
86,505
2004
27
1.8%
90,000
1,800
91,800
83,146
1,663
84,809
2003
15
1.0%
48
18
32
57
102
183
326
1.0%
1.8%
3.1%
5.4%
9.4%
16.4%
160
285
523
934
1.0%
1.8%
3.1%
5.4%
1,666
178
317
580
1,036
1,849
3,301
1,782
15,807
17,589
1,804
16,001
17,805
1,815
16,554
18,369
1,808
16,485
18,293
1,766
16,100
17,866
1,661
15,148
16,809
9.4%
2,974
16.4%
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
2003
91,000,000
1,167,818
2004
92,183,000
1,183,000
2005E
93,381,379
1,198,379
2006E
94,595,337
1,213,958
2007E
95,825,076
1,229,739
2008E
97,070,802
1,245,726
2009E
98,332,723
1,261,920
2010E
99,611,048
1,278,325
2011E
100,905,992
1,294,944
260,000
850,000
1,110,000
14,245
263,380
861,050
1,124,430
14,430
266,804
872,244
1,139,048
14,618
270,272
883,583
1,153,855
14,808
273,786
895,069
1,168,855
15,000
277,345
906,705
1,184,050
15,195
280,951
918,492
1,199,443
15,393
284,603
930,433
1,215,036
15,593
288,303
942,528
1,230,831
15,795
Growth Rate
Base case
2003
Share of Broadband Sites - 5 kW (assumes 25% of total)
Existing
277,500
New
3,561
Total - 5kW Broadband Sites
281,061
1.3%
2004
281,108
3,607
284,715
2005E
284,762
3,654
288,416
2006E
1.3%
2007E
1.3%
2008E
1.3%
2009E
1.3%
2010E
1.3%
2011E
288,464
3,702
292,166
292,214
3,750
295,964
296,013
3,799
299,811
299,861
3,848
303,709
303,759
3,898
307,657
307,708
3,949
311,657
3,702
60,453
64,155
3,750
61,239
64,989
3,799
62,519
66,318
3,848
63,332
67,180
3,898
64,155
68,053
3,949
64,989
68,938
241,812
60,453
244,955
61,239
250,075
62,519
253,326
63,332
256,619
64,155
259,955
64,989
Worst case
2003
Share of Broadband Sites - 5 kW (assumes 15% of total)
Existing
166,500
New
2,137
Total - 5kW Broadband Sites
168,637
Potential market size for a 5kW backup power unit
Additions - New Broadband Sites
Additions - Replacement Broadband Sites
Addressable cable/broadband backup power market
Cumulative replacement sites - total market
Cumulative sites - 15% of total market
2004
168,665
2,164
170,829
2005E
170,857
2,193
173,050
11
20
35
62
110
194
0.3%
0.5%
0.9%
1.6%
2.8%
4.9%
181
322
574
0.3%
0.5%
0.9%
192
341
609
1,080
1,915
3,394
3,691
60,272
63,962
3,730
60,917
64,648
3,764
61,944
65,708
3,786
62,313
66,100
3,789
62,350
66,138
3,754
61,789
65,543
2006E
2007E
2008E
1,018
1.6%
2009E
1,805
2.8%
2010E
3,200
4.9%
2011E
173,078
2,221
175,299
175,328
2,250
177,578
177,608
2,279
179,887
179,916
2,309
182,225
182,255
2,339
184,594
184,625
2,369
186,994
2,221
36,272
38,493
2,250
36,743
38,993
2,279
37,511
39,791
2,309
37,999
40,308
2,339
38,493
40,832
2,369
38,993
41,363
241,812
36,272
244,955
36,743
250,075
37,511
253,326
37,999
256,619
38,493
259,955
38,993
7
0.3%
12
21
37
66
117
0.5%
0.9%
1.6%
2.8%
4.9%
109
193
345
611
0.3%
0.5%
0.9%
1.6%
1,083
115
205
366
648
1,149
2,037
2,214
36,163
38,377
2,238
36,550
38,789
2,258
37,167
39,425
2,272
37,388
39,660
2,273
37,410
39,683
2,253
37,073
39,326
2.8%
1,920
4.9%
49
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Figure 33. Market Model: 5-kW Cable/Broadband Backup Power Market (continued)
Best case
2003
Share of Broadband Sites - 5 kW (assumes 40% of total)
Existing
444,000
New
5,698
Total - 5kW Broadband Sites
449,698
2004
449,772
5,772
455,544
2005E
455,619
5,847
461,466
2006E
2007E
2008E
2009E
2010E
2011E
461,542
5,923
467,465
467,542
6,000
473,542
473,620
6,078
479,698
479,777
6,157
485,934
486,014
6,237
492,251
492,333
6,318
498,651
5,923
96,725
102,648
6,000
97,982
103,982
6,078
100,030
106,108
6,157
101,330
107,488
6,237
102,648
108,885
6,318
103,982
110,300
241,812
96,725
244,955
97,982
250,075
100,030
253,326
101,330
256,619
102,648
259,955
103,982
50
18
32
56
99
175
311
0.3%
0.5%
0.9%
1.6%
2.8%
4.9%
290
514
919
0.3%
0.5%
0.9%
1,629
308
546
975
1,728
3,064
5,431
5,905
96,435
102,340
5,969
97,468
103,436
6,022
99,111
105,133
6,058
99,701
105,759
6,062
99,760
105,821
6,007
98,862
104,869
1.6%
2,888
2.8%
5,120
4.9%
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
We derived our $11 target price for Plug Power based on a discounted cash flow (DCF) model to 2014. We
have isolated our methodology to a DCF model given that the company has neither commercial sales nor
recurring cash flows that are predictable today and that Plug Power remains a development-stage company.
Note our DCF discount period uses ten years to account for the time when we believe PLUG will trade at a
multiple that is more in line with comparable industrial manufacturers.
DCF methodology: With that in mind, we base our DCF model on market assumptions and PLUGs
anticipated share for residential and small commercial fuel cell markets through 2014. After determining our
projections for free cash flow to 2014, we apply an 18% discount rate to the forecast period and a terminal 10x
EBITDA multiple to achieve our $11 target price (assumes 73 million shares). We believe the 10x multiple
adequately reflects a mature and highly profitable industrial company based on comparable industrial peers.
Discount rate explanation: The 18% discount rate assumes a 5% risk-free rate supplemented by a 6% equity
risk premium. Subsequently, we add a further 7% to adjust for the early stage of the industry, cash
consumption, and uncertainty in the timing of commercial product introductions. Note that our discount rate is
higher than an academically derived weighted average cost of capital of 13.08%, which incorporates a beta of
1.67 (sourced from Bloomberg, two-year adjusted), a risk-free rate of 5.28%, and an average 20-year market
return of 13.65%. We use a two-year beta given that the extreme volatility of the stocks between 1999 and
2001 is removed and skews the beta disproportionately upward. We use a generic 18% discount rate across
our fuel cell coverage list that places the companies on an even playing field but is differentiated by the
assumptions and free cash flows that fall out of our forecasts. We believe that all the companies face specific
risks, including market development and adoption, cost reduction, and reliability, none of which has been
definitively borne out by any specific company in our universe as of today. These costs have been
incorporated into our cash flow forecasts and thus our valuation analysis.
We should also note that we see less risk in Plug Powers business model today compared with some of its
peers for the following reasons: 1) Roger Saillant has executed a successful turnaround of the company since
2001 that has included a renewed General Electric distribution agreement, a 30%-plus reduction in the system
cost each year for the past two years and an expanded management team with extensive product introductions;
2) extensive development agreements with GE, Detroit Edison, and Honda Motor; 3) the most comprehensive
field test program of all fuel cell companies, with more than 200 systems having real-work experience
achieving greater than 95% reliability levels; and 4) consistently reduced cash burn rates.
Enterprise value/sales: To supplement our DCF-based valuation, we also employ an enterprise value/sales
multiple approach. Applying an approximate 9x multiple (high end of our fuel cell universe range of 4x9x) to
our 2007 per share revenue forecast (the year we expect PLUG to generate significant commercial sales) of
$1.39 and using an 18% discount rate yields a per share present value of $9.00. We believe the high end of the
range is warranted given that the end markets that PLUG is targeting (i.e., telecom backup) grow at above
average rates. Adding approximately $0.70 per share of cash and cash equivalents (amount reported at the end
of second quarter 2005) to our discounted price/sales result yields a valuation for PLUG of approximately $10
a share, further bolstering our DCF-driven 12-month target price of $11.00.
Risks
We rate PLUG stock Speculative risk because of the development-stage nature of the business, the ongoing
cash burn on a recurring basis for the foreseeable future, and the need to meet development milestones for
success in fuel cell markets. Importantly, we stress that PLUG is a development-stage company that does not
yet have commercial products available and accordingly does not record recurrent revenue streams typical of a
commercially viable entity. As such, the company does not generate operating cash flow and is a net
consumer of cash. We expect this to continue through 2005 and into 2006. The company may need to raise
additional funds in the future to continue product development, but the new management team has been
extremely progressive in lowering the cash burn rate. Similarly, there remain no guarantees that fuel cell
technology issues can be overcome, and mass market demand is yet unproven.
51
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Specific risks to Plug Power stock not meeting our target price could occur on specific company-related news
and macro-driven items. In particular, we expect that if the company does not meet its operating cash burn
targets, there could be downward pressure on the stock price. Similarly, if the company does not attain wider
scale market acceptance of the GenCore product, the shares may come under pressure. Lastly, the loss of a
strategic partner (i.e., GE, DTE Energy, or Honda Motor) would be viewed as a negative, in our view. In
terms of the macro environment, if negative news on government funding or acceptance of fuel cell
technologies were to emerge, we believe the stock and the sector would react negatively and act as a detriment
in achieving our target price.
If the impact on the company from any of these factors proves to be greater than we anticipate, the stock will
likely have difficulty achieving our financial and price targets.
Hydrogenics (HYGS-$3.45; 1S)
Valuation
We derived our $7.50 target price for Hydrogenics based on a discounted cash flow (DCF) model to 2014.
Our DCF discount period is ten years to account for the time when we believe HYGS will trade at a multiple
that is more in line with comparable industrial manufacturers. We have isolated our methodology to a DCF
model given that the company has neither significant commercial sales nor recurring cash flows that are
predictable today and that Hydrogenics remains a development-stage company. In addition, although the
company recently added approximately $45 million in sales with the Stuart acquisition, the impact on our DCF
model is minimal given that most of the value of the new Hydrogenics is encompassed in our terminal value
calculation.
DCF methodology: With that in mind, we base our DCF model on market assumptions and HYGSs
anticipated share for residential and small commercial fuel cell markets through 2014. After determining our
projections for free cash flow to 2014, we apply an 18% discount rate to the forecast period and a terminal 10x
EBITDA multiple to achieve our $7.50 target price (assumes about 91 million shares). We believe the 10x
multiple adequately reflects a mature and highly profitable industrial company based on comparable industrial
peers.
Discount rate explanation: The 18% discount rate assumes a 5% risk-free rate supplemented by a 6% equity
risk premium. Subsequently, we add a further 7% to adjust for the early stage of the industry, cash
consumption, and uncertainty in the timing of commercial product introductions. Note that our discount rate is
higher than an academically derived weighted average cost of capital of 13.24%, which incorporates a beta of
1.49 (sourced from Bloomberg, two-year adjusted), a risk-free rate of 5.28%, and an average 20-year market
return of 13.65%. We use a two-year beta given that the extreme volatility of the stocks between 1999 and
2001 is removed and skews the beta disproportionately upward. We use a generic 18% discount rate across
our fuel cell coverage list that places the companies on an even playing field but is differentiated by the
assumptions and free cash flows that fall out of our forecasts. We believe that all the companies face specific
risks including market development and adoption, cost reduction, and reliability, none of which has been
definitively borne out by any specific company in our universe as of today. These costs have been
incorporated into our cash flow forecasts and thus our valuation analysis.
We should also note that we see less risk in Hydrogenics business model today compared with some of its
peers for the following reasons: 1) the commercial test and control business has allowed the company to
approach turning cash flow positive ahead of any of its standalone fuel cell energy peers, thus lowering the risk
of a going concern issue; 2) the stationary and motive power module business Hydrogenics is focusing on
should begin releasing commercial products in the coming 2436 months alternatively, we do not expect
automotive markets to develop until the 201520 time frame; and 3) Hydrogenics has developed strategic
customer relationships with several blue-chip industrial companies that are developing fuel cells today,
including General Motors, Deere & Co., and Dow Chemical.
Enterprise value/sales: To supplement our DCF-based valuation, we also employ a enterprise value/sales
multiple approach. Applying a 5.3x multiple (current mean price/sales multiple for our fuel cell universe) to
our 2007 per share revenue forecast (the year we expect HYGS to generate significant commercial sales) of
$2.17 and using an 18% discount rate yields a new per share present value of $7.00. Adding approximately
$1.07 per share of cash and cash equivalents (amount reported at the end of second quarter 2005) to our
52
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
discounted price/sales result yields a valuation for HYGS of approximately $8.00 a share, further bolstering
our DCF-driven 12-month target price of $7.50.
Risks
We rate HYGS stock Speculative risk because of the development-stage nature of the business, the uncertainty
of recurring cash flows, and the need to meet development milestones for success. Similar to other fuel cell
companies we cover, the primary long-term risk to HYGS is the timing of eventual fuel cell market
development coupled with infrastructure concerns that make exact rollout of automotive opportunities highly
uncertain. However, we believe this risk is somewhat mitigated by Hydrogenics being regarded as the industry
leader in Asia, North America, and Europe for fuel cell test and control stations and eventual diagnostic
opportunities. In the meantime, we expect that near-term uncertainty may arise in fuel cell developers, slowing
capital spending plans for fuel cell development. But the competitive race to get products to market should
help abate this concern.
Specific risks to Hydrogenics stock not meeting our target price could occur on specific company-related news
and macro-driven items. In particular, the company expects to be cash flow breakeven by the end of 200506.
Should it appear this is unlikely due to the magnitude of quarterly losses, the stock price could weaken.
Similarly, if the company does not meet its product-related milestones, the target price may come under
pressure. The loss of a strategic partner (i.e., General Motors or Deere & Co.) would also be viewed as a
negative, in our view. Finally, any negative news regarding the integration of the recently acquired Stuart
Energy could also be viewed negatively. In terms of the macro environment, if negative news on government
funding or acceptance of fuel cell technologies were to emerge, we believe the stock and the sector would react
negatively and act as a detriment in achieving our target price.
If the impact on the company from any of these factors proves to be greater than we anticipate, the stock will
likely have difficulty achieving our financial and price targets.
53
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
ANALYST CERTIFICATION
Appendix A-1
I, David Smith, research analyst and author of this report, hereby certify that all of the views expressed in this research report accurately
reflect my personal views about any and all of the subject issuer(s) or securities. I also certify that no part of my compensation was, is,
or will be directly or indirectly related to the specific recommendation(s) or view(s) in this report.
IMPORTANT DISCLOSURES
Hydrogenics Corp. (HYGS)
Ratings and Target Price History - Fundamental Research
Analyst: David Smith
USD
3
1
6
5
4
4
3
Target
Closing
#
Date
Rating
Price
Price
1: 6 Sep 02 1S
8.00
3.63
2: 6 Jan 03 1S
8.00
3.88
3: 7 Aug 03 1S
*6.00
4.15
4: 5 Sep 03 1S
*7.50
4.98
5: 12 Sep 03 Stock rating system changed
6: 12 Sep 03 *1S
*7.50
4.82
*Indicates change.
2
SO ND J FMAM J J A S O ND J FM AM J J A S ON D J FM AM J J A
2003
2004
2005
Covered
Not covered
USD
10
7 8
6
1
3
2
4
10
Target
Closing
#
Date
Rating
Price
Price
1: 6 Sep 02 *1S
10.00
5.80
2: 3 Jan 03 1S
10.00
4.80
3: 25 Feb 03 1S
*8.00
5.71
4: 31 Jul 03 1S
*6.50
4.41
5: 5 Sep 03 1S
*7.50
5.15
6: 12 Sep 03 Stock rating system changed
7: 12 Sep 03 *1S
*7.50
4.99
8: 22 Oct 03 1S
*9.00
6.36
9: 28 Jan 04 1S
*13.00
9.50
10: 23 Feb 05 1S
*11.00
6.95
*Indicates change.
2
SO ND J FMAM J J A S O ND J FM AM J J A S ON D J FM AM J J A
2003
2004
2005
Covered
Not covered
Customers of the Firm in the United States can receive independent, third-party research on the company or companies covered in this
report, at no cost to them, where such research is available. Customers can access this independent research at
http://www.smithbarney.com (for retail clients) or http://www.citigroupgeo.com (for institutional clients) or can call (866) 836-9542 to
request a copy of this research.
Citigroup Global Markets Inc. or its affiliates beneficially owns 1% or more of any class of common equity securities of Hydrogenics Corp.
and Plug Power Inc. This position reflects information available as of the prior business day.
Within the past 12 months, Citigroup Global Markets Inc. or its affiliates has acted as manager or co-manager of a public offering of
securities of Plug Power Inc.
Citigroup Global Markets Inc. or its affiliates has received compensation for investment banking services provided within the past 12
months from Plug Power Inc.
Citigroup Global Markets Inc. currently has, or had within the past 12 months, the following company(ies) as investment banking client(s):
Hydrogenics Corp. and Plug Power Inc.
Citigroup Global Markets Inc. currently has, or had within the past 12 months, the following company(ies) as clients, and the services
provided were non-investment-banking, securities-related: Hydrogenics Corp.
Citigroup Global Markets Inc. or an affiliate received compensation in the past 12 months from Hydrogenics Corp. and Plug Power Inc.
Analysts' compensation is determined based upon activities and services intended to benefit the investor clients of Citigroup Global
Markets Inc. and its affiliates ("the Firm"). Like all Firm employees, analysts receive compensation that is impacted by overall firm
profitability, which includes revenues from, among other business units, the Private Client Division, Institutional Equities, and Investment
Banking.
54
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
The Firm is a market maker in the publicly traded equity securities of Hydrogenics Corp. and Plug Power Inc.
OTHER DISCLOSURES
Within the past 5 years, Citigroup Global Markets Inc. or its affiliates has acted as manager or co manager of a public offering of equity
securities of Hydrogenics Corp. and Plug Power Inc.
For securities recommended in this report in which the Firm is not a market maker, the Firm is a liquidity provider in the issuers' financial
instruments and may act as principal in connection with such transactions. The Firm is a regular issuer of traded financial instruments
linked to securities that may have been recommended in this report. The Firm regularly trades in the securities of the subject company(ies)
discussed in this report. The Firm may engage in securities transactions in a manner inconsistent with this research report and, with
respect to securities covered by this report, will buy or sell from customers on a principal basis.
Securities recommended, offered, or sold by the Firm: (i) are not insured by the Federal Deposit Insurance Corporation; (ii) are not
deposits or other obligations of any insured depository institution (including Citibank); and (iii) are subject to investment risks, including the
possible loss of the principal amount invested. Although information has been obtained from and is based upon sources that the Firm
believes to be reliable, we do not guarantee its accuracy and it may be incomplete and condensed. Note, however, that the Firm has taken
all reasonable steps to determine the accuracy and completeness of the disclosures made in the Important Disclosures section of this
report. In producing its research reports, members of the Firm's research department may have received assistance from the subject
company(ies) referred to in this report. Any such assistance may have included access to sites owned, leased or otherwise operated or
controlled by the issuers and meetings with management, employees or other parties associated with the subject company(ies). Firm
policy prohibits research analysts from sending draft research to subject companies. However, it should be presumed that the author of
this report has had discussions with the subject company to ensure factual accuracy prior to publication. All opinions, projections and
estimates constitute the judgment of the author as of the date of the report and are subject to change without notice. Prices and availability
of financial instruments also are subject to change without notice. Although Citigroup Investment Research does not set a predetermined
frequency for publication, if this is a fundamental research report, it is the intention of Citigroup Investment Research to provide research
coverage of this/these issuer(s), including in response to news affecting this issuer, subject to applicable quiet periods and capacity
constraints. This report is for informational purposes only and is not intended as an offer or solicitation for the purchase or sale of a
security. Any decision to purchase securities mentioned in this research must take into account existing public information on such security
or any registered prospectus.
55
Switch Signals: Fuel Cells in Distributed Telecom Backup August 24, 2005
Investing in non-U.S. securities, including ADRs, may entail certain risks. The securities of non-U.S. issuers may not be registered with,
nor be subject to the reporting requirements of the U.S. Securities and Exchange Commission. There may be limited information available
on foreign securities. Foreign companies are generally not subject to uniform audit and reporting standards, practices and requirements
comparable to those in the U.S. Securities of some foreign companies may be less liquid and their prices more volatile than securities of
comparable U.S. companies. In addition, exchange rate movements may have an adverse effect on the value of an investment in a foreign
stock and its corresponding dividend payment for U.S. investors. Net dividends to ADR investors are estimated, using withholding tax
rates conventions, deemed accurate, but investors are urged to consult their tax advisor for exact dividend computations. Investors who
have received this report from the Firm may be prohibited in certain states or other jurisdictions from purchasing securities mentioned in
this report from the Firm. Please ask your Financial Consultant for additional details. Citigroup Global Markets Inc. takes responsibility for
this report in the United States. Any orders by non-US investors resulting from the information contained in this report may be placed only
through Citigroup Global Markets Inc.
This report is made available in Australia to wholesale clients through Citigroup Global Markets Australia Pty Ltd. (ABN 64 003 114 832
and AFSL No. 240992) and to retail clients through Citigroup Wealth Advisors Pty Ltd. (ABN 19 009 145 555 and AFSL No. 240813),
Participants of the ASX Group and regulated by the Australian Securities & Investments Commission. If this publication is being made
available in certain provinces of Canada by Citigroup Global Markets (Canada) Inc. ("CGM Canada"), CGM Canada has approved this
publication. This report may not be distributed to private clients in Germany. This report is distributed in Germany by Citigroup Global
Markets Deutschland AG & Co. KGaA, which is regulated by Bundesanstalt fuer Finanzdienstleistungsaufsicht (BaFin). If this report is
made available in Hong Kong by, or on behalf of, Citigroup Global Markets Asia Ltd., it is attributable to Citigroup Global Markets Asia Ltd.,
Citibank Tower, Citibank Plaza, 3 Garden Road, Hong Kong. Citigroup Global Markets Asia Ltd. is regulated by Hong Kong Securities and
Futures Commission. If this report is made available in Hong Kong by The Citigroup Private Bank to its clients, it is attributable to Citibank
N.A., Citibank Tower, Citibank Plaza, 3 Garden Road, Hong Kong. The Citigroup Private Bank and Citibank N.A. is regulated by the Hong
Kong Monetary Authority. This publication is made available in India by Citigroup Global Markets India Private Limited, which is regulated
by Securities and Exchange Board of India. If this report was prepared by Citigroup Investment Research and distributed in Japan by
Nikko Citigroup Ltd., it is being so distributed under license. Nikko Citigroup Limited is regulated by Financial Services Agency, Securities
and Exchange Surveillance Commission, Japan Securities Dealers Association, Tokyo Stock Exchange and Osaka Securities Exchange.
This publication is made available in Korea by Citigroup Global Markets Korea Securities Ltd., which is regulated by Financial Supervisory
Commission and the Financial Supervisory Service. This publication is made available in Malaysia by Citigroup Global Markets Malaysia
Sdn Bhd, which is regulated by Malaysia Securities Commission. This publication is made available in Mexico by Acciones y Valores
Banamex, S.A. De C. V., Casa de Bolsa, which is regulated by Comision Nacional Bancaria y de Valores. In New Zealand this report is
made available through Citigroup Global Markets New Zealand Ltd., a Participant of the New Zealand Exchange Limited and regulated by
the New Zealand Securities Commission. This publication is made available in Poland by Dom Maklerski Banku Handlowego SA an
indirect subsidiary of Citigroup Inc., which is regulated by Komisja Papierw Wartosciowych i Gield. This publication is made available in
the Russian Federation through ZAO Citibank, which is licensed to carry out banking activities in the Russian Federation in accordance
with the general banking license issued by the Central Bank of the Russian Federation and brokerage activities in accordance with the
license issued by the Federal Service for Financial Markets. Neither this report nor any information contained in this report shall be
considered as advertising the securities mentioned in this report within the territory of the Russian Federation or outside the Russian
Federation. This report does not constitute an appraisal within the meaning of the Federal Law of the Russian Federation of 29 July 1998
No. 135-FZ (as amended) On Appraisal Activities in the Russian Federation. This publication is made available in Singapore through
Citigroup Global Markets Singapore Pte. Ltd., a Capital Markets Services Licence holder, and regulated by Monetary Authority of
Singapore. Citigroup Global Markets (Pty) Ltd. is incorporated in the Republic of South Africa (company registration number
2000/025866/07) and its registered office is at 145 West Street, Johannesburg 2196. Citigroup Global Markets (Pty) Ltd. is regulated by
JSE Securities Exchange South Africa, South African Reserve Bank and the Financial Services Board. The investments and services
contained herein are not available to private customers in South Africa. This publication is made available in Taiwan through Citigroup
Securities Investment Consulting Inc. which is regulated by Securities & Future Bureau. This publication is made available in United
Kingdom by Citigroup Global Markets Limited, which is regulated by Financial Services Authority. This material may relate to investments
or services of a person outside of the UK or to other matters which are not regulated by the FSA and further details as to where this may
be the case are available upon request in respect of this material. FSA rules require that a firm must establish, implement and make
available a policy for managing conflicts of interest arising as a result of publication or distribution of investment research. The policy
applicable to Citigroup's equity research products can be found at www.citigroupgeo.com. This publication is made available in United
States by Citigroup Global Markets Inc, which is regulated by NASD, NYSE and the US Securities and Exchange Commission. Unless
specified to the contrary, within EU Member States, this publication is made available by Citigroup Global Markets Limited, which is
regulated by Financial Services Authority. Compensation of equity research analysts is determined by equity research management and
Citigroup's senior management and is not linked to specific transactions or recommendations. This report may have been distributed
simultaneously, in multiple formats, to the Firm's worldwide institutional and retail customers. This document is not to be construed as
providing investment services in any jurisdiction where the provision of such services would be illegal. Subject to the nature and contents
of this document, the investments described herein are subject to fluctuations in price and/or value and investors may get back less than
originally invested. Certain high-volatility investments can be subject to sudden and large falls in value that could equal or exceed the
amount invested. Certain investments contained herein may have tax implications for private customers whereby levels and basis of
taxation may be subject to change. If in doubt, investors should seek advice from a tax adviser. This advice has been prepared without
taking account of the objectives, financial situation or needs of any particular investor. Accordingly, investors should, before acting on the
advice, consider the appropriateness of the advice, having regard to their objectives, financial situation and needs.
2005 Citigroup Global Markets Inc. Member of SIPC. Citigroup Investment Research is a division and service mark of Citigroup Global
Markets Inc. and its affiliates and is used and registered throughout the world. Citigroup and the Umbrella Device are trademarks and
service marks of Citicorp or its affiliates and are used and registered throughout the world. Nikko is a registered trademark of Nikko Cordial
Corporation. All rights reserved. Any unauthorized use, duplication, redistribution or disclosure is prohibited by law and will result in
prosecution. The Firm accepts no liability whatsoever for the actions of third parties. The Firm makes no representations or warranties
whatsoever as to the data and information provided in any third party referenced website and shall have no liability or responsibility arising
out of, or in connection with, any such referenced website.
ADDITIONAL INFORMATION IS AVAILABLE UPON REQUEST
56
US08P108