Value Investors Club / CALPINE CORP (CPN)

Description

Long CPN

As a merchant power generator/IPP, CPN’s profitability is generally driven by gas prices, as gas

prices set power prices in most power markets and forward power prices are a determinant in

capacity pricing. The sensitivity to gas prices is less than that of a coal or nuclear generator, as

CPN is a gas power plant operator. But the sensitivity is still there since CPN is an efficient gas

generator with operating heat rates below market clearing heat rates. So any thesis on CPN has

to be grounded with a view on gas prices.

Even as gas strip pricing has recently collapsed due to an oversupplied market (continued

production growth in the Marcellus and Utica), there are numerous dynamics at play that I

believe make CPN’s cash flow story attractive and sustainable. The dynamics discussed in this

write-up should support continued cash returns to shareholders via stock buybacks, regardless of

further declines in natural gas prices. Furthermore, whenever commodity markets balance, the

balancing is not smooth and production generally over-adjusts. Therefore, I see longer-term

natural gas prices as undervalued, as natural gas markets must balance in the short-term via a

production response.

CPN trades at an average ’15, ’16 free cash flow yield of 12% on current Consensus estimates. I

believe the underlying gas market dynamics at play support these cash flow estimates over the

next two years, and when combined with the magnitude of cash flow return to shareholders,

support a minimum 16-22% total return range on the stock over a 12-month horizon.

Importantly, this return estimate assumes no change to CPN’s cash flow multiple (i.e. it

continues trading at a 12% yield).

The energy market today is very fluid and uncertain, and finding anything with high conviction

is pretty difficult. But because I view CPN as having limited earnings sensitivity to natural gas

prices: a 12% free cash flow yield and a forecasted return to shareholders of 33% of its market

cap over the next 3 years make it one of the few near-term energy bets that it is palatable.

Summary:

- Northeast natural gas production growth will have to slow down in order to balance an

oversupplied gas market in 2015 and 2016. Even a Consensus slowdown in 2016 L-48

gas production growth to 2.5 Bcf/d from 3.8 this year will still leave Oct’16 storage at

very high levels and will again require incremental YoY C-to-G switching….a lot of

demand elasticity required of the power market

- Gas supply/demand models require substantial incremental YoY CtoG switching in ‘15 at

a time when:

1. Gas plant power burn already steals market share when gas is below $4.00

2. Coal is oversupplied and incremental gas demand from power gen needs

to today be done at more competitive PRB coal prices

- Oil and NGL prices have driven down well IRR’s and E&P cash flows, meaning break-

even gas prices from liquids-rich wells are much higher today. But the slowdown in oil

production (and associated gas production) is taking too long, implying substantial risk

for Oct ’15 storage levels

- Even if we get a material slowdown in associated gas production growth, it does not seem

to be enough to balance the market on its own; i.e. NE production growth is too high and

we’re already requiring a lot from CtoG switching.

Conclusion #1: there is meaningful risk that gas prices need to fall to levels that curtail NE gas

production growth in ’15 and ’16.

Conclusion #2: Gas production growth needs to slow now in order to balance markets near-term,

at time when gas demand begins to pick-up in 2016+. The production response at the bottom of

a commodity cycle always has risk of overshooting to the downside, meaning long-term gas strip

pricing seems undervalued after the recent correction. This is a positive for merchant power

companies

The problem is the natural gas balancing mechanism

According to Bentek estimates, gas-fired power burn for base-load generation reached a record

high of around 23 Bcf/d in Jan and Feb ’15 (when henry hub prices averaged $2.91). But we still

need more switching as total power demand falls in the shoulder season and production growth is

not slowing.

Morgan Stanley’s valid point: The Northeast needs 1.5 Bcf/d YoY incremental CtoG to balance

the regional market this year, and when combined with the fact that infrastructure is tight out-of-

basin, gas will need to find new NE pricing hubs to stimulate demand. But Northeastern coal

prices are already out of the money relative to gas:

Source: Barclays Research

So where does incremental demand in the NE come from? Coal retirements via MATS is one

area, but at least one sell-side shop has recently revised down its MATS retirement estimates due

to regulatory timeline extensions.

The graph above from Barclays is an interesting result of strong production growth and higher

power market share from gas via CtoG

i.e. higher HDD’s can be met by a reversal of CtoG but there is less room from incremental CtoG

when HDDs are very low.

What does this all mean and how should we interpret?

1. Because gas markets are oversupplied near-term, gas prices need to fall to a level that

encourages incremental YoY CtoG switching on a U.S.-wide basis.

a. The latest estimate from MS is a total of 2.7 Bcf/d of required incremental

YoY switching nationwide, and 1.5 Bcf/d of that will be needed in the

NE…this is on 5-6 Bcf/d of estimated current CtoG switching

b. Henry Hub will then need to compete with PRB coal prices, implying HH

prices of < $2.60 (see table above)

c. But the NE is take-away constrained near-term, implying prices may need to

fall to shut-in production (at least party via completion deferrals) absent

meaningful incremental CtoG switching in-basin.

i. Apply your own basis to sub $2.60 henry hub

ii. The fact that gas markets require another increase in YoY switching in

2016 should temporarily drive down strips gas pricing even more,

providing the incentive for natural gas E&Ps to slow activity

2. Because PJM power prices are set by coal economics, a further downdraft in NE gas

prices means gas power plant operators realize expanding spark spreads and increased

utilization through CtoG switching

a. Meaning as NE gas prices keep falling, PJM spark spreads expand as coal

plants set power prices

b. As this dynamic plays out, further declines in 2016 forward gas pricing may

not lead to tighter spark spreads in the NE because coal pricing will set a floor

on power prices

3. Given that Henry Hub prices are already close to $2.60, further declines in power

prices are some-what mitigated by the coal floor; i.e. coal begins at some point to set

power prices in Texas; but importantly, incremental CtoG switching is needed to

balance the market, so gas power plants realize increased utilization via CtoG

switching.

4. Bottom line is that gas prices need to force lower natural gas production growth,

particularly in the Northeast, if the needed CtoG doesn’t materialize at a high enough

price level

a. Gas power plants realize expanding spark spreads, where coal begins to set

power prices, and increased utilization via CtoG. Then, as the natural gas

commodity cycle develops (i.e. 2016 expected gas production comes down),

strip gas pricing then rises, which is beneficial for merchant power stocks and

sellside valuation of these stocks.

Risks to the above interpretation:

1. Oil production growth begins correcting immediately, bringing down associated gas

production growth enough to allow minimal incremental CtoG to balance gas markets

a. Consensus 2015 gas production growth is 4 Bcf/d YoY with associated gas

production growth of 2 Bcf/d, implying that if you took out the 2 Bcf/d of

associated gas growth, you would still need incremental YoY CtoG switching to

balance gas markets this year, albeit a much smaller amount

i. Or, you would need oil production to decline in ’15

ii. Or, you need NE gas production growth to slow

2. CtoG switching is sufficient to balance gas markets at a gas price that does not slow

down NE production growth, meaning we never get the upward bias in strip pricing post

2015 shoulder season.

a. This would mean that NE gas finds power demand sensitivity at < $2.60 henry

hub, but in this case, gas turbines get the incremental utilization

3. MATS retirements create a tighter power market than anticipated, increasing gas demand

more than expected YoY

a. In this case, a tighter power market means higher power prices and higher spark

spreads

4. Spark spreads decline

a. The risk is that coal prices start to come down materially in a short period of time

which requires a lower and lower gas price to displace coal in the power stack.

This drives down power prices and thus drives down spark spreads of efficient

gas plants (see more on this below)

b. However, it may be difficult to realize much lower coal pricing given implied cost

of production.

Historical coal prices:

Source for coal data/graphs: Howard Weil

The thrid graph abovce is the Supply response.

Coal is generally priced on contracts with fixed pricing for a term, making it

difficult of power plants to get competitive pricing in the short-term

Finally, the $53 per ton Appalachian coal price in the table above is very close to cost of production support levels (based on current coal demand):

Why play CPN here?

CPN is the perfect fit for the gas and power market dynamic discussed above:

CPN is a high-efficiency gas power plant company with merchant exposure in PJM and Texas.

In PJM, we have sustained spark spreads driven by coal prices setting power prices, and so

further drops in NE gas prices are all margin to CPN. In Texas, we have more power price

exposure due to more competitive coal prices (PRB), but gas prices have already corrected to

low levels and there is not much further for gas to go to start competing with Coal. The

necessity of CtoG to balance the market means higher utilization for CPN plants in Texas,

offsetting lower spark spreads; and if gas prices fall enough, CPN realizes the same margin

impact as in PJM, creating a floor on earnings downside.

While CPN is realizing increased utilization and higher spark spreads in PJM, offsetting some

margin pressure in ERCOT (Texas), the anticipated gas production response eventually lifts

longer-term gas strip pricing, which is beneficial to any power producer (where gas sets power

pricing). In other words, CPN has short-term earnings tailwinds at time when gas prices are

dropping. Then eventual strip revaluation driving long-term earnings estimates higher because

CPN has a below-market-clearing heat-rate (meaning is still long gas prices).

The attractiveness of CPN is not really earnings though: The macro dynamics simply make me

comfortable with the sustainability of cash flow generation that will support continued cash return to shareholders.

The bottom line here is that we can have a high degree of confidence in CPN’s 2015 guidance

range (the bias is upwards in my opinion given the above dynamics). But the more important

aspect is that we can have a high degree of confidence in CPN’s free cash flow yield.

CPN trades at an 11% free cash flow yield this year (free cash flow is CFO minus total capex),

but importantly has a policy of returning excess cash to shareholders via stock buybacks.

Therefore, Free cash flow per share is growing at 16% this year and 22% next year based on

Consensus EBITDA.

The way I view valuation is that the fundamental dynamics support free cash flow generation,

and the return of free cash flow generation leads to a higher growth rate in free cash flow per

share.

So applying an 11% yield to forward year free cash flow per share that is growing 16-22% means

a 16-22% total return potential (alternatively, the company is shrinking EV while growing

EBITDA). The idea that CPN still trades at an 11% yield (at least) is based on the fundamental

factors discussed above.

CPN Summary ($mm)

Current Market Cap 8,212

Current Debt 11,282

Current Cash 717

Leverage 5.31x

Current share price $22.00

Low Mid High

CPN guided 2015 EBITDA 1,900 2,000 2,100

2014 2015 2016 2017

Consensus CPN EBITDA 1,990 2,014 2,102

Consensus Capex (497) (423) (306)

Interest expense (645) (645) (645)

Working capital (50) (75) (75)

Free Cash Flow to Equity 798 871 1,076

Cash Balance 717 715 786 862

Stock Buybacks (800) (800) (1,000)

Cumulative buybacks as % of mkt cap 31.7%

Shares outstanding 375 339 303 257

Free cash flow per share $2.03 $2.35 $2.88 $4.18

Growth 16% 22.3% 45.3%

Free cash flow yield 10.7% 13.1% 19%

I do not hold a position with the issuer such as employment, directorship, or consultancy.
I and/or others I advise do not hold a material investment in the issuer's securities.

Catalyst

The catalyst here is the realization of cash flow return to shareholders made possible by visibility and realization in actual earnings. The stock price cannot ignore actual cash returns to sharedholder of this magnitude without drastically decreasing the fair EBITDA multiple at which the stock trades (i.e. EV will decline materially while EBITDA grows).

show sort by

			2015	2016
Price:	23.00	EPS	0	0
Shares Out. (in M):	375	P/E	0	0
Market Cap (in $M):	8,619	P/FCF	0	0
Net Debt (in $M):	0	EBIT	0	0
TEV (in $M):	0	TEV/EBIT	0	0