Long Copa Holdings (CPA), short LAN Airlines (LFL). We are recommending this Latin American airline pair trade, as we believe these are two highly comparable companies where an unwarranted (and unprecedented) valuation disparity has emerged.

Based on consensus estimates, CPA is trading at 8.6x 11E P/E, while LFL is at 18.9x. On an EV/EBITDAR basis, CPA is at 7.5x while LFL is at 10.7x. In our opinion, LFL's wide premium to CPA is incosistent with the fact that CPA is expected to grow EPS at a 17.4% CAGR from 2010-12, while LFL should grow at a marginally higher 19.8%.

Background

CPA and LFL are two of the most profitable and best-managed airlines in the world. Their LTM EBITDA margins are 21-22%, in line with Ryanair, and well ahead of LUV and JBLU (11.6% and 14.7%) or any of the Brazilian airlines (GOL and TAM at 9%).

Panama-based CPA has boasted remarkably stable EBITDA margins in the 21-23% range over the past five years. As far as airlines go, this type of consistency is unseen in any other company. CPA's model is based around a single hub in Panama City's Tocumen airport, where the company has 80% market share. Tocumen is an unparalleled asset: its size, benign weather and low altitude make it the most efficient connection hub in Central America. Thus, 50-55% of CPA's traffic comes from passengers that catch connecting flights via Tocumen to any of CPA's 46 international destinations.

14% of CPA's traffic is from their AeroRepublica segment (recently rebranded Copa Airlines Colombia). CPA acquired this domestic Colombian airline in 2005, and turned it into a feeder of connecting traffic via its Tocumen hub.

Chile-based LFL has grown from being a solely Chilean carrier to becoming a pan-South American player, ramping up local operations in Peru, Ecuador and Argentina. Thus, LFL operates a 4-hub model (Santiago, Lima, Guayaquil, Buenos Aires). LFL has wisely established its local subsidiaries in countries that lacked an incumbent flag carrier, which has obviously helped in maintaining high pricing amid limited competition. This has been the case until the recently announced (still to be closed) merger with TAM of Brazil and initiation of domestic operations in Colombia (more on this later). The other key aspect of LFL's is the importance of its cargo segment: cargo accounts for about a third of LFL total sales, while it's only 5% in the case of CPA.

LFL's rise over the past few years has not just been driven by traffic growth but also from playing margin catch-up by improving its yield management and reducing costs. This has allowed LFL to grow EBITDA margins from 8.7% in 2005 to 21.5% LTM.

Recent developments and Variant perception

While news flow over the past few months has been more favorable to LFL, we believe trading action on the two stocks represents a textbook case of over-reaction (since April, LFL has outperformed CPA by 100%). As sanity returns to investor's minds and fundamentals prevail, we believe valuation multiples should return to historical levels, making the trade work.

1. We believe the market is underestimating CPA's growth potential. While LFL is guiding passenger capacity (ASM) growth of 15-17% in 2011, which the sell-side is happily buying into, CPA still hasn't detailed its growth guidance for 2011, other than saying it would add 5 planes during the year. While 5 new B737s should bring 7-10% capacity growth, there are still analysts out there like CS with ASM growth of only 2% in 2011.

Consensus 2011 capacity growth for CPA is around 8%. However, we are fairly confident that the growth rate will be in the mid-teens (all centered on Panama flights), as the company announces new planes on top of the 5 already announced. CPA's CEO increased 2011 capex guidance from $150mm to $200-300mm last week, yet this went largely unnoticed as the market was busy worrying about CPA's September traffic numbers. CPA will announce preliminary 2011 guidance in its 3Q call in November. We believe the sell-side will be forced to raise their ASM growth estimates when the company guides to ASM growth of around 15%. Note that this is almost twice the growth rate currently assumed by the Street.

It is true that CPA is more Central America/Caribbean based and LFL is more South America-based, but both regions post very similar rates of growth and their respective airline markets show comparable competitive dynamics (limited competition, except in Colombia). The IMF recently expected Panamanian GDP to grow 6.5% next year, vs 5.5% for Chile, 7.1% for Peru, and 5.8% for Brazil. It seems reasonable to assume that air traffic and capacity growth rates should be very similar for CPA and LFL.

2. The market's been spooked by CPA's September monthly traffic release, which showed a sequential decline in load factor of 460bps to 70.4%. In the three trading days since the release came out on Oct 8^th, LFL has outperformed CPA by 12%. While load factor came in below expectations, September has always been a seasonally weak quarter for CPA. In both 2007 and 2008, September load factor fell sequentially by over 500bps, and yet it managed to bounce back to the mid-70%s in the seasonally strong months of Nov-Dec.

4Q load factors would implicitly need to reach 79.4% in order to meet full year guidance of 77%. While this number may be a bit high, the market's reaction seems to extrapolate September's load factors (and worse) into infinity, even though Nov-Dec load factors should recover well into the mid-high 70%s.

3. The market's been concerned about increased competition in CPA's Colombian operations (AeroRepublica). The aggressive entry of new player Aires, coupled with the existence of a strong incumbent (Avianca/Taca) has depressed yields in Colombia to the point that CPA's AeroRepublica is currently only breaking even on a cash basis. Our understanding is that Aires' financial situation is already stretched and might soon have to cut down on its expansion. The fact that LFL is entering the market later this year through AerOasis won't help CPA, but would obviously also dilute LFL's margins.

What the market seems to ignore here is the fact that Colombia accounts for only 17% of CPA's revenues, yet it's punished CPA as if it was the Panama-based business that was in jeopardy. Hence, you don't even have to count on Colombia yields recovering for this trade to work. Our estimates simply assume stable Colombia yields from here, in line with CPA's expectations.

4. Some investors seem to perceive CPA as an orphan company that's been left in no man's land in the midst of all the industry consolidation going on. Also, the fact that LFL chose TAM as a merger partner cut off hopes that CPA might be acquired by LFL, at least in the short term. The argument goes on that CPA is not even party to any of the big three airline alliances.

Against this argument, CPA has managed to grow and defend its business for years on the sole basis of its alliance with Continental (Copa participates in Continental's OnePass frequent flyer program, and has code-share agreements with Continental), which is currently in effect through 2015. CPA joined the SkyTeam alliance in 2007 and then exited it in 2009, following Continental's exit. CPA is currently pondering whether to join StarAlliance (which Continental/United are part of), and we expect an announcement by the end of the year which should remove the lingering overhang.

5. LFL's merger with TAM of Brazil to create LATAM, announced on Aug 13^th, was warmly welcomed by the market, even though the exchange ratio offered implies a 47% premium to TAM's share price at the time. The merger also implies LFL's entry into a much more competitive market and presents integration challenges.

We understand the rationale behind the deal, and we are familiar with similar mergers that have been successful (KLM+AF being the prime example), but we are suspicious about management's guidance of $400mm annual synergies: $290mm (72% of the total) of these synergies are expected to come from the revenue side, when time and again it's been proved that cost synergies are much easier to realize than revenue synergies in these types of airline mergers. We assume $200mm synergies by 2012.

6. LFL's revenue growth has outperformed CPA this year, driven largely by a strong bounce-back in the very cyclical cargo segment, which had tanked during the recession. Cargo traffic YTD has grown 28% for LFL, vs 10.2% in the passenger segment (CPA's traffic is up 11.5%). Coupled with recovering yields, this has driven 60.3% revenue growth for LFL cargo, and 24.5% for passenger, vs. CPA 8.9%.

While a premium on near-term earnings could've been justifiable for LFL based on leverage to a cyclical recovery in cargo, the rebound witnessed in 2010 cannot be extrapolated into the future, and thus LFL's premium valuation shouldn't be justified on this basis. A good indication is the fact that LFL plans to grow cargo capacity by 11-13% in 2011, below passenger growth of 15-17%.

Valuation

We've already gone through recent developments for both airlines, which have led 2011 EPS estimates 8% higher for LFL and 10% lower for CPA since April. Yet LFL's outperformance has been of 100%. Clearly, this has been a story of multiple expansion/compression above anything else.

It's worth charting a historical fwd P/E chart of the two companies to realize what an aberration current valuation levels have become. Based on 2011 consensus estimates, CPA's trading at 8.6x P/E while LFL is at 21.2x. CPA's 2010-12 EPS CAGR is 17.4% CAGR, while LFL's is 19.8% (again, based on consensus).

Fwd P/E chart: http://tinypic.com/view.php?pic=30sc35z&s=7

LFL standalone is trading at a 12.6 turns of P/E premium to CPA, when the historical premium has been 1-5 turns. Even if we factor in the LATAM merger and assume $200mm run-rate synergies by 2012, the premium is  8 turns. With LFL priced for perfection and CPA so depressed, we believe the margin of safety in this trade is very wide.

Because our earnings estimates are quite in line with consensus, and because 95% of the payout on this trade is just on multiple reversion to normal, I won't go into our detailed modeled assumptions, but just lay out what the relative multiples look like based on consensus estimates.

• CPA issuing 2011 guidance, including capacity growth - 3Q call
• CPA joining Star Alliance or other partnership - 4Q10
• LATAM merger moving ahead: arb spread currently 12%
• LFL cargo growth moderating