Thursday 8 September 2016

Flybe – A dog is a value investor’s best friend?

I think there is one thing we can all agree on – no one has much love for Flybe (FLYB.L). Even after a recent small bounce it has lost almost 50% of its value over the last year and is now trading at a PTBV of just 0.8 and an EV/EBITDA of 1.4.

And many would say that’s fully justified. For example a few comments from here:


The real killer for me is these persistently low level load rates despite the generally favourable conditions for air travel.’

‘They must surely be near rock bottom on load factors. Generally when LF's get to 65% is frequently game over for an airline.’

‘Also, if they barely make money with low oil prices & the economy moving along nicely, what happens when we have a recession linked to high oil prices?’

‘Anyone short or considering shorting here?’

UK Small Cap Analyst & Blogger Paul Scott also turned negative on the stock following poor Q1 trading:


‘However, for me this is one disappointment too many, and I'm currently in the process of ditching my position in the company today. The load factor dropping to 70% is worrying me, and I think it's possible that this company may never move back into proper profitability.’

Here’s the main reason for being bearish in 1 graph:

Load factor is passenger numbers divided by seat capacity. This doesn’t look like the turnaround story investors were promised. It can be hard to make money when your average plane is less than 70% full.

Things are unlikely to change soon either early indications for Q2 in the Q1 trading statement indicate that there is a risk that H1 won’t be significantly above break even.


And after all this is to be expected, Warren Buffett has given us repeated warnings against owning an airline:

"I like to think that if I'd been at Kitty Hawk in 1903 when Orville Wright took off, I would have been farsighted enough, and public-spirited enough--I owed this to future capitalists--to shoot him down. I mean, Karl Marx couldn't have done as much damage to capitalists as Orville did."

So that’s that then. £FLYB is a dog and best avoided…Or is it?

The opposing view

Flybe popped up on my radar again when it appeared as the highest risk-reward & ROOIC pick for a value investing Hedge Fund:


When smart value investors suggest a 10x upside then I certainly start to take notice.

Essentially their argument is:
  • CEO is a smart guy with extensive industry (EasyJet) and restructuring (Gores Group) experience who is committed to only flying profitable routes and will quickly take action to eliminate routes that are facing too much or loss-making competition.
  • Clean under-leveraged balance sheet following 2014 rights issue means that there is very little chance of bankruptcy and the company can easily whether periods of poorer trading due to external circumstances like terror attacks.
  • Have now dealt with and ring-fenced legacy issues surrounding leases on unprofitable E195 regional jets.
  • 2.5 years into a 3 year turnaround plan with key issues like plane utilisation and staffing levels now at industry norms.
  • Plane choice of Q400 Turboprop allows profitable routes from smaller airports at lower passenger numbers.
  • Main competition is often rail travel which is more expensive and slower in most cases.
  • Flybe has a long runway (pun intended) of short profitable routes in Europe that it can add to drive growth over the medium term.
  • Their valuation metrics on an EBITDAR, EBITDA & EBIT basis are so undemanding that a move to trade at an average for the sector would see the share price multiples of today’s price.

The Alternative Load Factor Story

Now a lot of that makes sense but on the surface doesn’t seem to line up with the story that the load factor graph is telling us. However when you graph the components of the load factor separately and apply a trend line to help remove seasonality you get a different story:


So the load factor decline is due to passenger numbers increasing at a slower rate than the seat capacity increase. According to the CEO this is as expected since a new route takes 2 years to develop the volume to become profitable and the company has added seat capacity recently.

We can actually model the lag in passenger numbers following an increase in seat capacity. Taking the Q1 figures to avoid seasonality I am going to assume a route starts at 55% load factor when opened and then hits 65% after 1 year and 75% after 2 years. I assume mature existing routes stay at 75% load factor. This is what you get when I model this against past Q1 numbers (to avoid seasonality effects I compare just Q1 to Q1):


Notice how closely the model matches the actual Q1 passenger numbers. When the seat capacity growth moderates the maths means the load factor tends to return to the 75% that I’ve assumed for a mature profitable route. This year Flybe have moderated their seat capacity growth in response to a weaker market and will only add 6% this year (calculated from the plane deliveries this year.) If I assume zero seat capacity growth after this year the load factor trajectory will probably be something like this:


i.e. a return to a 75% load factor 2 years after seat capacity growth stops.

Of course this model doesn’t prove that growth in seat capacity is the sole reason for the drop in load factor but does give some confidence that it is at least part of the explanation.

The reality is also more complex since there is a strong relationship between load factor & yield (revenue per passenger.) There is high price elasticity meaning that small changes in pricing have big impacts on passenger numbers. The company will adjust pricing to maximise revenue per flight just this revenue will be bigger for more mature routes that need less marketing for public awareness and get scale from airports.

Valuation

Flybe is very low on multiples of EBIT, EBITDA & EBITDAR compared to its peers and with those metrics likely to increase in the future they will look even better comparative value unless the share price responds. However in general I’m not a big fan of sector multiple comparisons when valuing a business since the whole market or sector could be under/over-valued or there may be differences in tax rate etc. that are not properly accounted for. For this reason I prefer to value businesses on a Discounted Cash Flow basis. It is clear that a DCF is at best an attempt to understand the factors influencing intrinsic value not a way of determining an absolute value. They are equally as dependent on assumptions as sector multiple comparisons but at least the assumptions are explicit and can be tested or varied as required. Here I am going to make some pretty conservative assumptions.

Assumptions:
  • Load factor as per above profile with 75% maximum.
  • Zero growth in seat capacity beyond this year.
  • Contract revenue flat.
  • Other operating revenue flat.
  • Fuel pricing drops 32% in line with oil price hedges next year +6% seat capacity increase.
  • Staff costs +2% for 3 years as per recently announced pay deal.
  • Airport & route charges, ground operations & maintenance track seat capacity.
  • Aircraft rental charges flat as capacity expansion is through ownership.
  • Marketing & distribution, Finance costs flat.
  • Other Operating Expenses includes things like Insurance, Travel Costs, Property Rent, etc. so it is conservative to assume this is proportional to seat capacity.
  • Cash Flow = profits over the long term. i.e. depreciation = maintenance capex and working capital flows are neutral.
  • Grounded lease cost £20m this year, £10m next year.
  • 20% UK Tax Rate
  • 15% discount factor.
  • 6.67x terminal multiple (equivalent to 15% discount factor applied to no growth scenario) from FY19/20.
  • £49.4m net cash = £62.2m net assets - £7.8m restricted cash - c.£5m share purchase (see next point)
  • Fully diluted shares in issue of 218.7m (note employee 5% share aware will be met with on-market purchases.)

This gives a fair value estimate of £1.23 or 2.3x current share price even after the recent rise from sub 40p to the current 53p.

A less conservative discount rate of 10% and hence 10x final multiple would see a fair value estimate of £1.88 or 3.6x current share price.

This is based on a management team doing the basics of running an airline right: scheduling, pricing utilisation but not much else. If we assumed that the CEO can actually do what he says and add further profitable routes to the network, oil price stays low & revenue/passenger increases slightly then 10x share price is not unreasonable.

Obviously reality will not match my model of smooth growth. Each quarter, half or year will be noisy as external factors like fuel pricing or consumer confidence provide head-winds or tail-winds. With net cash on the balance sheet though Flybe has the ability to weather these periods of poor trading and capitalise on the periods of good trading.

I see three major risks to the type of upside I am modelling here. The first is a long severe recession in the UK that would seriously impact passenger numbers across the industry for an extended period. This was certainly a risk with the Brexit vote however PMI indicators seem to have bounced back recently. I am actually surprised at the apparent strength of the UK economy in the wake of the Brexit decision. Flybe has significant economic exposure but given its higher proportion of business travel than most other low cost carriers it is much less impacted to weak sterling reducing leisure travel that could be a challenge for other low cost carriers. And of course if you were concerned about the severe recession scenario you would own no UK stocks. The second risk is another cash-rich airline competing aggressively on the shorter regional routes that Flybe fly. Unless this competitor is willing to fly Q400 or similar aircraft Flybe will have the cost advantage but as Warren Buffet alludes to we shouldn’t underestimate the willingness of airlines to burn shareholders capital on unprofitable route expansion. I think the risk of an airline competing across a large proportion of Flybe’s routes is limited by the fact that it would be cheaper to simply buy Flybe. Which is my third risk. That someone else runs these numbers and Flybe gets taken over for less than a conservative estimate of its intrinsic value.

All things considered I’m struggling to find a better risk/reward situation in the UK market at the moment. At least for investors willing to live with the volatility of short-term results.

Disclosure: Since my analysis indicates a very significant potential upside to Flybe shares I am long the equity. I may be wrong. I reserve the right to change my mind at any time if I perceive the facts have changed.

6 comments:

  1. Excellent post and I fully agree with your risk/reward analysis. it seems a great self help story which the market ignores, preferring to fret over bombs and Brexit. You also make a good point that loads are declining - but as part of a bigger plan.

    BTW For 16/17 you expect £20m of Blackbird costs: Haven't we already been told that will only be £10m?

    Many thanks and I await further updates!?

    ReplyDelete
  2. Re:second risk
    Would it be even cheaper for the competitor to just take advantage of its white label program? Instead of competing or buying Flybe.

    ReplyDelete
  3. Handel & Co,

    Yes I think you are right. I had taken my numbers from the note GreenWood sent me (You can request it from them here if you are interested: http://www.gwinvestors.com/report-request/) but looking at it again they maybe had done their model prior this announcement:

    http://www.investegate.co.uk/flybe-group-plc--flyb-/rns/flybe-announces-completion-of-project-blackbird/201511020700091255E/

    We currently anticipate that the financial cost of resolving the remaining E195s will be £20m this year, reducing to around £10m next year, and then to £6m the year after and £4m in the following year.

    If I use these figures I get a slight uplift in my NPV15 to £1.27.

    Igor,

    I must admit the white-label part of the business is one I don't have particularly good insight into. However I doubt that Flybe would fly white-label for someone else in competition to their own routes. If an airline wanted to add their own unique route and fly Q400 due to economics then Flybe could probably operate it for them. However if they wanted access to an existing Flybe route they would offer a codeshare deal which would be mutually beneficial. Happy to hear further thoughts if you understand white-label better than me though.

    Cheers,

    Mark

    ReplyDelete
  4. The CEO just resigned! "Saad Hammad has by mutual agreement stepped down from his role as Chief Executive Officer of Flybe with immediate effect". He was a centerpiece of the investment thesis: a cost cutter, hands on, massive incentive package.Why did he leave? Is the situation so dire the turnaround is impossible? Was he fired? I thought his track record was very good? Any explanations/theories?

    ReplyDelete
    Replies
    1. It's pretty clear that Hammad was pushed given that he left with immediate effect with full pay for his notice. This can only really be a boardroom disagreement of some kind but I don't really have any further details. Like you I think he was doing an OK job. He probably ramped up capacity a little bit too fast given the weak macro conditions but a certain amount of this was unforeseeable - also it seems he may have underestimated the time to build load factor on a regional airline competing with road & rail on new routes rather than other airlines but seems hardly disastrous compared to the good progress on other fronts.

      The upcoming results probably won't be great but then they never were going to be. And if they were significantly different to market expectations you would expect them to have said with the announcement of the departure of Hammad.

      Although it adds to the short term uncertainty this has always been a 3-5 year investment for me so I'm happy to wait and see the impact of new management before making my mind up.

      Delete
  5. flybe just three weeks away from posting another loss for the business the two failed directors have resigned and we await ANOTHER five year turnaround for the new CEO. will COW look for another cash injection to keep afloat and keep the company going

    ReplyDelete