Singapore Airlines

[weijian]

The Spore Gov caps the downside risks for SIA.

Singapore wants to keep borders open even if Covid variant emerges

We have more tools in the toolkit. It's not as if you always only have a hammer. Now I've got you know, a wrench and a screwdriver and all kinds of other things as well

https://www.businesstimes.com.sg/governm...nt-emerges

[weijian]

I thought it would be interesting to look at the hedging gain/losses over the years. Luckily, SIA put it very nicely in their AR.

Year: Hedging gain or loss / Extra fuel spending or savings / Net gain or loss - Hedging duration as per AR

FY18/19:  331.5 / -737 / -405.5 -- 20 quarters
FY17/18:  72.5 / -466.7 / -394.2 -- 20 quarters
FY16/17:  -269 / 56.2 / -212.8 -- 20 quarters
FY15/16:  -926.6 / 1682 / 755.4 -- 8 quarters
FY14/15:  -456.9 / 733.9 / 277 -- 8 quarters
FY13/14:  71.1 / 218 / 289.1 -- 8 quarters
FY12/13:  27.7 / 72.5 / 100.2 -- 18months
FY11/12:  19.9 / -1254 / -1234.1 -- 15months
FY10/11:  -49.9 / -848 / -897.9 -- 15months
FY09/10:  -460 / 1507 / 1047 -- 15months
Net:  -1639.7 // 963.9 // -675.8 

Std dev of hedges = 366mil
Std dev of extra fuel spending or saving = 980mil
Std dev of net gain/loss = 706mil

My thoughts from the above data:

- Hedges are supposed to run in opposite direction to your cost. There has been 7/10 years where hedging and prices are in different directions (ie. there is hedging losses but savings in fuel cost and vice versa)

- In years where fuel prices crashed (2008 and 2015), they always produce big savings but with big hedging losses as well.

- FY11 and FY12 were years that they spent big on fuel but hedging didn't really help. These were coincidentally the good old times when offshore was booming. They lost close to 2bil in those years.

- The standard deviation of hedges (366mil) is ~60% lower than the standard deviation of extra fuel spending/savings (980mil). OPEC's hanky panky and market changes seem to fluctuate more than Mgt's hedging methodology. The standard deviation of net gain/loss was 706mil, which is a 274mil reduction or 25% reduction in fluctuation. I think hedging did its job in the last 10 years.

- Over 10 years, the net loss was ~675mil. SIA spent about 50bil of fuel over that last 10 years. The loss translates to ~1% of their actual fuel cost in this period. I would like to infer that probably no gambling was done by SIA mgt, at least not in large scale.

During the covid era, SIA recognized ~845mil of realized losses coming from fuel hedging ineffectiveness. To recap in a nutshell, fuel hedging ineffectiveness comes in when hedges are not offset by actual usage --> When fuel prices drop, their hedges will lose money but the ops side gain via using cheap fuel (and vice versa). The "ineffectiveness" comes in when fuel prices drop causing their hedges to lose money, but they can't take advantage by using the fuel for ops and that was what happened during covid lockdown.

4 years ago when the fuel hedging ineffectiveness losses (marked to market) were revealed by Mgt, many VBs were actually blasting them for been speculative and reckless. So I thought t would be interesting to look at what has happened since then:

---------------------------------------
Year: Hedging gain or loss / Extra fuel spending or savings / Net gain or loss
FY23/24:  333 / 1114 / 1447       
FY22/23: 638 / -1734 / -1096
FY21/22:  196 / -1001 / -805
FY20/21:  -302 / 387 / 85
FY19/20:  -105 / 380 / 275
Net:  760 // -854 // -94 (FY20-24)

FY18/19:  331.5 / -737 / -405.5 -- 20 quarters
FY17/18:  72.5 / -466.7 / -394.2 -- 20 quarters
FY16/17:  -269 / 56.2 / -212.8 -- 20 quarters
FY15/16:  -926.6 / 1682 / 755.4 -- 8 quarters
FY14/15:  -456.9 / 733.9 / 277 -- 8 quarters
FY13/14:  71.1 / 218 / 289.1 -- 8 quarters
FY12/13:  27.7 / 72.5 / 100.2 -- 18months
FY11/12:  19.9 / -1254 / -1234.1 -- 15months
FY10/11:  -49.9 / -848 / -897.9 -- 15months
FY09/10:  -460 / 1507 / 1047 -- 15months
Net:  -1639.7 // 963.9 // -675.8 (FY10-19)

------------------------------------------------

(1) In the decade of 2010-2019, SIA gained a nett 963mil due to ops but lost 1.639bil from hedges. This resulted in a nett 675mil loss, which wasn't good.

(2) In the next 5 years 2020-2024, SIA lost a nett 854mil from "more expensive" fuel from ops. This was due to the Ukraine-Russia war I suppose. But the rise in oil prices allowed a nett hedging gain of 760mil. Consider the total amt of fuel hedging ineffectiveness of -845mil, the past 5 years of hedging gains made back 90% of the fuel hedging ineffectiveness losses.

(3) Over a 15 year period, considering all the hedging gain/losses, once-in-lifetime fuel hedging ineffectiveness and ops gain/losses --> (760 - 1640) - 845 - (-854+964) = -880 - 845 + 110 = -1615mil. So in essence, we could represent that the hedging lost SIA 1.6bil over 15years (or ~107mil amortized over 15years). Without the one-in-lifetime covid issue, the annual loss would be halved to ~50mil/annum.

(4) On hindsight, one could say that SIA is better off without hedges - and that is mostly true in theory. But in practice, one can easily see that hedging reduces the annual P/L volatility. It is akin to choosing a stable 10% annual gain over a volatile 11% gain. And I guess it makes total sense for Mgt that has little skin in the game to choose a stable gain that is less. What really matters is having more stable results and staying out of trouble OVER becoming a hero in 1 good year to zero in another.

[postage paid]

I thought it would be interesting to look at the hedging gain/losses over the years. Luckily, SIA put it very nicely in their AR.

Year: Hedging gain or loss / Extra fuel spending or savings / Net gain or loss - Hedging duration as per AR

FY18/19:  331.5 / -737 / -405.5 -- 20 quarters
FY17/18:  72.5 / -466.7 / -394.2 -- 20 quarters
FY16/17:  -269 / 56.2 / -212.8 -- 20 quarters
FY15/16:  -926.6 / 1682 / 755.4 -- 8 quarters
FY14/15:  -456.9 / 733.9 / 277 -- 8 quarters
FY13/14:  71.1 / 218 / 289.1 -- 8 quarters
FY12/13:  27.7 / 72.5 / 100.2 -- 18months
FY11/12:  19.9 / -1254 / -1234.1 -- 15months
FY10/11:  -49.9 / -848 / -897.9 -- 15months
FY09/10:  -460 / 1507 / 1047 -- 15months
Net:  -1639.7 // 963.9 // -675.8 

Std dev of hedges = 366mil
Std dev of extra fuel spending or saving = 980mil
Std dev of net gain/loss = 706mil

My thoughts from the above data:

- Hedges are supposed to run in opposite direction to your cost. There has been 7/10 years where hedging and prices are in different directions (ie. there is hedging losses but savings in fuel cost and vice versa)

- In years where fuel prices crashed (2008 and 2015), they always produce big savings but with big hedging losses as well.

- FY11 and FY12 were years that they spent big on fuel but hedging didn't really help. These were coincidentally the good old times when offshore was booming. They lost close to 2bil in those years.

- The standard deviation of hedges (366mil) is ~60% lower than the standard deviation of extra fuel spending/savings (980mil). OPEC's hanky panky and market changes seem to fluctuate more than Mgt's hedging methodology. The standard deviation of net gain/loss was 706mil, which is a 274mil reduction or 25% reduction in fluctuation. I think hedging did its job in the last 10 years.

- Over 10 years, the net loss was ~675mil. SIA spent about 50bil of fuel over that last 10 years. The loss translates to ~1% of their actual fuel cost in this period. I would like to infer that probably no gambling was done by SIA mgt, at least not in large scale.

During the covid era, SIA recognized ~845mil of realized losses coming from fuel hedging ineffectiveness. To recap in a nutshell, fuel hedging ineffectiveness comes in when hedges are not offset by actual usage --> When fuel prices drop, their hedges will lose money but the ops side gain via using cheap fuel (and vice versa). The "ineffectiveness" comes in when fuel prices drop causing their hedges to lose money, but they can't take advantage by using the fuel for ops and that was what happened during covid lockdown.

4 years ago when the fuel hedging ineffectiveness losses (marked to market) were revealed by Mgt, many VBs were actually blasting them for been speculative and reckless. So I thought t would be interesting to look at what has happened since then:

---------------------------------------
Year: Hedging gain or loss / Extra fuel spending or savings / Net gain or loss
FY23/24:  333 / 1114 / 1447       
FY22/23: 638 / -1734 / -1096
FY21/22:  196 / -1001 / -805
FY20/21:  -302 / 387 / 85
FY19/20:  -105 / 380 / 275
Net:  760 // -854 // -94 (FY20-24)

FY18/19:  331.5 / -737 / -405.5 -- 20 quarters
FY17/18:  72.5 / -466.7 / -394.2 -- 20 quarters
FY16/17:  -269 / 56.2 / -212.8 -- 20 quarters
FY15/16:  -926.6 / 1682 / 755.4 -- 8 quarters
FY14/15:  -456.9 / 733.9 / 277 -- 8 quarters
FY13/14:  71.1 / 218 / 289.1 -- 8 quarters
FY12/13:  27.7 / 72.5 / 100.2 -- 18months
FY11/12:  19.9 / -1254 / -1234.1 -- 15months
FY10/11:  -49.9 / -848 / -897.9 -- 15months
FY09/10:  -460 / 1507 / 1047 -- 15months
Net:  -1639.7 // 963.9 // -675.8 (FY10-19)

------------------------------------------------

(1) In the decade of 2010-2019, SIA gained a nett 963mil due to ops but lost 1.639bil from hedges. This resulted in a nett 675mil loss, which wasn't good.

(2) In the next 5 years 2020-2024, SIA lost a nett 854mil from "more expensive" fuel from ops. This was due to the Ukraine-Russia war I suppose. But the rise in oil prices allowed a nett hedging gain of 760mil. Consider the total amt of fuel hedging ineffectiveness of -845mil, the past 5 years of hedging gains made back 90% of the fuel hedging ineffectiveness losses.

(3) Over a 15 year period, considering all the hedging gain/losses, once-in-lifetime fuel hedging ineffectiveness and ops gain/losses --> (760 - 1640) - 845 - (-854+964) = -880 - 845 + 110 = -1615mil. So in essence, we could represent that the hedging lost SIA 1.6bil over 15years (or ~107mil amortized over 15years). Without the one-in-lifetime covid issue, the annual loss would be halved to ~50mil/annum.

(4) On hindsight, one could say that SIA is better off without hedges - and that is mostly true in theory. But in practice, one can easily see that hedging reduces the annual P/L volatility. It is akin to choosing a stable 10% annual gain over a volatile 11% gain. And I guess it makes total sense for Mgt that has little skin in the game to choose a stable gain that is less. What really matters is having more stable results and staying out of trouble OVER becoming a hero in 1 good year to zero in another.

Hi weijian,

Thanks for doing the hard work to deliver such detailed and insightful analysis.

I believe in practice, the benefit of hedging goes even further than P&L volatility reduction down to survival. Hence, its more of an insurance type cost.

Its more of while large, unexpected oil prices falls can boost profits sky high, large unexpected increases in oil prices in the past often coincided with geopolitical factors, such as war that reduce demand at the same time supply costs are high.

This is in addition to the already decreased demand from higher ticket prices.

If prolonged, or if there is a threat such issues are prolonged, it will cause the airline to run into cashflow and insolvency threats which can snowball quite fast as suppliers and financiers restrict credit.

Of course, its not a binary hedge/ dont hedge 100%/0% choice.

The skill of management is to hedge well, based on rigorously tested assumptions and projections. Its quite hard to evaluate their process from the results, but if the insurance cost translates into 1% of fuel costs, which is an even smaller % of total costs, its not bad.

[weijian]

hi postage paid,

I agree one has to look at it like buying insurance for a blow-up. The amortized hedging losses I calculated makes up ~1-2% of fuel costs and so maybe it is worth paying after all. I suspect most of us are probably paying something similar in % for our own personal "blow up" insurance. At the same time, OPMIs getting "passive income" via selling call options are pretty happy too.

As for "down to survival", I did think about it too but I thought it is actually not very applicable for SIA. After all, covid-19 has shown us that Ah Gong is there to backstop SIA. Of course, the prodigal son cannot do anything he wants. But he could probably survive without buying blowout insurance!

[specuvestor]

"The "ineffectiveness" comes in when fuel prices drop causing their hedges to lose money, but they can't take advantage by using the fuel for ops and that was what happened during covid lockdown."

Thanks Weijian for painstakingly compiling the annual figures but in my humble opinion the COVID loss should be excluded from the calculation cause nobody in the right mind would think that operational fuel usage (and matching revenue) would go down to almost zero

Agree that Ah Gong is there to backstop SIA so from a counterparty / on-going entity per se hedging might not matter as much as say commodity businesses but without hedging the management accounting on the price of the tickets and hence volatility of earnings might fluctuate erratically

Actually hedging is not betting; not hedge is betting... that's a common misconception. It's equivalent to saying buying insurance is betting. Hedging helps in managing pricing and costing with more certainty. Another option is not to hedge so your pricing will fluctuate with your cost and assuming demand inelastic. With oil usuallly at cotango means SIA always pay a premium to spot so there is a cost for certainty ie insurance premium.

To make things even more complicated, operational companies if adopt a hedging policy will have to manage the duration of the hedge. If say your company is with limited life, your hedge is quite defined say 3 years or 5 years. But when you hedge fuel for an on-going concern that needs jet fuel every day you have to decide the tenor and that's where the judgement comes. You may have hedged a good price today for next year but does not mean your hedge going forward next year for the following year would be good. So do you hedge 2 years instead of one?

In short do you buy say 3 year term life or a whole life policy?

Those who took the rights issue would have done ok. And Temasek will continue to backstop and depending how long this COVID last, will determine the stake of Temasek

At least it's a fair market driven "nationalisation" unlike in the words of US taking over Fannie Mae or Freddie Mac it is conservatorship

If we look at the Structure (of A-B-S) we will be able to conclude. Obviously the directors know that as well.

But I guess they shouldn't use the word "beneficial"... maybe "give shareholders opportunity to participate pro-rata" as per Temasek...

MCB is not for investors. It’s to signal to investors and business partners Temasek is backstopping