tphuang

VX has definitely at least operated all year around in 2016 and 2017. I have the BTS data to show that they were definitely operational. They were normally at once daily during summer, but they were still operational. So please stop repeating that FLL-SFO was a seasonal route. And unless situation somehow gets better for AS by this winter, I don't see that route returning. If B6 adds more frequency on this route, AS is definitely not making a comeback.

As for your second statement, B6 is definitely not at the bottom of the transcon ex-JFK. That would be AS/VX. Considering DL has a larger operation in NYC than B6 and also has a hub at LAX, it's not surprising they have a revenue premium there, but B6 definitely has higher margin on that route. And on JFK-SFO, DL has a very small revenue premium on B6.

On all the ex-BOS transcons, B6 has gone from the bottom to the top once they added a First class cabin. Same with FLL. And same with happen on JFK-SAN/LAS once we have the numbers.

Based on how well FLL-SFO/LAX is performing, I think SEA-FLL will happen. Simply because they need to serve this route at some point to complete their FLL hub and A320 doesn't have the leg to make it without a tech stop in winter time. And the average fares are really good on SEA-FLL/MIA, better than SEA-BOS/JFK

fly18725

I don't have an answer about transcon strategy. My gut is that in a stabilized market there's room for a competitively priced (and lower cost) "standard domestic" non-premium product. From a cost perspective, it definitely makes sense to invest more in the soft product, which Alaska has yet to announce, and maintain a denser, standard hard product that can be deployed throughout the network.

While Mint has addressed a significant weakness at JetBlue, we shouldn't pretend its a panacea. JetBlue's financial performance and outlook are not particularly stellar.

There were periods were VX operated SFO-FLL seasonally, or not at all. But, I don't care about being correct since this point is irrelevant to the larger discussion.

Well, I used the past tense. Still, the data I've seen is quite different than what you're saying. JetBlue's yields have improved, but they don't lead in any competitive transcon markets ex-JFK or BOS.

ashill

Emphasis added. Though that's my instinct too (that premium transcon routes will have to reach a higher price before they stabilize profitably for B6), I'm a bit less convinced running numbers. B6 Mint gets 16 seats in front of the L2 door on an A321; B6 regular gets 52, VX gets 34 (8J), and AA non-transcon gets 38 (16J). So each B6 Mint seat takes (1/16) / (1/52) = 3.25 more floor space than a coach seat (a mix of regular and Even More Legroom), so you'd think that prices would have to stabilize at a point where they get 3.25 times more for Mint than for coach to be profitable (neglecting costs of the soft product, cost savings due to reduced weight, and lost ancillary revenue from coach passengers).

A domestic F seat takes about twice as much space as a coach seat (figure you get 5 rows of 6 Y into the space you'd get 4 rows of 4 F), so the premium has to be a factor of 2 or so to be profitable. That's not actually all that much space for an intermediate price. Now if flat bed business class stabilizes at more like 5-10 times the cost of coach (which I think is where it was for JFK-SFO/LAX before B6), there's definitely space for AS to charge 2-4 times as much as coach (including getting elites to upfare to higher coach fares for an upgrade into that average) and make a profit. But if flat bed business stabilizes at only 4-6 times coach, B6 could still make Mint profitable (and therefore keep AA, DL, and UA J in that price range) and AS really wouldn't have a lot of room to undercut them profitably with their intermediate product.

I started writing this post expecting to come down on the other side, but these total back-of-the-envelope armchair CEO [insert other pejorative term here; I'd be happy to be proven wrong by someone who has actually bothered with the real revenue data!] guestimates make me doubt that the transcon F market really will shake out in a way that supports AS's strategy. It really comes down to how small B6 is willing/forced to keep their margins on Mint.

tphuang

JetBlue has above average margin every quarter despite having very few monopolies. And mint has only increased their margins.

I use data from BTS with my on algorithm filtering out redempted tickets. They provide the average fare of top market share carrier and lowest fare carrier. My basically match their number, so I'm pretty sure of my numbers. Just to give you an example.

BOS-SFO in 2016Q1, pre-mint
AA 1228 327.21
B6 5618 280.38
VX 5503 328.13
UA 10293 401.18
2017 Q2.
DL 2226 307.56
B6 5970 384.69
VX 5855 325.49
UA 15968 357.64
2017 Q3 - after UA added lie flats and DL entered market
DL 4197 280.71
B6 7592 317.88
VX 6628 279.65
UA 14250 350.68

BOS-LAX
2016Q3 (prior to mint entrance)
AA 8704 300.84
DL 4968 287.71
B6 6248 250.34
VX 5772 291.19
UA 5650 280.34

2017Q3
AA 8892 294.38
DL 5913 290.42
B6 6234 307.55
VX 5665 263.37
UA 4975 283.85

beckoa

Didn't read the full context, my apologies. No, this route is not currently operated.

AS and VX have both operated this route previously. Think its been 4-5 years on VX and around a decade on AS. I'd like it brought back, but there are several other flights ex-Alaska that have supported nonstop flights that are no longer operated by AS, DEN for one.

milypan

fly18725

While interesting, I don’t think BTs data on average fare is conclusive about yields. Thanks for the continued JBLU promotion though.

ashill

Thanks for the numbers, though the relevant number is really revenue per plane (ie average fare times number of seats sold on each flight), not revenue per ticket sold (assuming only one aircraft type, eg A321; otherwise you have to adjust for that too). With denser configurations and/or higher load factors, an airline can of course get more revenue with a lower average fare. And cheaper tickets tend to have more ancillary revenue associated with them, I'd bet.

So if I'm reading your numbers right, B6 has gone from 280.38 * 200 = $56,076 per flight to 384.69 * 159 = $61,166 per flight to 317.88 * 159 = $50,543 per flight on BOS-SFO (assuming A321s with 100% load factors for mathematical simplicity), and $50,068 to $48,900 on BOS-LAX by adding Mint. It doesn't look to me like that's a clear win in revenue, certainly not evidence that B6 has settled into sustainable fares that justify widespread deployment of Mint. Now, that's not a long baseline to let Mint fares come into equilibrium, so maybe this doesn't say much about the long term revenue from Mint. I also think they've switched from A320s without Mint to A321s with Mint, so you have to account for both the different capacities and operating costs to do this right. And of course load factors matter.

Yes, though the difference is that ANC-SFO is now hub to hub, which it wasn't before for either VX or AS (unless you can't AS's old tiny focus city in SFO as a hub) and ANC-DEN certainly wasn't.

Yeah, though of course ANC is about as seasonal a market as there is. If ANC is 51st overall from SFO, I'm sure it's considerably higher than that in season. And of course ASVX will now be able to fill the flight with connections at both ends, whereas each only had much feed on one end before.

milypan

ashill

AS has historically used a mixture of 1 and 2. They only sold full F, strategy 1. (No discounted first class sold as such at all until recently.) But elites always had what was in many circumstances effectively discounted F through the time-of-purchase upgrade eligible fares. Essentially AS uses/used the U bucket as the revenue management for your strategy 2, with the discount to get cheaper first class varying by elite status. Then for those elites who don’t buy an instant upgrade fare, there are complimentary upgrades as in strategy 1.

The problem (from AS’s point of view) with this strategy, which you also brought up, is that AS loses much of their ability to price discriminate. The upgradable fares are upgradable whether purchased long in advance (when it’s a buy up) or close-in (when it’s probably cheapest available coach anyway). The big three have been pretty successful in finding ways to price discriminate for the F cabin recently; AS’s upgrade-focused strategy leaves that money on the table unless the upgrades really do drive loyalty (ie customers paying more to fly AS even when they’re in coach than the competition charges for coach).

tphuang

Yes, the numbers I posted are flawed in the sense that it's just the average ticket sold between 2 destinations, so it doesn't account for load factors or what the connection traffic yields are or any other revenue source. Also it discounts award tickets (which BTS stats also discounts). And for legacies especially, the numbers may be inflated by one-stop ticket itineraries between the cities (even if direct flights are offered). But it does provide a good indication imo of the yield in various markets.
That's what I see, yes although I haven't done the associated LF calculation, which is also available. Just because I'm not sure how to account for redemptions and connections and such. You can see a number beside the average ticket. That's 10% of the number of tickets sold between the 2 destinations in one quarter (that was accounted for in my calculation). So for example, UA in the first example had 10293 passengers or about 103k that quarter in total across both directions.

According to B6, CASM went up 6% from A320 to A321 mint config, but then again A320 is in 150 seat config and will be reconfigured to 162 seat going forward, so that difference will change. So if we say a A321 mint replaced A320 on the same route, the # of seats is about constant, the CASM went up about 6 to 15%. And RASM needs to be up more than that to be net beneficial. I do see that's the case in all the example I looked up. It's harder to compare A321 core to mint, because too many more y tickets to sell would crash the average fare.

btw, one thing I would say regarding your example is that mint was reconfigured from previous 190 seat configuration of A321, so they replaced 47 y seats with 16 J seats, which is under 3:1 in ratio.

Interesting things I see regarding to AS from the numbers I looked up.
- fares are really good to HI (especially the monopolies)
- fares are really good in all SEA routes (as expected)
- fares out of SEA are at least similar if not better compared to DL to most destinations that are not DL hubs. Which tells me that DL is at best making a slight profit in SEA.
- fares are bad on the routes they dropped (as expected)
- WN get a premium on SJC/SAN routes to non-PNW destination even though AS use regional planes. That's why I advocate continued effort in SFO expansion.

SF1K

What is making people think this? Any actual word from AS on this? This would totally suck in my strategy of making MVPG. The SFO-FLL cut is already going to hurt on that. I don’t want to go back to UA. ;(

BOB W

According to the reports at the time of the proposed merger, AS and AA were required by government regulators to drawdown their partnership in order for AS to get approval. They are not allowed to offer code shares or FFM on competing routes. If AS flies a route and AA does not (or vice versa), then nothing changed. That is my understanding of how it works now.

I just flew ANC-SEA-DFW (AS) then DFW-TPA on AA, all purchased through AS on one PNR. I got the miles and bonuses for all segments with the AA miles showing up in my partner miles bucket. My routing did not include any competing routes that would have impacted mileage accruals. The ticket was purchased about 5 months out, well after the merger process began.

formeraa

Great Analysis!

rustykettel

My observation was limited to the frequent flyer partnership drawdown, not the codeshares. The FF portion was not due to regulatory action, afaik.