I think coverage of the A380 often misses the point that at the time, Boeing's 747 simply had no competition, whereas all of Airbus's planes did. So Airbus pretty much had to create the A380 in order to protect its other lines from cross-subsidy by this unopposed cash-cow.
And, also missing from almost all the coverage is the fact that this worked: if the A380 is struggling, the 747 is moribund or dead. Official backlog for the A380 is ~100 planes (actual may be less), the 747's is 21. "the 747 line could be closed in the third quarter of 2019".
And I don't quite get how Boeing somehow played this better, it's not exactly that they decided not to develop a jumbo, they just already had one, whereas Airbus had a bunch of mid-size twins (in fact, the company created and was founded on the mid-size wide-body twin, the A300).
Now you can argue that both planes "were"/might be killed by the shift to smaller twins (and the shift in regulations allowing much longer ocean crossings with just two engines, heck, I recently read that they are now doing the Atlantic in 737s!), but I don't think this is necessarily so, at least not by itself.
Boeing might have been right that there wasn't space in the market for two such planes, but that's a comfortable position if you have the monopoly in this, the high-end segment of the market.
I also am not convinced that the reason given for the A380, constrained slots at major airports serving the major routes, is entirely invalid. With air travel growing the way it does, this problem is not going away, and the shift to more point to point routes may only have been a temporary reprieve.
As an example, Lufthansa recently used jumbo jets on the domestic Frankfurt / Berlin route after their competitor Air Berlin went belly up.
Boeing made a bet on the 787. However, as you note, that's an easier bet to make when you already have the 747. Boeing already had a jumbo. It's easy to make a bet on twin-engines getting certified for longer ocean crossings when you already have the 747. There's a lot of things that Boeing bet on with the 787 and you're totally right that it's way easier to make those bets when you already have the 747. However, they did make those bets and, in hindsight, it looks like they were better moves.
The A380 is really built for hub and spoke systems which is one of the reasons Emirates uses it so much. Emirates almost exclusively goes through Dubai and so they're able to, for example, load up a plane from London to Dubai and then Dubai to Singapore. The airline industry started moving away from that and toward more point-to-point routes in most cases. A 787 lets an airline serve a route that just wouldn't fill a 747 or A380 and provide customers with a better experience. For most airlines, that's a win.
The A380 isn't just a 747 competitor. It's a bet that the 747 didn't go far enough and that what airlines really needed was something huge. Boeing's bet was that a fuel-efficient twin that could serve point-to-point routes (rather than hub-and-spoke) would provide the economics for better, more customer-friendly routes, better utilization due to fewer empty seats, etc. Airbus didn't build a 747 competitor, they wanted to one-up it.
It is true that some airports have constrained slots, but is this mostly a London Heathrow problem? A slot-pair at Heathrow can go for over $50M. By contrast, the DOJ valued 12 slot-pairs at JFK at $44M or $3.7M a piece. American sold 17 at LaGuardia and 52 at Reagan for $381M or $5.5M a piece. That's certainly money, but it seems like the problem at Heathrow is an order of magnitude worse. When you move down from the most congested markets (and DCA kinda counts due to artificial restrictions), slot constraint seems a lot less interesting. Is the A380 a plane to solve the problems of a few airports? Is solving that problem enough?
If you're Emirates, London -> Dubai might be your most important route and the A380 means you can service it for a lot of people. Of course, that's where most of the A380's support is coming from. For airlines that don't operate such a hub-and-spoke model, who use less constrained airports than London Heathrow, etc. it seems that the 787 has been a lot more attractive.
You're right that when they started the A380, it wasn't clear that the market would turn the way it did. Honestly, if Airbus had made a 747 competitor instead of something much larger than a 747, it might have fared better. But the number of routes that can fill an A380 is small, the number of airports where landing slots are unduly expensive is small, and customers like the convenience of point-to-point (and the larger windows, better headroom, better humidity, less noise of a 787).
The A380 isn't a complete bungle, but it was a bet that not only did airlines like the 747, but they wanted something larger. Boeing bet that if they could have something a bit smaller for a lot of the routes the 747 was on, they'd like that. The A380 found a big customer in Emirates, but the 787 has much broader appeal. That doesn't mean the A380 doesn't have utility and is certainly useful for certain routes that are popular and slot-constrained, but the 787 seems to be more useful for a larger number of situations.
Excellent points, and I agree that the switch in the market certainly was a large component, and probably the larger component. I still think that Airbus simply could not not do the A380, just as a defensive move and hedge, and that it may have served a commercial purpose even if it didn't manage to be profitable on its own.
> constrained slots [..] certainly money [..]
Are slots primarily a money problem? My understanding was that the constraint is not so much their expense, but simply non-availability. For example, one of the most valuable assets of Air Berlin apparently is their slots[1]. AFAICT, these were/are not available on the open market, they get doled out by some mechanism.
Non-availability is what makes them expensive. It's kind of two sides of the same coin. If the slots were widely available, they wouldn't be a valuable asset. They are a valuable asset only because they aren't available.
The point of looking at the monetary value is that it's a way to quantify how unavailable they are. Land costs money. Where land is scarce, it becomes expensive. Where land is abundant, it's cheap. If landing slots can be acquired cheaply, it means there is availability.
It looks like EasyJet bought their Berlin Tegel operation for €40M including 25 leased A320s. Most likely the landing permits weren't that valuable. Lufthansa is taking on 81 aircraft for €210.
Let's say that each plane has 2 slot-pairs per day. That's €800k/slot for EasyJet and €1.3M for Lufthansa. But let's say half of the value was the staffing and leases.
Land costs money even when it's not in a popular location. If landing slots command a certain amount of money, that's not a bad thing. It gets bad (bad enough that you want a work around) when they become very expensive.
For example, let's say the A380 takes 2x the passengers of a 787. Let's say that based on passenger load, fuel, profitability of point-to-point over hub-and-spoke, etc. flying the A380 costs you $2M more per year than flying two 787s. Well, it looks like you can get a landing permit for well under $1M so it makes more sense to buy the landing permit off someone else.
Things are generally available, even if scarce, for a price. Airlines are choosing the 787 to optimize for their profit and including the scarcity and cost of landing slots as part of that. At most airports, it seems they're available for reasonable amounts of money.
--
In terms of Airbus not being able to not do the A380, do you mean that they had to create a 747 competitor to be prudent or that they had to build the A380?
You might be right that, based on the information and the market at the time, prudence might have demanded a response to the 747. But did it demand building something bigger? Maybe they could have build something marginally smaller than a 747 and been better off.
Airbus had a dream that a much larger aircraft than a 747 would be a winner. It is for certain routes, but not most. For most routes, companies wanted a smaller 747 and Boeing gave that to them in the 787.
At the time, I remember it looking like Airbus was going to get a win with the A380 and that the 787 would be the mistake. Turns out that larger didn't work out.
> The point of looking at the monetary value is that it's a way to quantify how unavailable they are. Land costs money. Where land is scarce, it becomes expensive. Where land is abundant, it's cheap. If landing slots can be acquired cheaply, it means there is availability.
Sure, but I think you're missing the point about the "non-availability." I don't think the cost shows everything.
When there's lots of slots available then the price will be low. When there's few slots available then the price will be high. When there's _very few_ slots available then prices are harder to compare. There's an upper limit on what a company will pay (a slot can only provide so much profit, after all, even if you expect to own it for many years). Prices will presumably start to depend more on who's bidding for them and what deals can be struck.
Free market economics only works if the market is able to respond (i.e. it's relatively liquid).
Yes, I think Airbus probably had to do a 747 competitor. Did it have to be larger? Dunno. My guess is that it didn’t make much sense to make it the same size or slightly smaller, as shown by the A340, and they already had a 787-size twin, the A330.
this debate is extremely informative; thanks for sharing and engaging so clearly and politely! since you felt a smaller plane would be more desirable for most routes, could you elaborate why you felt airbus was going to win at the time? thanks!
In the early years of the post-communist era state disintegration that is currently going on in Bulgaria, the national airline "Balkan" was sold for a funny sum of $150k [1] and subsequently liquidated supposedly netting a lot of unaccounted money for its valuable slots.
Speculatively, by (i) having a more negative effect on the profitability of their major competitor in a duopoly and (ii) by having indirect positive effects on their sales of other aircraft. As unsuccessful airliner programs go, the A380 wasn't that bad, with over 200 actually delivered. Though it would have looked a lot worse if it wasn't for Emirates
There are also likely technologies developed for the A380 which will have long run benefits when adapted for newer generation versions of other Airbus aircraft.
Not just technologies but also operational gains. The problems with the development of the A380 forced Airbus to synchronise their CAD/CAM software and development practices between their previously semi-autonomous offices in Germany and France. It also led to further organisational rationalisation and the removal of French and German government influence on commercial decisions. It led to the removal of a lot of identical functions on both sides of the border and removal of the rules stating that work had to be split rather than concentrating similar functions in certain countries. That in itself was a big gain for Airbus, probably worth a quarter of the money spent on the development of the A380 if amortised over the length of the project. Once the costs started building up it was easier to convince the various governments to back off a bit.
Methinks the operational gains could be done for 1/100th or less of the $30B+ the A380 cost. Failing at a huge project isn't a way to get governments to back off, it's a way to get them to realize they should stop funding you.
But there is also the human factor to consider, especially in organisations that are fully or partly government owned. They don't have their "own" money (it's taxpayer money), and in some ways they aren't allowed to pursue their own agenda (it's not their money right?), it is political; so rocking the boat by pushing change can be difficult. So for these types of organisation sometimes it takes catastrophic events to foment change.
i) 747 sales were already declining and it required expensive investment to continue.
ii) How would A380 sales help sell other Airbus aircraft? It's such a low volume plane that it offers virtually zero benefits to purchasing other Airbus models.
200 deliveries for a $30B+ development costs is a disaster. Airbus has already admitted the development costs will never be recouped, they'd have to sell over 400 planes to ever have a shot.
And any technology you want to develop for other planes would have been far cheaper to just develop for the other, presumably profitable, planes.
Boeings brilliance was sucking Airbus into this market. Instead of canceling the 747, they kept the 747 production line open with a low cost update (the 747-8) allowed them to crater A380 pricing. That cost Airbus a huge amount of capital, and delayed/hamstrung their efforts to compete with the 777 and 787.
Airbus has lost at least 10x as much on the A380 as Boeing lost on the 747 since 2000.
Nobody is saying Airbus has made a profit on the A380, but absorbing development costs over 250 aircraft delivered leads to much lower losses than absorbing them over 14 aircraft delivered like a certain innovative indirect ancestor of Airbus...
And much as program costs are accounting fiction, Boeing's written down far more than a 10th of the A380 program costs as 747-8 development programme specific losses, never mind the bigger issue of losing a natural monopoly cash cow - even a declining one - and the tag of being the undisputed best at large widebodies which has wider ramifications. Not sure the 747-8 was ever really a significant factor in A380 pricing, because Emirates were even less likely to buy something only fit for freighters as their flagship passenger aircraft than they were to take any notice of what the actual sticker price for the A380 was when negotiating deals.
It's an open question whether the A350 would have done much better if Airbus had doubled down on launching at the same time as the 787, or whether it will do better in the long run for being the more recent aircraft model for its first couple of generations.
Airbus isn't absorbing development costs over 250 aircraft at all, those aircraft are barely going to be profitable on their own construction costs, let alone chip away at the $30B+ developmental costs.
The 747-8 cost only $2.5B to develop. The accounting fiction is Airbus's attempt to minimize how much it actually spent on A380 development. The sole ramification of being the undisputed leader in large wide bodies is apparently limited to massive capital losses.
What boggles me a bit about the A380 vs the 747 is the 747 development was a bit of an accident based on 1960's era irrational exuberance. At the same time they were working on that they were trying to develop an SST. The SST was supposed to be the future of passenger aircraft. It failed and the 747's succeeded. And managed to side line some smaller competitors like the DC10.
But since it went into service no one built a comparable or larger aircraft for 30 years. And if there was a huge untapped market for super large passenger jets, why weren't more 747's in service?
2 points: Airbus had a competitor to the 747. The A340 was close to the 747 with all the benefits of a single airframe supplier.
Second point: Airbus was in a tough spot with the A380. They had to make it huge because Boeing was always threatening a cheaper 747 stretch. I almost feel like Boeing only ever spent money on the post -400 stretch as a "special teams" move to force Airbus into the A380.
Making a money losing investment isn't a "tough spot". You are only damned if you choose to do it. Not only did they know that 747 sales were falling, they knew point to point was becoming more popular.
It was a really easy decision and they blew it. Engineering the world's greatest aircraft is an addictive endeavor, but the numbers have to work and they never did.
In the 1970s the 747 averaged 32 orders per year. In the 1980s it averaged 43 orders per year. It peaked in 1990, taking in 122 orders that year alone, but from 1992-2000 orders dropped to average 27 per year.
Ironically in 1993, the year A380 design work began, the 747 had it's worst sales year ever (till then), with only 2 orders.
The A380 was approved for production in 2000. In 2005, Boeing announced the 747-8 to better compete with the A380.
From 2000-2009, the 747 had 214 total orders, almost all after 747-8 was announced. The A380 had 212. Combined that's 43 orders a year. The market did not grow, despite substantially better, more capable, and more efficient, products.
From 2010-2017, the 747 and A380 have booked 18 orders per year combined. It is by far the smallest jumbo market since it was created. And what point to point advocates predicted.
I think part of the parent's point is that, even in a stagnant market, Airbus basically took half of it from what was previously a monopolist. Those 43 orders per year used to be all for Boeing, and now they only get half. That's a victory of sort, regardless of where the market is going for both companies.
Again, it’s not a victory, it’s been a horrific loss. Airbus disintegrated $30B of their own capital, while Boeing walked away after spending only $2.5B more.
And you see low numbers. All throughout the discussion here, I see mostly references to the 787 and 747. No. It was the 777 all along that constrained the A380. And the 777X was the coup de grace.
This..
Airbus product line (cash cow) is based on midsized fuselage cross section twin. A300 fuselage cross section. This is what funded the A380. 4 engines (A340) dont make economic sense after ETOPS.
Airbus does not have large cross section twin. Hence the A350xwb is critical to their future, not A330neo. Large cross section ETOPS twin 777X is the real 747 replacement. A380 is a halo product and not 747 replacement. The market is too small for A380 production run vs RnD costs.
Now Airbus product line is more secure. A350xwb completes their prouct line-up. Single aisle, medium cross section twin, large cross section twin.
BTW: I forgot that Boeing wasn't the only company to understand how poor the prospects were for the Super Jumbo market. McDonnell Douglas designed a new super jumbo just before Airbus it did, but saw the writing on the wall when they tested customer demand, and canceled it.
What I can't understand though is how it even makes sense at LHR.
A380 requires 'super heavy' spacing between planes AFIAK, as opposed to 'heavy' for 747 and below.
Surely all the gain you get from more passengers is lost as you need significantly more spacing between the plane and the next one? So you might as well just run two 787s instead and extract more point to point incremental revenue?
Please do reply, I've wondered this for a while and haven't been able to have a good answer.
The difference between heavy and super heavy spacing is only 30-60 seconds (depending on airport), which is a lot less than an additional arrival/departure slot (which come in pairs for obvious reasons).
Wasn't another thing that Boeing believed that fuel economy would become more important in the future and thus developed airframes with fuel efficiency in mind?
At the time the Dreamliner was being announced people were initially underwhelmed by the proposal but Boeing persisted and tried to impress upon the aviation industry that they had made the right choice. I don't think at the time everyone thought the Dreamliner was the best answer to the A380.
Turned out it wasn't. The 777 is the real profit engine at Boeing, the Dreamliner may never earn back it's cost thought it will never be as bad an investment as the A380.
Apparently the Dreamliner is doing better than anticipated, for it has begun turning a profit[1]. With the A380 being developed, Boeing couldn't just stand arms crossed. They had to develop a response --and in this case develop new materials to achieve better fuel efficiency --the discoveries thereof can be reused in future frames.
That article is misleading, the Dreamliner is at least a decade away from true profitability, if it ever gets there. When Boeing announces it "turned" profitable, they are saying it now costs less to make each Dreamliner than they are selling for. But that is ignoring over $30B in development costs (plus interest) that it needs to earn back before it can ever generate a true economic profit.
"In the second quarter of 2015, Boeing lost $25 million on each 787 delivered but was planning to break-even per plane before the year end. After that Boeing hopes to build 900 Dreamliners over six years at an average profit of more than $35 million each. But with deferred costs peaking in 2016 at $33 billion, Leeham analyst Bjorn Fehrm considers Boeing can't make an overall profit on the program. Ted Piepenbrock, an academic affiliated with the MIT and the University of Oxford, projects losses decreasing through the first 700 airliners, forecast the cumulative deferred costs to peak beyond $34 billion and its model most favorable to Boeing projects a program loss of $5 billion after delivering 2,000 Dreamliners. Boeing’s original development investment, estimated at least at a further $20 billion, isn't included in these costs.[177]"
Thanks for the insight and clarification. This has to be one of the most difficult businesses to be in, given a poor choice can risk bankruptcy for the developer/manufacturer. If relative success means turning a profit 20 years hence, that's one hell of an industry to be in.
"Boeing made a bet on the 787. However, as you note, that's an easier bet to make when you already have the 747. Boeing already had a jumbo."
It's also easier to make this bet when the passenger version of the 767 is struggling for new orders because of the A330. Boeing had the 747 against nothing, the 777 winning and killing the A340, and the 767 losing to the A330. So Boeing made a better A330. That's why I don't see the 787 as such an insightful bet in terms of capacity (size, length, weight and range): it really is pretty close to the A330. And a modern A330 can pretty much do anything the 787 can do, albeit burning more fuel on the way. The real risk was in the development process and industrialization of the 787 though, which proved risky and costly.
The 787 has a significantly longer range than the 767/330. That fact alone a very large impact:
1/ It is much heavier than those, and costs a lot more in fuel on takeoff and landing, which makes it a bad choice for the short-but-heavy-load routes that the 767 and 330 excel at (HKG-TPE, CTS-HND, etc). So it's a bad replacement and competitor in those markets.
2/ On the other hand, it opens up ultra-long-range point-to-point flights that weren't quite possible before without a jumbo (e.g. UA on SFO-SIN, or routes to secondary Chinese cities like XIY, CTU, CAN, etc.)
Remember that Boeing killed off a trans-sonic plane and decided to build the 787 instead. They weren’t always sure this was the only way to go. I think the economy tanking and fuel costs rising made the decision for them.
But thanks to the 787’s extended range variants, that’s not necessarily going to be the case for long. Qantas starts Perth to London direct in March on the 787.
The other thing that the 787 does is open up availability of more gates at more destinations. If you're an airline and having difficulty getting gates at the traditional international airports, having the 787 is a competitive advantage.
> if the A380 is struggling, the 747 is moribund or dead
Boeing has built and sold over 1,500 747s [1]. It first flew in 1969 and is now being end of lifed. It has been obscenely profitable over its lifetime. The A380 first flew in 2005 [2]. It has been a commercial disaster.
The 747 was a great plane for 1969. Airbus built the finest horse-drawn carriage just as the Model T came out.
> constrained slots at major airports serving the major routes
Between 30 and 40% of passengers departing from JFK or Newark will go on to a connection [3]. Consider what happens when they start flying directly to their destinations.
The 747 was a problem that was disappearing on it's own. When the A380 was approved, the 747 was no longer obscenely profitable, it was already facing lower demand and the requirements for an expensive update.
Airbus crippled itself trying to eliminate it with the A380, they blew a huge amount of capital they'll never get back.
The hub and spoke model is dead and won't be coming back. New slots can be easily created by adding runways or building new airports. Flying point to point isn't just more convenient and faster, it is cheaper and more fuel efficient.
> New slots can be easily created by adding runways or building new airports.
I wouldn't call that "easy". How many new runways are going to be built next year? How many new airports? I'll make a guess that the answers are "none" and "very few or none".
Well, demand is the driver of that. Right now the need isn't that great. If it becomes that great, a far easier solution than the A380 is new runways and new airports.
Demand may be the driver, but the transmission is broken. There's zoning, and environmental regulations, and NIMBY opposition, and on and on.
I mean, how's Heathrow's new runway coming? The demand is clearly there, but the runway is not. It received the support of the government clear back in 2009. After much back and forth, it's ready for a final decision this winter, with the earliest possible year of operation being 2025. And that's with the demand being clearly there.
It hasn't been a 1-for-1 replacement. At least for BA the 747 has been replaced by the A380, 777, and 787 on many routes. All 3 have different capacities to scale with demand; 787 < 777 < A380. And all 3 are more fuel efficient per passenger than the 747.
The first three are from countries where Eads doing well is a political thing: Britain Germany and France own chunks of airbus and have a lot invested in it doing well. The former flag carriers are only nominally independent from the state.
Tell me more about this nominal independence. The biggest state owner of IAG (BA’s corporate parent) is Qatar. Lufthansa is fully privatized. France owns ~15% of Air France-KLM.
For example, an airline might get preferred access to airport berths. Governments, and officials of privatised government entities serve to organise and stabilise this sort of cartel.
> The first three are from countries where Eads doing well is a political thing: Britain Germany and France own chunks of airbus and have a lot invested in it doing well. The former flag carriers are only nominally independent from the state.
OTOH, Lufthansa was the initial customer for 747-8.
So while you correctly pointed out that influence, carriers aren't quite as dependent on their host nations as you assume.
Qantas don’t have any 777s (or any 777 orders). I think that going so hard on the A380 and never getting any 777s was one of their biggest mistakes. Now they’re doing better having cancelled their remaining A380 orders (after they got 12) and have started using the A330 on some routes with smaller demand and have finally got their first 787.
That, and their Emirates alliance have finally turned things around (they weren’t profitable internationally for several years until about two years ago).
What do you mean "started using the A330" ? It has been Qantas's go-to medium-haul airliner more than 10 years, long enough they they've all been refurbished recently with "A380 like" interiors.
Um, the 787 is what kiled the 747. As in Boeing decided, as all companies do, to retire a line in favor of a more modern one. It’s really weird you don’t acknowledge that at all.
> The real question is if cargo operations will continue to drive 747 demand.
The reality is that we're going to have loads of 747-400s that aren't life-expired available, and I imagine many operators will just turn to freighter conversions of them rather than order new 747-8s.
Till now, cargo operations are built on the backbone of older aircrafts that require relatively inexpensive modifications to haul pallets. The economics of the business rarely encourage brand new aircraft purchase with the exception of UPS and FedEx and ad-hoc specialized operators.
In the future, this might not be the case. In the era of fuel efficiency most passenger aircrafts were not designed to haul cargo. In-order to save on weight the main deck on the 777 was made of lightweight materials that is acceptable for passenger traffic. This in-turn makes it very expensive for cargo conversions.
However the efficiency allows long, thin routes. Essentially the 787 and now the newer 777 allow more agile routing over longer distances which opens up airlines to greater routing efficiencies that aren’t as dependent on super-hub airports. There are over 600 787 routes, of which 368 are long haul and 90 are completely new routes. For example, ANA flies Tokyo to San Jose and BA flies London San Jose, those I think the BA version runs the new 777s now.
My point is that passenger capacity isn’t such an important metric when comparing the ‘47 and the ‘87. And the A380 is an airplane built for last century’s model of air travel.
The general point being nobody wants a passenger plane with the A380 or 747's capacity anymore. Small, efficient, long-distance planes like the 787, A350 and C-Series are rendering the larger format obsolete.
.. just to piggy back on this and add a "why": the smaller long range planes are allowing airlines to connect smaller cities with direct flights while still making money. This is easier for customers, and is insanely popular.
An example here is British Airways' service from New Orleans to London. Obviously, some amount of people have always wanted to get from New Orleans to London, but previously they've been flying though Dallas/Atlanta/New York, etc. The 787 allows BA to offer a much better product to those customers, and the size/cost to operate a 787 is what makes the flight commercially viable. When the sky is full of airplanes operating flights like that, there's no one group of people in one place to fill up a jumbo.
But New Orleans/London isn't two smaller cities, it's one smaller city and one major intercontinental hub. Doesn't expecting to have a nonstop flight from London to every city in the world put unsustainable pressure on landing slots in London?
I think routes like San Jose-Manchester is a better example of why A350s and 787s are great. You're also seeing a lot of 787 flights out of Oakland instead of all bay area international travel being from SF.
When I was a kid and lived in Santa Cruz, we would always have to drive to SFO for international flights because SJC's runway was too short for some of the bigger jumbos. Now with 350s and 787s, its not an issue.
Yes. But that's a slightly different problem, and one that could potentially be solved in other ways (an extra runway at LHR, going to a different airport in another European capital if London is unwilling to make the changes needed to keep their airport competitive for connecting passengers transiting through, stuff like that).
Slot utilization at busy airports is one of the better pro-jumbo arguments going around at the moment, though. There's just not enough market pressure to cause airlines to push passengers away from direct flights and towards less popular routing options right now.
Does the comparison not work if the smaller city is in Europe and he major hub is one of the ones in the US? Or even two US cities?
I think perhaps using London as an example is overshadowing the point being made. If you aren't aggregating passengers at hubs which you then use larger planes for the common leg of the journey, you don't need as many larger planes.
Direct flights to more destinations from any hub still means a superlinear increase in the number of slot pairs required. Are you suggesting that that wouldn't also be problematic at hubs in North America?
I'm suggesting, as I took sjm-lbm to be originally, that there are more direct flights that are not using the hubs, therefore less need for larger planes to support aggregated passengers. If flights from New Orleans to New York used to go through Atlanta and the Atlanta to New York flight was full of aggregated smaller flights to Atlanta, that would require a larger plane. If more flights are going direct from New Orleans to New York using smaller planes, that lessens the aggregation in Atlanta, and lessens the need for another 747 at that location.
I was just being snarky about the implication that Chicago is a new opportunity opened up by the 787 when it’s actually one of the largest cities in the world that already gets multiple flights daily from most major world cities.
Off topic: any chance you'd be willing to discuss your transition from algo trading to what you're doing now (per your profile)? As an ex- algo trader and pilot, I'd love to get your perspective.
Posting this from a throwaway because my colleagues read HN, but the email in my profile is valid. Thanks!
The A320s and 777s (and soon A350s) are just more efficient for passenger ETOPS flights. Hell, United will probably even fly 737MAXs on Atlantic routes like they did with 757-ERs.
Flying jumbo jets is bad business these days, but the immediate reason for why these carriers dropped the 747 in 2017 was because starting in 2018 they wouldn't have been allowed to fly due to new FAA rules around fuel tank inerting.
Of course putting "we won't be allowed to fly them since the regulator thinks they're too likely to explode coming next year, but we're still selling flights on them!" in your own press releases is bad for business, so most of the news only mentions other reasons.
Start with this thread[1] from HN a couple of months ago (search for "inert"). This has been a long time coming[2] and is in response to TWA 800 which was a 747 that exploded in 1996 due to lack of fuel tank inerting[3].
The FAA issued a directive in 2008 saying that by December 26, 2017 (not 2018, sorry) [4] passenger planes flying within or into the US had to have fuel tank inerting. Many operators such as Delta and United (can't find a better source for that now than this reddit comment [5]) decided not to retrofit and flew their 747s pretty much right up to the deadline.
Other operators, such as BA (presumably) have already retrofitted their 747s and will continue flying them.
So obviously Delta and United weren't very interested in flying the 747s anyway since they consider it more economical to turn them into beer cans or cargo planes than continue flying them, but the immediate reason they're stopping at this time is because they couldn't legally continue past the deadline.
Which is all I'm pointing out here, that they've known this was coming since 2008 but didn't value the safety of their passengers enough to retrofit the planes, and continued flying them right up to the deadline.
It's worth calling out corporate doublespeak and convenient omissions whenever they occur.
The 747 without fuel tank inerting is still has a very good safety record. It's been over 20 years since one has had a fuel tank explosion. FAA estimates are without inserting a passenger jet risks a fuel tank explosion every 60M hours, or roughly every 30,000 years. Lastly, it's unclear how well the FAA inerting requirements will even work, they only reduce oxygen levels to 12%, while military jets require 9% to eliminate explosion risks.
It's silly to spend a large amount of money to retrofit planes near the end of their useful lives with safety equipment of questionable benefit. Especially when every bit of equipment introduces it's own new risks.
Agreed, overall they're safe. I'm just pointing out that the time of the phase-out in the US is effectively FAA-mandated due to the airlines not wanting to retrofit the fleet, as e.g. BA has apparently done.
This is an interesting bit of trivia, and not something you'll find in airline press releases since for better or worse it's going to make them look worse in the eyes of their customers.
There's ~500 747s[1] in service. So 30k/500 =~ 60yrs, or a 25% chance that one will blow up in the next 15 years. So this seems like a reasonable retrofit given the cost.
I assume that the FAA knows what it's doing and 12% oxygen helps, even if it's not 9%, similarly I assume they've taken into account the risk of retrofitting existing planes, since they started explicitly mandating that in 2008.
I was just passing through Atlanta, and there was a Delta liveried 747 at what I think was Terminal F (we were taxiing to C). I'm guessing it was there for the farewell tour for their employees.
The 747 was already in decline before the A380 came along. It was getting long in the tooth (fuel burn too high) and Boeing needed to invest billions to update it to keep it viable.
In the 1970s the 747 averaged 32 orders per year.
In the 1980s it averaged 43 orders per year.
It peaked in 1990, taking in 122 orders that year alone, but from 1992-2000 orders dropped to average 27 per year.
In 2000, Airbus approved spending €9.5B to build the A380.
In 2005, Boeing announced the 747-8 to better compete with the A380.
From 2000-2009, the 747 had 214 total orders. The A380 had 212. Between them that's 43 orders a year. The market did not grow, despite substantially better, more capable, more efficient, products.
From 2010-2017, the 747 has booked 27 orders, or 3 per year. The A380 has booked 115, or 15 per year. 18 orders per year combined is by far the smallest jumbo market since it started.
There is no doubt that Airbus hurt Boeing's profitability by building the A380. But estimates are that Airbus spent €20B-€30B developing it, most of which it will never recoup. And the bulk of Boeing's profitability is in small to medium sized jets. It mints money with the 777 and 737. It never needed the low volume 747 to "subsidize" them.
Losing tens of billions so your competitor can lose a few billion is not a smart strategy. Good business strategy is making profitable entries into your competitors most profitable niches, not hemorrhaging money in them. Airbus and Boeing were both told that point to point was the future. Boeing heeded that advice (helped by a ton of their own customer research), Airbus mostly ignored it, and the end result is that Airbus crippled themselves while Boeing is as successful as ever.
The argument about airport congestion was always a false one. Not only can and will (edit) new runways and airports be built, but the A380's massive separation requirements gives back much of it's slot efficiency benefits.
One small nitpick - because of the rise in real estate costs and nimbyism, new airports are not as easy a solution as they once were. But I take your point that the 3A80 doesn't really give much better slot efficiency.
>And as far as airport congestion goes, building new airports is far easier than new rail lines.
Might tell Berlin that :-)
It may be true in general if there's undeveloped space within an hour or so drive of a major city center but major airport construction, at least in Western countries, usually hasn't gone smoothly.
Wow, there is so much space around Berlin I don’t even know where to start with this, there’s even an airstrip in town still. I know it’s a park now but seriously of all the major cities I’ve lived in building a new airport should be relatively easy... unless it’s a conscious decision to prefer the train?
Tegel is horrible but it’s also really quick - you can and I have arrived 30mins before a flight and make it just fine.
The Berlin mention is a reference to the Brandenburg airport which has been under construction since 2006 and was scheduled to be completed in 2012. They are now hoping to open in 2020 or 2021.
Yes as I was checking the spelling of Tegel I came across the Brandenburg airport information and linked to it above. It makes Terminal 5 look fast and flawless by comparison!
And how do you expect to move people from that empty space into Berlin? People aren't happy to be flown 50 miles outside the city. When that happens, all those flights coming and going into that enpty space end up addin an extra couple of hours to the travel time.
To make matters worse, that empty space tends to be connected with city centers by.... Train.
Fun fact: there's a single train that serves the already completed train station under the not completed BER airport -- the tunnels were designed with the assumption of some air flow coming from the trains. Now that single empty run ensures the tunnels don't mold.
Haha that's amazing, as a kind of thank you how about this:
I overheard on the Bakerloo line two Transport for London employees talking about safetly and one pipes up and says "of course most of the inside of the Bakerloo line is covered in asbestos, they have to spray the inside of the line with glue every year and check the air quality - it can't ever be removed of course - as soon as you disturbed it you'd be completely fucked".
"It may be true in general if there's undeveloped space within an hour or so drive of a major city center but major airport construction, at least in Western countries, usually hasn't gone smoothly."
I remember DIA (Denver) being built relatively quickly and without trouble - is my memory faulty ?
"As expected, Mayor Wellington Webb announced today that the new Denver International Airport, already six months late and more than $1 billion over budget, will not open on May 15, missing its fourth deadline." [1]
The IATA code for Denver International Airport is DEN. I originally heard it by DIA and was confused many years ago, so I looked it up. Not a huge deal as most airline folks know both versions, but wanted to share the knowledge in case it was useful.
As others have noted, it was hardly flawless though arguably just par for the course troubled except for the baggage system. And that was for an airport basically built in the middle of nowhere.
It's not a bad airport. Its downside is that it was popped down in empty land to the northeast of the city so it's significantly farther from downtown and farther from many of the mountain destinations to the west than the old airport was.
The 747 was killed by loosening the regulations on ETOPS. Going to 4 engines used to be the only way you could run certain routes (particularly transpacific and trans-antarctic routes). Now that big jumbo twins are allowed to run those routes, running a 747 is unnecessarily expensive.
I'm sure this is a VERY complex issue. But two things come to my mind.
1 - Maintenance and tooling for the 747 series has been around for a long time; so moving to 380 would likely be a much bigger deal in that regard than moving to the latest 747 variant.
2 - Fewer airport options for 380 vs 747 means less flexibility of how the fleet is deployed.
Boeing could have chosen to compete with the A380 and built their own Superjumbo, but they chose not to, instead working on the Dreamliner.
The 744, which was the last update to the 747 to see large market penetration, wasn't all that cost effective either:
> with 70 percent of its seats occupied, used more than 95 percent of the fuel needed by a fully occupied 747
Yes, the A380 is more fuel efficient, and cost per seat is much lower, but I suspect with the exception of Emirates, most airlines did not need bigger planes. Other airlines such as BA, Quantas and Singapore bought A380s to replace their oldest 744s, but still have a lot of 744s in service that won't be replaced with A380s.
I expect Qantas have all the A380s they’re going to get (twelve). They originally planned for 20, but their eight remaining orders are postponed indefinitely.
The 787 will be a big part of their strategy from here on in. Their first has finally entered service in the last month or two.
Piling on another point (mdasen is absolutely right about the route-model aspects, and the fact that megahub approaches "lost" to smaller planes flying routes to non-megahubs):
Airbus made one more big, big blunder with the A380: they deliberately and knowingly designed a jumbo that could never have a cargo variant or a cargo conversion. Although the 747's cargo conversion famously is due to Boeing thinking supersonic planes would take over the passenger market in the 70s (and thus the 747 needed a backup role it could take on once it had been displaced), Boeing stuck to the convertibility of the 747 through all those decades, and it's given them business both directly (through selling dedicated 747 freighters) and indirectly (through customers who bought used passenger 747s and wanted to convert, incurring all sorts of work on the airframes).
The A380 can never do this. This is not a "they don't want to" situation. This is a "the way they designed this plane, it can't be done".
First, the A380's middle-deck (the "ceiling" for the lower passenger level/"floor" of the upper level) is a structural component. You can't just yank it out of there and use the whole interior as a vast cargo space. It also cannot have a nose cargo door like a 747; the 747's upper-deck hump with elevated flight deck was created specifically to allow turning the nose cone into a cargo door, while the A380's flight deck is situated much lower to the ground for compatibility of training with the A330/A340.
Second, although the A380 is big, it's either not big enough, or not powerful enough. The increased interior volume over a 747 is large, but the increased capacity to carry weight... is not so large. In fact, an A380's payload capacity is lower than a 747's![1] Which means that if you solve the structural-deck issue, or work around it, you're getting a bunch of space that you can't use without going over the plane's weight limit.
[1] Technically, an A380 can take off and land heavier than a 747 (A380 max takeoff weight is 575,000kg, max landing weight is 394,000kg, compared to 447,700kg/346,091kg for a 747-8). However, an A380 is also heavier to begin with (by about 80,000kg), and must carry more fuel. The maximum payload for an A380 is thus about 84,000kg, compared to 132,630kg in a 747-8 freighter.
Yes, it will never have the capacity of a plane that was designed solely to be a big cargo carrier.
But if you're a civilian cargo-carrying airline not named "Antonov", the 747 is the biggest thing you can get, and you'll never be able to get a cargo A380.
The new twins collect telemetry data from the entire flight. Maintenance is scheduled based on cumulative stresses on the engine and not on miles flown.
They end up going longer between service windows (reduced labor) but it also means they’re more of a known quantity and shouldn’t fall off over the Atlantic. And as we’ve discussed many times in SpaceX threads, having more moving parts may increase the likelihood of a failure instead of reducing it.
So you have bigger engines, more efficient airframes, fewer moving parts, and better maintenance records, why not fly them over the ocean with only two engines?
I don't think the a380 has an analog from Boeing... the 787 is a smaller plane. Boeing gambled on "more choices, more planes" with the 787, whereas AirBus is riding the trend of hauling larger amounts of cattle to established infrastructure.
I don't think the Boeing 747 really created much of a cross-subsidy for their smaller models. Most of the major airlines with 747s had a mix of Boeing and Airbus airplanes.
The 747-400 set the stage for large aircraft that could be operated with only a pilot and co-pilot. That, in turn, paved the way for the 777's fancy new avionics. Those avionics then went down market, including the then-new 737NG.
At least, that is my insider's perspective from the era (95-00).
And, also missing from almost all the coverage is the fact that this worked: if the A380 is struggling, the 747 is moribund or dead. Official backlog for the A380 is ~100 planes (actual may be less), the 747's is 21. "the 747 line could be closed in the third quarter of 2019".
And I don't quite get how Boeing somehow played this better, it's not exactly that they decided not to develop a jumbo, they just already had one, whereas Airbus had a bunch of mid-size twins (in fact, the company created and was founded on the mid-size wide-body twin, the A300).
Now you can argue that both planes "were"/might be killed by the shift to smaller twins (and the shift in regulations allowing much longer ocean crossings with just two engines, heck, I recently read that they are now doing the Atlantic in 737s!), but I don't think this is necessarily so, at least not by itself.
Boeing might have been right that there wasn't space in the market for two such planes, but that's a comfortable position if you have the monopoly in this, the high-end segment of the market.
I also am not convinced that the reason given for the A380, constrained slots at major airports serving the major routes, is entirely invalid. With air travel growing the way it does, this problem is not going away, and the shift to more point to point routes may only have been a temporary reprieve.
As an example, Lufthansa recently used jumbo jets on the domestic Frankfurt / Berlin route after their competitor Air Berlin went belly up.