Airbus and Boeing: two opposite bets on how people would fly, and why the smaller plane won

The situation

On 19 December 2000 the shareholders of Airbus approved the launch of a new aircraft, the A3XX, which was renamed the A380 that day. It was to be a double-deck airliner with about 555 seats in a three-class layout, far larger than anything flying. Airbus put the development cost at $10.7bn. Outside analysts thought the real figure would be nearer $14bn. Six customers had signed for 50 firm orders and options on 42 more: Air France, Emirates, the lessor ILFC, Qantas, Singapore Airlines and Virgin Atlantic.

The bet rested on a forecast. Airbus expected airlines to need about 1,200 aircraft with more than 400 seats over the next 15 years. Boeing, looking at the same market, put the figure at roughly a third of that. Analysts quoted by CNN at the time said the A380 would need between 250 and 300 sales to break even.

Three and a half years later, on 26 April 2004, Boeing’s board approved the opposite bet. It launched the 7E7, later called the 787 Dreamliner, with an order from All Nippon Airways (ANA) for 50 aircraft worth about $6bn at list prices. The 7E7 was a twin-engine jet with 200 to 300 seats, a range of 3,500 to 8,500 nautical miles and, Boeing said, fuel burn 20% lower than aircraft of a similar size then in service.

The two programmes were not head-to-head rivals in size. They were rival views of how long-haul flying would be organised over the next 30 years. Airbus thought passengers would keep flowing through a small number of large, congested hub airports, so airlines would need more seats per flight. Boeing thought airlines would use efficient mid-sized twins to open direct routes between pairs of cities and skip the hub. Each company was committing billions of dollars and a decade of engineering effort to its answer, and neither could easily change course once it had started.

Background

Airbus

By 2000 Airbus was the only serious competitor to Boeing in large passenger jets. At the time of the A380 launch it was 80% owned by EADS, the Franco-German-Spanish aerospace group, and 20% by Britain’s BAE Systems. EADS agreed in 2006 to buy BAE’s stake for EUR 2.75bn. The French state held 15% of EADS, and the group was run under a 2000 agreement that gave it two chairmen and two chief executives, one French and one German in each post.

The company’s record with shareholders was mixed. Howard Wheeldon, an analyst at Prudential-Bache, told CNN in 2000 that Airbus had been a great commercial success over its history but had not rewarded its shareholders. He also said government loans would cover about two-thirds of the A380’s cost over the following three years. Boeing’s 747 still held the very large end of the market, and the A380 was meant to end that.

Boeing

Boeing had led the very large end of the market since the 747’s first revenue flight in January 1970. It had answered the A3XX with proposals for a stretched 747, known as the 747X, which analysts estimated would cost about $4bn to develop, a fraction of the A380’s budget. Keith McMullan of Aviation Economics told CNN in 2000 that Boeing would spend a fraction of what Airbus needed to spend. Boeing never built the 747X. It launched a smaller update, the 747-8, in 2005.

Its next big idea was the Sonic Cruiser, a design that would fly close to the speed of sound. Airlines showed more interest in operating costs than in speed, and Boeing dropped the Sonic Cruiser in December 2002 and switched to a 250-seat conventional airliner, known internally for a time as the “Super Efficient Airplane”, which it hoped to launch within 12 months. Boeing’s chief executive Harry Stonecipher announced in December 2003 that the board would offer the plane for sale. Alan Mulally, head of Boeing Commercial Airplanes, led the April 2004 launch.

Both firms make money in the same way. They spend heavily for many years on development, sell aircraft at large discounts to list price, and recover the investment over hundreds of deliveries as unit costs fall along a learning curve. Boeing uses program accounting, which spreads expected costs and revenues across a block of aircraft, so early losses on each jet can be deferred and recovered from later deliveries. That method matters for reading the 787’s results.

The industry

Large commercial aircraft is a duopoly. In 2025 Airbus delivered 793 aircraft and Boeing 600, and they ended the year with backlogs of 8,748 and 6,713 aircraft respectively, according to Forecast International’s tally of company data. Barriers to entry are very high: a new widebody takes about a decade and tens of billions of dollars to develop, certification is slow, and airlines want proven support networks worldwide. No new entrant has broken into the widebody market.

The buyers are concentrated too. A handful of large airlines and lessors place most widebody orders, and they bargain hard by playing the two makers against each other. Engine makers form a third powerful group. In the widebody segment there are usually only one or two engine choices per aircraft, and engine improvements set much of the gain in fuel burn from one generation to the next.

Demand grows with air traffic. Airbus’s 2025 Global Market Forecast expects passenger traffic to grow by 3.6% a year and projects demand for about 43,400 new passenger and freighter aircraft between 2025 and 2044, of which 34,250 single-aisle and 9,170 widebody. About 44% of these deliveries are expected to replace older aircraft. The split between narrow and wide bodies, and within wide bodies between very large and mid-sized jets, is the question the A380 and 787 answered differently.

Forecasts for the very large segment diverged for years and did not converge until Airbus had stopped building the A380 (Exhibit 1).

Exhibit 1: Airbus and Boeing forecasts for very large aircraft

Sources: CNN Money, 19 December 2000; Leeham News, 16 March 2008; Reuters via Malay Mail, 9 June 2017.

Three forces worked against very large aircraft over the period. The first was engines. New engines developed for the 787 offered about 15% better fuel consumption, according to Aviation Week, and Airbus’s long-time sales chief John Leahy later told the magazine that Airbus had been caught out by that improvement. The second was airline economics. Airlines generally find a smaller aircraft easier to fill on a given day, and they can fly it more often, which business travellers value. The third was the arrival of efficient large twins. Aviation Week names the 787, the A350 and the 777-300ER as the aircraft that undercut the A380’s case on cost per seat.

Hub congestion pointed the other way. Airports such as London Heathrow had few spare slots, and Airbus argued that a bigger aircraft was the only way to carry more passengers through them. In January 2016 Airbus said 8% of Heathrow’s 75 million passengers in 2015 had flown on an A380, and that 36 of 47 megacities had regular A380 services. That was the strongest evidence for its forecast, and it explains why some hub airlines liked the aircraft.

The decision

Airbus’s options in 2000

Airbus had three broad choices. It could leave the very large segment to Boeing’s 747 and concentrate on mid-sized twins. It could build a smaller, cheaper improvement on an existing design. Or it could build an all-new double-decker that would beat the 747 on seat-mile costs. It chose the third. The case for it was that a 747 monopoly gave Boeing pricing power and profits that funded its other programmes, and that traffic growth at slot-limited hubs would need bigger aircraft. McMullan put Boeing’s dilemma plainly at the time: it could not afford to give away market segments, but the A380’s failure would be hard for Airbus to absorb.

The case against was concentration. A 555-seat aircraft made sense only for a limited number of airlines with hubs, slots and traffic to fill it. The break-even of 250 to 300 aircraft quoted by analysts in 2000 was a high bar in a segment Boeing thought would need only about 400 aircraft in total over 15 years.

Boeing’s options in 2002 to 2004

Boeing considered building a larger 747, building the Sonic Cruiser, or building a conventional mid-sized twin with a new carbon-fibre airframe. It chose the twin. Mulally said in the launch release that “the size and speed of this order validates our view of the market”. It planned production from 2006, first flight in 2007 and service entry with ANA in 2008.

The same decision contained a second, less visible choice: how to build the plane. Boeing planned to outsource far more of the airframe design and manufacture to partners in Japan, Italy and the United States than on earlier jets, with large sections arriving at Everett almost complete. Supporters argued that this would share the financial risk and cut costs. Critics inside Boeing disagreed. In a February 2001 lecture at Boeing’s leadership centre, senior technical fellow John Hart-Smith argued that Boeing relied too heavily on outsourcing and concluded that “outsourcing all of the value-added work is tantamount to outsourcing all of the profits”. The paper became public in 2003, before the 7E7 launch.

What they did

Building the A380

The A380 first flew in April 2005. In June 2006 Airbus announced a further delay, and EADS shares fell 26% on 14 June. The cause was wiring. Each aircraft had about 300 miles of cable, and Aviation Week later linked the delays to problems with the wiring and with the CATIA design software used across Airbus’s sites. Christian Streiff, who took over Airbus that summer, called the wiring “the only weak link in the production chain”. EADS co-chief executive Noel Forgeard and Airbus chief executive Gustav Humbert both left in early July 2006. Humbert called the delays a major disappointment for customers, shareholders and employees. Louis Gallois became EADS co-chief executive and Streiff, a former deputy chief executive of Saint-Gobain, replaced Humbert.

In October 2006 EADS announced another year’s delay. It said the delays would cut EUR 4.8bn from operating profit over four years, and warned of a EUR 600m charge for losses on some A380 contracts. Delivery targets were cut to 13 aircraft in 2008 and 25 in 2009, from 28 and 40. Emirates, then holding orders for 45 aircraft worth more than $13bn at list prices, said it was reviewing all its options. EADS also set a target of EUR 2bn a year in savings from 2010 under a cost-cutting programme called Power8.

Singapore Airlines received the first A380 on 15 October 2007 and flew the first scheduled service, Singapore to Sydney, on 25 October 2007. Emirates began A380 flights in August 2008 and Qantas in October 2008. Deliveries were slow at first, 12 in 2008 and 10 in 2009, and peaked at 30 a year in 2012 and 2014, short of the 45 a year Airbus had once planned to build.

Emirates became the programme’s anchor. At the 2013 Dubai Airshow, when it had 39 A380s in service and 51 on order, it ordered 50 more. Its president Tim Clark spoke of an optimal fleet of about 120. Other airlines bought in small numbers or not at all. Qantas cut its planned fleet from 20 to 12. Only five aircraft went to China, all to China Southern, and Cathay Pacific never ordered.

Building the 787

The 787 ran late too, for different reasons, which critics later tied to the new outsourced supply chain. The first flight took place at Paine Field, Everett, on 15 December 2009, more than two years behind the 2007 date in the launch plan. ANA took the first delivery on 25 September 2011, more than three years late, having taken extra 767s and 777s to cover the gap. Business Traveller reported a price of about $185m per aircraft. ANA’s senior vice-president Satoru Fujiki said the three-year wait had had an impact on the airline.

In January 2011 Boeing’s commercial aircraft chief Jim Albaugh told students at Seattle University that the outsourcing strategy had backfired, according to a Seattle Times report summarised by World Airline News. Albaugh had not been at Boeing when the decision was made.

On 16 January 2013 the US Federal Aviation Administration (FAA) issued an emergency airworthiness directive grounding US-registered 787s after lithium-ion battery failures on two aircraft released flammable electrolyte, heat and smoke. All 50 Dreamliners then in airline service worldwide were grounded. Flights resumed only after regulators approved a redesigned battery system.

The accounting cost was large. By 2015 Leeham News estimated that Boeing had spent close to $20bn on 787 development and had built up about $32bn of deferred production costs, the excess of early unit costs over the average cost assumed for the first 1,300 aircraft. Whether later deliveries could recover that sum became a standing argument among analysts.

What happened

The A380 winds down

Airbus said the A380 programme broke even for the first time in 2015, when it delivered 27 aircraft, 10 years after first flight. Break-even here meant that the cost of building each aircraft was covered by its price; it did not mean the development cost had been recovered. Aviation Week later put Airbus’s research and development spending on the aircraft at about EUR 25bn, money it said would never be recovered.

New orders dried up. In July 2016 Airbus cut its planned output to 12 a year from 2018. In July 2017 it said it would build eight a year from 2019. In January 2018 Leahy said that without a new Emirates deal Airbus would have to close the line, and Emirates then ordered 36 more. That reprieve lasted a year. On 14 February 2019 Emirates cut its order book from 162 aircraft to 123 and ordered 40 A330-900s and 30 A350-900s instead. Tom Enders, Airbus chief executive, said the company had “no substantial A380 backlog and hence no basis to sustain production”. He called the announcement painful and said 3,000 to 3,500 positions could be affected over three years.

Emirates received the last A380, its 123rd, at Hamburg-Finkenwerder on 16 December 2021, ending a 16-year production run of 251 aircraft. That was at the bottom of the 250 to 300 range analysts had given for break-even in 2000.

Exhibit 2: A380 deliveries in selected years

Sources: Aviation Week; AP via AviationPros, 4 October 2006; Airbus 2015 results and Engineering News, 12 January 2016; Fortune, 15 January 2018; Scramble, December 2021.

The 787 sells, then stumbles, then recovers

The 787 sold in large numbers. Boeing delivered a record 145 in 2018 and 158 in 2019. Then production quality failed. In September 2020 Boeing found manufacturing flaws in new 787s. It cut output for the fourth time in 18 months in December 2020, after delivering none in November. The FAA paused deliveries in May 2021 while it reviewed Boeing’s inspection methods, and later reported a new quality problem on undelivered aircraft. In October 2021 Boeing found titanium parts that had been made incorrectly over three years. In January 2022 it reported $4.5bn of fourth-quarter charges linked to the delays, customer concessions and production costs, and Reuters estimated the total cost of the problems at at least $5bn. In February 2022 the FAA said it would carry out final inspections itself rather than let Boeing certify each new aircraft. Deliveries resumed in August 2022 after the FAA approved Boeing’s inspection and repair plan.

Exhibit 3: Boeing 787 deliveries, 2018 to 2025

Sources: Forecast International, January 2020 and January 2026; Tech Xplore (AFP), January 2022; CNBC, 10 January 2023 and 9 January 2024; AirData News, January 2025.

Demand held up through all of this. In 2025 Boeing delivered 88 787s and took 381 gross orders for the type, ending the year with a backlog of 1,076, according to Forecast International. The SC Daily Gazette, reporting Boeing’s results in January 2026, gave a lower 787 backlog of 1,026. The difference probably reflects how orders at risk of cancellation are counted, but the sources do not explain it. Either way the 787 backlog alone was more than four times the number of A380s ever built.

Airbus, meanwhile, had moved its own long-haul effort to twins. In 2025 it delivered 57 A350s and 36 A330s and held backlogs of 830 and 294. Exhibit 4 compares the two original programmes.

Exhibit 4: The two bets compared

Sources: CNN Money, 2000; Boeing, 26 April 2004; AirlineRatings, October 2017; Aviation Week; Business Traveller, 2011; Leeham News, 2015; Scramble, 2021; Forecast International, January 2026.

Other views

Airbus’s own leaders later gave the bluntest verdicts. Enders said the industry had “misjudged the future of the twins” and underestimated demand for point-to-point flying, according to Aviation Week. Fabrice Bregier, who ran Airbus’s commercial aircraft business, said the A380 had been launched at least ten years too soon. Leahy and another former Airbus executive quoted by the magazine argued that new engines, an A380neo, could have saved the programme. Airbus did not re-engine the aircraft.

Some critics pointed to design choices. The A380’s wing was sized for a stretched version, the A380-900, that was never built, so the base aircraft carried extra weight. Leahy told Aviation Week that you do not build extra weight into an aircraft so that you can stretch it later. Airline planners also disliked inflexibility: an airline could not easily move seats from the upper deck to suit a weaker route.

Supporters of the A380 had evidence on their side too. Emirates’ Tim Clark said in 2017 that the aircraft continued to be hugely popular with customers, and Airbus cited passenger surveys in which 20% of respondents said they would pay a premium to fly on it. At slot-limited airports the aircraft did carry large numbers, as the Heathrow figures show. The difficulty was that too few airlines had both the network and the traffic.

The 787 had its own critics. Hart-Smith’s 2001 warning about outsourcing was widely quoted after the delays. The engineers’ union SPEEA took part in the 2003 debate the paper started, and the Seattle Times reported then that Boeing already outsourced about 64% of the work on its aircraft by value, about half of it engines. Academics have used the programme as a teaching example of supply chain risk in new product development, for example ManMohan Sodhi and Christopher Tang in their 2012 book Managing Supply Chain Risk. The FAA’s decision in 2022 to take back final inspections from Boeing was, in effect, a regulator’s judgement on the company’s quality control.

Boeing’s view was that the 787’s market thesis had been proved. Its executives pointed to the backlog. Boeing Commercial Airplanes chief Stephanie Pope said in January 2026 that the company continued to see strong demand for the 787 family.

Reading it through a framework

Real options

Real options thinking values the right, without the obligation, to expand, delay, switch or abandon an investment as uncertainty clears. It is a useful way to compare the two bets, because both were made under deep uncertainty about network structure.

The A380 contained few options. It was a single very large aircraft, aimed at one network model, with a wing sized for a stretch that only made sense if the model won. If hub traffic grew as Airbus forecast in 2008, with about 1,700 very large aircraft needed over 20 years, the programme would earn a large return. If airlines moved to direct flights, as Boeing’s forecast of 960 implied, there was no smaller version to fall back on. The main remaining options were to slow production or stop. Airbus exercised both: output fell from 30 a year at the 2012 and 2014 peaks to a planned eight a year in 2019, and then the line closed. The decision not to re-engine the aircraft was, in options terms, a decision not to buy a costly option on a small market.

The 787 bet held more options. A 200 to 300 seat twin can fly point-to-point, but it can also feed a hub or replace a larger aircraft on a thinner route, so its value did not depend on one network model winning. Boeing also kept a family option, stretching the design into larger versions. The case shows the limit of the idea, though. Holding the right option does not protect against failure to execute it. The 787’s real options were worth a great deal, but the supply chain and quality failures cost a three-year delay, about $32bn of deferred costs and, in 2020 to 2022, at least $5bn more.

Porter’s five forces

For the A380, the strongest force was buyer power. Emirates took 123 of 251 aircraft, 49% of all production. That gave one airline the ability to decide the programme’s future. When Emirates swapped 39 A380s for 70 smaller Airbus jets in 2019, the line closed, as Enders said in his statement. A supplier whose output depends on one buyer has little pricing power, and Airbus had to support the programme with incentives and goodwill to keep Emirates buying for as long as it did.

The threat of substitutes was the second force. For most airlines, the substitute for one 555-seat aircraft was two mid-sized twins flying more often or direct, which is exactly the market the 787 was built for. As twins became more efficient, by about 15% in fuel use with the new generation of engines, the job the A380 did could increasingly be done another way. The 787’s 2025 backlog of 1,076 aircraft, plus 830 A350s on order, shows how large that substitute market became.

Rivalry and entry barriers explain why both companies made these bets at all. In a duopoly with high entry barriers, each firm’s main fear is leaving a segment to the other. Airbus wanted to break the 747’s hold on the top end. Boeing wanted to protect its mid-sized widebody business from Airbus’s A330 and from the A350 XWB, which EADS launched in December 2006 as the 787 piled up orders.

Since then

The A380 has outlived its production line in service. Emirates, which planned in 2021 to fly the type for about 15 more years, said in November 2025 that it would keep it until 2041 and aimed to raise its active A380 fleet to 110 by the end of 2026 from about 95, bringing aircraft back from storage after maintenance and cabin upgrades. Elsewhere the second-hand market was weak: as early as 2017 the lessor Dr. Peters was considering breaking up its A380s for parts. In the very large segment Boeing also stopped: the last 747, a 747-8 freighter, was delivered to Atlas Air at the end of January 2023, after 1,574 had been built. The 747-8 passenger version sold only 46 aircraft.

The 787 programme has grown again. In January 2026 Boeing’s chief executive Kelly Ortberg said the North Charleston factory had reached eight aircraft a month, up from five a year earlier, and that Boeing would assess readiness for ten a month later in 2026. Boeing broke ground in November 2025 on a second assembly building, part of a $1bn expansion due for completion in 2027. Boeing’s finance chief Jay Malave forecast 90 to 100 787 deliveries in 2026, and the company said rework hours in the 787 factory fell by nearly 30% in 2025.

On 12 June 2025 an Air India 787-8, flight AI171 to London Gatwick, crashed shortly after take-off from Ahmedabad, killing 241 of the 242 people on board and 19 people on the ground. It was the first hull loss of a 787. India’s Aircraft Accident Investigation Bureau said in its preliminary report in July 2025 that both engine fuel control switches had moved from run to cutoff shortly after take-off. The report did not say why and reported no evidence of mechanical failure. As of late July 2026 the final report had not been published; the bureau told India’s Supreme Court that a draft would be filed in a sealed cover in early October 2026. Readers should check for the final findings before drawing any conclusion about the aircraft.

What to take from it

• A product designed for one view of the future carries more risk than one that still works if the future turns out differently, as the A380’s dependence on hub traffic showed.
• Relying on one customer for almost half of a programme’s output gives that customer the power to end it, which is what Emirates’ 2019 decision did.
• Forecasts made by a company that has already committed to a product deserve extra scrutiny, because Airbus’s very large aircraft forecast stayed high for years while orders did not.
• Getting the market right does not remove execution risk, since Boeing’s correct bet on the 787 still brought a three-year delay, a battery grounding and delivery halts costing at least $5bn.
• Outsourcing design and manufacture moves risk to suppliers on paper but not in practice, because the aircraft maker still answers to customers and regulators for every delay and defect.

Questions to think about

1. Analysts in 2000 put the A380’s break-even at 250 to 300 aircraft, and Airbus built 251. Using Exhibit 2, at what point should Airbus have concluded that the programme would not reach the upper end of that range, and what should it have done then?
2. Airbus forecast about 1,700 very large aircraft over 20 years in 2008 and Boeing forecast 960. Which assumptions about airline networks, engines and airports would you need to test to decide between them?
3. Should Airbus have re-engined the A380, given that Emirates was the only airline likely to buy it in volume? Set out the case using real options reasoning.
4. Emirates took 49% of all A380s. How should a manufacturer manage a programme when one customer becomes that important?
5. Boeing’s 787 delivered three years late and built up about $32bn of deferred production costs, yet its backlog exceeded 1,000 aircraft in 2025. Was the programme a success, and by what measure?
6. What does the FAA’s 2022 decision to take back final 787 inspections suggest about the limits of delegated certification in a duopoly industry?

Concepts at work

Real options, Buyer power, Hub-and-spoke versus point-to-point. Explore them in Insights.

Further reading

Oxford University Press covers this story as a teaching case in Cases in Marketing on Oxford Business Trove. The sources below are the public record this page was written from.

Written from the public sources listed below. Figures are as those sources report them.