Sound the trumpets! Bang the gong! Ninety years! In Hong Kong!!

Today British Airways celebrates 90 years serving the Hong Kong route. They started out with a bi-plane to Kai Tak. Today they have an Airbus A350-1000 to Chep Lak Kok. Happy 90^th birthday BA in Hong Kong!

On 24-Mar-1936 an Imperial Airways 10-seater de Havilland 86 bi-plane arrived at Kai Tak Airport.

Using today’s airport code’s, it’s route was something close to PEN-SGN-DAD-HKG. Presumably at some point the plane came from London. Imperial Airways was one of the forerunners of the airline we know today as British Airways (BA).

An 80^th anniversary article in the South China Morning Post, Hong Kong’s local daily, tells us that there were 16 mail bags and one passenger on board the first flight.

Advertisement:

I cannot tell a detailed history of BA in Hong Kong. Standard industry data sources only go back so far.

When the route celebrated it’s 80^th birthday ten years ago, BA widely publicised a launch fare of £175 (£16,040 in 2024 according to Measuring Worth, an online tool that calculates the current value of old money).

My “Mapping the Airways” book of BA route maps by Paul Jarvis shows multis-stop routes that took a week or more.

Things are much easier for me today. These days I whizz across Europe and Asia on a data-collecting Airbus A350 twin-aisle.

Hong Kong is a city I enjoy visiting and know a bit about. It has a fascinating history. And I have personally been flying there for more than 25 years. A big chunk of BA’s operating history.

So in today’s article, in honour of the occasion, I will be ruminating on Hong Kong’s place in the travel world past and present.

I will tell my story flying BA to and from Hong Kong. The highs and, occasionally, the lows. Like with BA itself, my journey to and from Hong Kong has changed. It started with Economy and has gone through First and Business too.

Occasionally a double-decker Airbus A380 was involved. A pint-sized Boeing 787-8 and Airbus A330 were involved too.

For this homage to Asia’s World City, I hope readers will forgive me a little indulgence.

The history of BA’s product & service to Hong Kong during my flying career

During my business travel career BA has operated up to three daily flights between London and Hong Kong. I think it has been at least 15 years since they had triple-daily. Sadly my data sources do not go back that far – readers are welcome to correct me…

When I first started studying the route there were Boeing 747 double-deckers operating almost every flight. Boeing 777s appeared every now and then.

The jumbos were not BA’s most premium heavy. A config with 38 Club World sleeper seats operated some of the flights. The config with 70 Club sleepers was harder to find – these tended to shuttle back and forth on BA’s trans-Atlantic routes.

When BA rolled out their remodelled Club World in around 2007-8, the config switched to 52 Club sleepers. Triple-daily would have meant more than 150 Club World sleeper seats and 42 First suites. A lot of premium capacity.

During this time I was still learning about the industry. I learned a lot of lore from the travellers on FlyerTalk.

Stories of shiny Gold card holders looking forward to middle seats when booking weeks or months ahead because that was all that was left.

Tales of BA overbooking by 100 passengers or more for the Friday night flights back to London and still going out with empty seats.

Rumours of the last-minute full-fare First Class bookers needing two front-cabin seats to Hong Kong tonight and, despite flights being full, somehow getting them.

Posts on Internet bulletin boards like FlyerTalk should be taken with a pinch of salt. But they also contain more than a few grains of truth. Twenty years ago, BA flights to and from Hong Kong really probably were like this.

In the years before COVID Airbus A380 super-jumbos and Boeing 777-300ERs ploughed the route. These two aircraft combined probably matched or close to matched the old 747 jumbo jet services, even at triple-daily, for premium capacity.

I am not quite sure when the service down-gauged from triple- to double-daily, but I can give a precise date for the change to single-daily – 17-Jul-2024. I wrote about it here on Airline Revenue Economics.

The route was down-gauged to a Boeing 787-9 with only 42 Club World seats and a mere eight in First. There are slightly more Club Suites (56) on today’s Airbus A350 but no First Class at all.

Times change. With the rise of middle eastern aviation and the inevitable decline in business traffic post-handover from 1997, Hong Kong does not need as much capacity direct from London as it used to. Cathay have the best choice of times with four or five flights a day.

But although capacity has changed a great deal, one thing that has not changed is the flight time. And these are timed for business travellers.

Flights leave Heathrow in the evening or the night, arriving in Hong Kong in the afternoon or evening local time.

I like the earlier flight. After arriving I pick up the bag and take the Airport Express into town.

Then I am exiting Tsim Sha Tsui station (lift at entrance L4) just in time for hotel check-in and a trip to the Orient Exchange in the Hankow Centre over the road to buy my Hong Kong Dollars.

The flight back to Heathrow is a grueller. With no chance of over-flying Russia at the moment it takes 14 hours via what I call “The Corner” at Mt Bazarduzu in Azerbaijan.

It leaves Hong Kong late and heads through the darkness. Passengers can eat, sleep and watch a movie, and it will still be dark outside.

Supposedly, business travellers like that flight because it arrives into Heathrow at zero-dark-hundred hours. They can impress the boss by appearing early at their desk.

As BA begins rolls out Starlink wi-fi I wonder what the future of flight times like this will be. As far as corporate travel buyers are concerned, is a day of working on a flight better than an OK night’s sleep and then being bleary-eyed all day at the office? Time will tell.

My memorable redemption experiences to and from Hong Kong

My flights to Hong Kong on Cathay Pacific are stories for another day. My trip on Alitalia – an Airbus A330, in Economy, in my late teens, in 2000 – is a distant memory.

Two of my BA trips count among my greatest points redemptions ever. With only one flight a day it is not easy to find redemption seats.

In fact, my own experience suggests that BA make it deliberately harder to secure redemption seats on this route than others.

When traditionally tough points routes like California, South America and even South Africa have special points seats available using the UK’s Amex voucher to open them up, Hong Kong does not.

So back to my two awesome redemptions…

As an organisation buying business travel, the Limited company that I control is a member of BA’s On Business (OB) loyalty programme for small scale enterprises. I can double-dip on points, earning both Avios and OB points.

OB points are much more powerful than Avios – they book into revenue fare classes. My November 2024 holiday to Hong Kong was booked with these. The return was especially good value, at only 29k OB points plus £35.10. Not bad for 6,000 miles in a flying bed.

There is no fuel surcharge ex-Hong Kong due to local regulatory requirements. So I booked two one-ways and avoided BA’s YQ, which I would not have done had I booked a return ex-Heathrow.

The second awesome redemption I believe came from a BA mistake. In summer 2025 BA switched their Revenue Management logic around. They moved from Point of Sale to Point of Commencement.

Reports on FlyerTalk suggested that Avios availability was almost completely dry, even using the Amex voucher which is supposed to open up special seats in Club World and Club Europe business class cabins.

So I took a look at Hong Kong. Availability there was wide open, so I grabbed what I needed and enjoyed my November 2025 holiday. A few days later everything was closed again.

I told this story to some BA RM chaps I met a T2RL Engage in London last year, an industry shindig. They claimed that the logic change did not impact Avios availability with the Amex voucher.

But they also said I was “lucky” to get redemption seats to Hong Kong. I believe that it was BA’s mistake and I was indeed lucky. My best guess is that when the system logic changed over, seats that were meant to be kept open were closed. Meanwhile seats that were meant to be kept closed were opened. Thanks BA!

Mistake aside, the difficulty of finding redemption seats on Hong Kong shows something about the route’s performance.

Loyalty is highly profitable for BA’s owner IAG. So profitable that they make efforts to reward Avios collectors with the Amex voucher that opens Club seats and guaranteed redemptions on every flight.

No or few Club seats opened with the Avios voucher shows that BA expects to sell plenty of seats at a good yield on Hong Kong. Good enough not to earn the profits from loyalty in that market.

My memorable flying experiences

Of my five BA flights from London to Hong Kong, four have left in the early evening – 5.40pm, 5.55pm, 6.25pm and 6.40pm. This time makes for a perfect evening meal service.

On an A380 operated flight in 2019 I was taking my Mum to Hong Kong on Avios and Amex voucher – seats were not so hard to find with large planes on the route.

This flight was truly remarkable. Mum and I were having dinner together and I was in the buddy seat. Due to atmospheric refraction and fortunate timing we watched the Sun set three times and rise twice before twilight finally arrived.

I count that a once-in-a-lifetime experience.

Another flight, this time on the 777, also arrived in Hong Kong just in time for breakfast, UK time. I woke up after a great sleep as the Captain came on to let us know that we were only 40 minutes out.

One of the crew saw me waking and came over to ask if I needed anything.

“Is there still time for breakfast”, I asked.

“There’s always time for breakfast” said the crewman, swiftly followed by delivery of a full English, toast and pastry selection. He must have had it ready and waiting for me!

Great service like this helps keep passengers buying tickets, boosting aggregate demand and so BA’s revenue.

Not all my flights have left early evening. When BA operated double-daily I took a flight leaving Heathrow at 8.55pm. This arrived into Hong Kong much later.

I took it because the better timed flight did not have First Class. Thanks to extensive business travel I had a Gold Guest List “Joker” priority reward voucher, which magically opens seats.

It arrived early enough to see the city in daylight from the plane, but late enough that by the time I left the airport it was dark. Plus by the time I reached Kowloon the Orient Exchange was closed. I would not fly at that time again if I had the choice, even in First.

Flight catering to Hong Kong

The dinner and breakfast service out of Heathrow is quite standard. It has been well reported elsewhere.

In First you get something like this (actually pictured on a Philadelphia flight, but with the same menu as I had on Hong Kong).

The Club meal is also good. Dinner looks like this:

Breakfast looks like this:

This late flight had a particularly unusual second meal service. Rather than breakfast, it was a duck roll and a chicken burger. I had both and they were delicious!

When it comes to catering on the way back to London, the starters are normally better than the main courses ex-Hong Kong. Here are some of the delicious starters in First and Club – duck with hoisin sauce for First and tuna tataki for Club:

And here are the corresponding main courses that are “quite good” rather than great – Hainan chicken for First and steak for Club.

BA generally excels in it’s pudds and this is just as much the case on Hong Kong as elsewhere.

One of the best bits about flying into Hong Kong on any airline, including BA, is the views

For those who have not flown the route, flying into Hong Kong is visually spectacular. You might catch a glimpse of historic Macau (top left in the image below). Depending on which side of the plane you are on there are spectacular views of either the New Territories (bottom right) and Shenzhen, or Hong Kong island and the Kowloon peninsula (bottom left). Or both if you are lucky.

The way back from Hong Kong overnight also has interesting views. The airport is lit up. Look at Hong Kong island carefully and you can see the paths up and down Victoria Peak, including Old Peak Road and the way down from the pavilion to HKU.

Flying over China is also interesting. Every Chinese city seems to have at least one bright red light. These are static – they do not flash.

The larger the city the more there are of them. What these are I have no idea – answers on a post-card…

I normally get to sleep somewhere a bit after Chengdu.

But flying BA to and from Hong Kong has not always been perfect

There was only one time when I was actually annoyed with BA on the Hong Kong route. On my 2023 flight they did not deliver my bag in both directions.

I would not have minded if they had told me on the plane, or put up a sign at the carousel like the BA team at Tokyo Haneda do. I had to stand at the carousel for 45 minutes while everybody else picked up their bags.

If they had told me straight off I could have simply obtained the Property Irregularity Report online and gone into town. The delivery of the late bag was fine. The communication was not. That is modern aviation in a nutshell.

One other time things went wrong I was not annoyed with BA because these things happen and the service recovery was great. My flight was delayed only a few hours away from departure due to lack of a part.

We all needed to wait until the replacement part came from London on the next BA flight. BA did a good job of looking after us – you can read about it here.

A historical perspective on why Hong Kong became an important BA market

To truly understand why Hong Kong became such an important centre of trade and finance warranting hundreds of First, Business and Premium Economy Class seats a day to and from London, a spot of history is required.

Last year I read Patrick Conner’s fascinating book “The Hongs of Canton”. A Hong, also known as a factory, was essentially a national trading post. There were thirteen of them and the first was established around about the 1680s.

Read more

Airline bosses in the Gulf are now taking some of the hardest decisions of their careers.

In the face of conflict with Iran they must first decide whether to up-gauge re-launched operations, maintain a controlled evacuation schedule or make a strategic suspension.

Their decisions now will influence their market share, customer trust, brand perception and even possibly the entire future of mass market Middle Eastern aviation hubs.

Next they must figure out how to handle ticket holding passengers who have either not started their journey or are part of the way through, but not actually in the Gulf.

Unfortunately their rebooking policies reveal a state that is operationally savvy yet strategically underprepared.

And finally they must look after their passengers stranded in transit. This group have perhaps the worst outcome of all. They are away from home in a region at war and little if any practical alternatives to get out.

Monday’s article addressed the first two challenges. In today’s article I will be looking at the third. Specifically, I wanted to find out just how big a problem stranded passengers are and which hubs are most affected.

I will be using schedule data and my own arrivals and departures model to answer four questions.

Q1: How many passengers were arriving in Gulf airports ready to transit when airspace was closed?

Q2. How many passengers were about to leave Gulf airports when airspace was closed?

Q3. What was the net impact of arriving and departing passengers on transit volumes when the crisis began?

Q4. Which airports were worst affected, relatively speaking?

Advertisement:

This article was prepared using data from OAG Schedules Analyser: visit oag.com. Thanks OAG!

Q1: How many passengers were arriving in Gulf airports ready to transit when airspace was closed?

I had to make four sets of assumptions to get the model going.

First up was seat factor for large carriers who use the Gulf airports as their home base. I used reported average seat factors from the incumbent home carrier’s annual report. These were;

Abu Dhabi = 89.9%, based on Etihad’s published Jan-26 numbers

Bahrain = 81.6%, based on Gulf Air’s reported 2024 numbers

Doha = 85.0% based on Qatar Airways’ 2025 annual report

Dubai = 78.9% based on Emirates’ FY24-25 annual report

Kuwait = 74.5% based on news reports about Kuwait Airways from 2017…

Kuwait’s stat is relatively old so please treat it as indicative rather than precise.

Next I needed to make assumptions about seat factor for other airlines. I used the same numbers as for the incumbent at each separate airport.

The third assumption was about the proportion of passengers connecting. Arriving passengers not in transit do not have quite as raw a deal – they already had visas and hotels booked or homes to go to. They are not covered by this model.

I assumed that at Abu Dhabi, Doha and Dubai 75.0% of passengers were in transit if they were on Etihad, Emirates Flydubai, Qatar Airways or (for Doha only) British Airways, who partner with Qatar Airways. For Bahrain and Kuwait I assumed 50.0% for Gulf Air and Kuwait Airways passengers.

For offline transits, possibly on separate tickets, I assumed 10.0% for Abu Dhabi, Doha and Dubai. I assumed 5.0% for Bahrain and Kuwait.

I also assumed that before the strikes on Iran began, the airlines were operating normally with every flight roughly on time.

The chart below shows my estimate of arriving passengers at five Gulf airports on:

1. The day before the crisis began – Friday

2. On the day the crisis began (Saturday) but before the first strikes on Iran

3. The time after the crisis began just after 6am GMT but before airspace was closed just after 9am GMT (the blue box)

4. The time after airspace was closed (the orange box).

The blue line represents the number of passengers set to enter transit at each time, according to my model.

After airspace was closed, the blue line of passengers arriving in the orange box becomes hypothetical. Many of these passengers would have been on flights that diverted to Muscat, Riyadh or elsewhere, or turned around and headed back for home.

The charts show that the timing of the strikes were bad news for all the major hubs.

In Abu Dhabi the first strikes on Iran were just before an arrivals peak. This is unfortunate because many passengers will have started their airport transit just before the strikes began and not all of them will have had time to make their connections.

At least it was the smallest such peak of the day. My model suggests that up to roughly 3,000 passengers an hour will have entered transit at Abu Dhabi around that time.

At Doha the strikes on Iran started roughly in conjunction with their second largest arrivals peak. So Qatar Airways will have suffered the same issues as Etihad but with a higher proportion of their daily passengers.

More than 6,000 passengers an hour could have entered Doha transit at this time.

Dubai suffered roughly the same story, at their largest arrivals peak of the day, My model suggests that 8,000 passengers entered transit from Emirates and Flydubai, among others, in the busy hour just before the crisis began.

The charts also show that the number of affected passengers transiting Bahrain and Kuwait may have been quite small. With this in mind it is easy to see why news reports focus on Abu Dhabi, Doha and Dubai.

For the rest of this article I will show Bahrain and Kuwait in the charts but leave you the reader to form your own commentary on these two airports.

Q2. How many passengers were about to leave Gulf airports when airspace was closed?

The next set of charts show the same five airports as before, but with gross flight departures instead of arrivals.

The charts show that if flights were going out on time, Etihad had almost completed their morning departures bank by the time that airspace was closed. Up to 3,000 transit passengers an hour could have escaped Abu Dhabi in time.

At Dubai on the other hand, the busiest hour was over and around 9,000 transit passengers should have escaped in time.

But the departures peak extends over several hours and it was likely that around 8,000 transit passengers were left behind immediately after UAE airspace was closed.

My model suggests that a similar story may have been played out in Doha.

Q3. What was the net impact of arriving and departing passengers on transit volumes when the crisis began?

The modelling needs to get a little more complex at this stage. Unfortunately not all passengers arrive and then head off on the next flight.

Some linger because their only viable connections are long. Others choose a long connection to achieve a price advantage. So I have made some more assumptions about how long passengers take to connect.

These produce a new set of curves for the number of passengers in transit at any one time. Because of the “lingering” effect I mentioned, these curves are flatter and less peaky than those shown in the previous two questions.

I assumed that:

1. All passengers make their connections within eight hours – with a seven hour window to connect.

This is on the basis that when I was at Qatar Airways we made eight hours the lowest limit for a free Doha stopover. I doubt that many passengers with more than eight hour connections can be truly said to be in transit.

2. I assumed that two sevenths of passengers (28.6%) leave two hours after they arrive and the same proportion leave three hours afterwards

This is also aligned with my experience at Qatar Airways. Many passengers enjoy rapid connections, even if some do not.

3. I assumed that one seventh of passengers (14.3%) leave four hours after arrival

4. I assumed that one fourteenth of passengers leave each hour in the fifth, sixth, seventh and eighth hour after arrival.

These assumptions and the same data as used to answer the previous two questions produced the results shown in the following charts.

At Abu Dhabi somewhere between three and six thousand passengers are likely to have been stranded in transit.

At Doha, my model suggests that around seven and a half thousand transit passengers were stranded, but possibly up to twelve thousand, almost as high as the airport can cope with.

This is because the outbound peak coincided with the airspace closure. If passengers were on flights that got out in time they were lucky.

And the same may be true at Dubai. Between 10,000 and 20,000 could have been stranded in transit.

Q4. Which airports were worst affected, relatively speaking?

The next set of charts show the same results as in the previous section. But they have been redrawn to show passengers in transit as a percentage of the day’s maximum.

Read more

It is far too early to evaluate the impact of war in the Middle East on Gulf aviation. Scenarios ranging from rapid recovery to prolonged disruption and even catastrophe are all plausible at this stage.

If the conflict ends swiftly and Iranian regime change transforms a pariah state into a friend of the west, the region and it’s aviation hubs could boom like never before. Iran Air could become the new darling of global airlines.

On the other hand the region’s airport infrastructure is subject to more physical and operational risk than ever before.

Runways are hard to destroy – craters can be filled in and concreted over. But remaining aircraft, hangars, fuel farms, lighting, baggage and cargo systems, terminal buildings and Air Traffic Control towers are not. There might be a serious incident.

In bad cases, the airlines might never recover.

One thing is for sure. Information appearing from the region should be treated sceptically. These countries do not tolerate criticism of their state apparatus or their leaders. And aviation is very much a part of the state out in the Gulf.

In this article I wanted to evaluate three things. First I will present my thoughts on what the crisis really is and why it matters for Gulf airlines.

Then I will analyse published rebooking policies for the major airlines in the region. I find that they suggest the airlines were unprepared and possibly naïve. I also present my model rebooking conditions that I think airlines should adopt.

In the next article (this Wednesday) I will present my analysis of arrivals and departures in the Gulf around the time the war began. The timing, which led to airspace closure just after major arrivals banks, was unfortunate. In fact, it could hardly have been worse.

This article was written using data from OAG Schedules Analyser: visit oag.com. Thanks OAG!

Advertisement:

What this crisis really represents for aviation in the Middle East

At the time of writing the large airlines are either operating an extremely limited number of flights or not operating at all.

I looked on FlightRadar24 as I wrote this article. The table below summarises what I found. Across the region it shows an up to 94.9% reduction in flights.

The first column shows flights that were scheduled to operate on the day the crisis started. The second column the flights that were operating at lunchtime one week later.

The third column some notes on the nature of the flights, showing that not all Qatar Airways flights were to or from Doha, and that Qatar Airways had cargo flights operating. If aircraft were turning their transponders off they might be missing from this data.

The re-launch of operations is not, in my view, a return to normal.

They look to me first of all like an attempt for the Sheikhs in charge to save face.

Emirates Group including both airline operations and dnata produced $39.6bn annual revenue and $6.4bn profit in their 24-25 financial year. Etihad generated $476 million profit and $6.8bn revenue in 2024 according to their own statements.

Nominal GDP in the United Arab Emirates was $552.3bn (£412.9bn) in 2024, according to the World Bank. Take away Abu Dhabi’s oil and gas, and aviation represents a significant slice of this pie. The Sheikhs will want to keep up the appearance of things going normally.

The re-launched flights also appear partly designed as a controlled evacuation, like lifeboats on a ship. Launching and operating the “boats” involves a degree of risk. So does staying on a “ship” that may or may not be sinking. Individual passengers must judge the risk for themselves.

Flight movements like this present three sets of challenges. First is staff rostering. I would hope that the operating airlines are crewing the flight only with volunteer Flight and Cabin Crew, and providing appropriate risk pay.

If flying is compulsory for Flight and Cabin Crew or risk pay is not thought fair, some staff may not be keen to return to duty once they reach a safe country. Eventually the airlines could have aircraft stranded and crew shortages.

The second challenge is safety. Airborne Early Warning and Control monitoring aircraft appear to be patrolling the skies to ward commercial planes away from risks, according to FlightRadar24. But no avoidance system is perfect and Iran’s behaviour may be unpredictable.

In addition the downing of three F15s by friendly fire in Kuwait opens the possibility of a terrible yet unintentional tragedy.

The third challenge is in customer expectations. Air travel is difficult, unusual and expensive for most consumers. Travellers based outside of the Gulf may trust the airline implicitly and not form a realistic judgement of the risk of travel themselves.

Leaving passengers stranded in transit at the beginning of the conflict is terrible but forgivable. Allowing it to happen again is not.

Middle Eastern airlines are now faced with an unprecedented cocktail of risk.

High fuel prices, loss of revenue, unknown future impact on demand and the risk of aircraft being shot out of the sky are bad enough.

But while these airlines score highly on safety audits, they also prize public image. Some have a legacy of an overbearing management culture. None of the countries impacted have a free press or a culture of disputing the government’s line.

Combining these factors with current events could produce a high stress operating environment prone to error and misjudgement.

They must now decide whether to up-gauge the re-launched operations, maintain a controlled evacuation schedule or make a strategic suspension. Their decisions now will influence their market share, customer trust, brand perception and even possibly the entire future of mass market Middle Eastern aviation hubs.

Airline rebooking policies suggest they are unprepared and possibly naïve

I lived in Doha working for Qatar Airways between 2007 and 2012. From time to time I asked people what they thought about the risk of conflict with Iran. I was lucky to be able to discuss this topic with both senior executives and my peers.

Most people had not given serious conflict a second thought. Almost none were happy to discuss it.

The American, European and Australian members of the leadership team at Qatar Airways I spoke to appeared to have little if any knowledge of contingency plans. One senior executive, a CEO direct report, told me in that he had heard there were plans but did not know what they were.

Understandably, the Gulf airlines have not yet made significant public comment about the crisis. What they have published is their rebooking policies. These are defined by Revenue Management, normally the head of Inventory Management. They are absolutely within the core scope of Airline Revenue Economics.

Reading the rebooking policies, it looks to me like the airline’s senior leadership team are unprepared. They also appear naïve.

There are five causes for concern.

First is an apparent focus for immediate cash preservation rather than long-term passenger retention.

Qatar Airways only permits rebooking to flights within 14 days and only for flights before 15 March. Emirates allows rebooking to 30 April but only for passengers booked until 31 March. Etihad is slightly more generous in one sense but not another – they allow rebooking to 15 May but only for flights on or before 21 March.

Gulf Air seems to have copied Etihad. Flydubai is allowing rebooking within 30 days of the original journey. I cannot even find a rebooking policy on Kuwait Airlines website – if you can and I am missing something obvious, please let me know.

The airlines appear to still be in the seasonal world of Revenue Management, preserving seats for the Easter and Summer peaks. If the crisis continues, those peaks are probably already gone.

Instead the airlines should permit any passenger on any flight to rebook or cancel, up to a reasonable limit.

I would suggest at least 30-June for the time being, and increase this limit day-by-day. Most passengers book at most three or four months ahead.

Knowing that they can rebook or cancel will give passengers confidence to book alternative airlines, while they are still available at reasonable prices. It should also help passengers who have not yet started their flights feel that they can still trust flying with a Middle East airline.

Such generous terms for rebooking and cancellation will come with a significant operational impact for the airlines though. If permitted without limit they will flood the call centre with requests about flights far in the future and the call centre will not be able to cope.

This leads to the second cause for concern with the rebooking policies – they need to reassure travellers that their journey will be addressed, but not necessarily immediately. I would suggest a statement that only journeys up to five to seven days in the future will be handled by the call centre.

Qatar Airways says that passengers should only contact them if flying with 48 hours. That seems sensible for the immediate term but might be too little as time moves on. Starting with 48 hours and moving to five to seven days later could work.

Emirates, Etihad and Gulf Air all remind passengers that they should approach their travel agent if the flight was not booked with Emirates. That also seems sensible.

If I were running the call centre, I might advertise accepting changes for flights five days in the future but informally accept changes from customers with flights up to seven days, provided this is operationally feasible at the time.

The third cause for concern is confusion about bookings made with loyalty points. Reports on FlyerTalk, a bulletin board, suggest that rebooking requests are being rejected if tickets were paid for with points.

Airline loyalty generates large amounts of income for airlines. It is unreasonable for airlines to treat changes for loyalty bookings different to cash bookings at this time.

Next up if flippant coverage of refunds. The policies should make it clear to passengers that if a refund is issued the travellers will lose their rights and consumer protections.

It should also be obvious that a refund may not be sufficient to cover a passenger’s costs of buying a new ticket if they are currently part way through their journey.

Airlines should not necessarily be encouraging passengers to take a refund at this time, especially if the travellers are only partially through their journey.

They have a moral and in some markets a legal duty of care to their passengers. It should be clearer that passengers will understand that they relinquish their consumer protections when requesting a refund.

The fifth and final cause for concern is the lack of clarity regarding the ability of Middle Eastern airlines to re-book on other IATA carriers.

There will be many re-protection agreements in place, specifically designed for cases of extreme disruption. Rebooking policies should make it clearer exactly when and where agents are authorised to rebook to partner airlines.

All the causes for concern suggest that Revenue Management have designed rebooking policies for short-term operational disruption and not achieving long-term strategic objectives.

Good rebooking policies must protect passengers, preserve goodwill and reduce pressure on the call centre servicing bookings. Unfortunately, the current policies appear to achieve little or none of these.

To some extent this scenario is beyond Revenue Management’s pay grade.

Airline CEOs and management boards have had decades to consider the risk of an Iran crisis, which while somewhat unexpected at this exact time was not a bolt from the blue.

The published rebooking policies present cause for concern. They reveal airlines that are operationally savvy yet strategically underprepared. The lessons they learn now and how quickly they adjust to emerging events could impact their ability to operate profitably for years to come.

My model rebooking policy

This proposal is more focused on protecting passengers, preserving goodwill and ensuring the operational stability of the call centre.

Read more

Delft Blue houses are a KLM signature. The Dutch flag carrier has been handing them out to their long-haul Business Class passengers since the 1950s. The iconic ceramics are in the shape of historic Dutch buildings and filled with gin.

The little houses are intended to reflect Dutch history, culture and tradition. I hope to pick up one each way when I fly KLM for the first time in a few weeks.

Which got me thinking. Airline networks are shaped a great deal by the history of their home country. Especially in the old world. So I decided to investigate the influences of history on the Air France-KLM group’s network design.

In particular I wanted to investigate whether economic history or home country culture matters more for airline network design. I analysed the data to answer two questions:

Q1. To what extent are today’s airlines serving their home country’s former colonial possessions?

Q2. Is Air France shaped by glitz and glamour and KLM shaped by finance and trade?

Advertisement:

This article was written using data from OAG Schedules Analyser: visit oag.com. Thanks OAG!

The two countries are quite different, culturally speaking. France is flamboyant and known for it’s love of culture, cuisine and couture. The Netherlands is a bit more direct. Both nations were built on long-distance trade.

Data & methodology

I took a look at both airlines’ networks next week (9-Mar-2026 to 15-Mar-2026) using OAG schedule empire. I classified their non-European destinations as follows:

Former Dutch colonial empire: Aruba, Bonaire, Curacao, Formosa (Taiwan), Indonesia, Sint Maarten, South Africa, Suriname

Sabe & Sint Eustatius are not served by Air France-KLM next week

I did not include Dutch trading posts elsewhere in the Caribbean, Latin America, Africa, the Middle East, South-East Asia & Japan that were not formally part of the Dutch empire. I did not count former Netherlands possessions like New York (New Amsterdam) in North America

French overseas departments: French Guyana, Guadeloupe, Martinique, Reunion

Air France-KLM does not appear to fly to Mayotte in Comoros next week

Former French colonial empire: Algeria, Benin, Cameroon, Chad, Congo, Djibouti, Equatorial Guinea, Gabon, Guinea, Lebanon, Madagascar, Mauritania, Mauritius, Morocco, Senegal, Togo, Tunisia, Vietnam

Burkina Faso, Cambodia, Central African Republic, Cote d’Ivoire, French Polynesia, Guinea-Bissau, Laos, Mali, New Caledonia, Niger, Saint Barthelemy, Saint Martin & Syria are not served by Air France-KLM next week.

I did not count former French possessions in North America (Louisiana and Quebec) as part of the former French empire

Fifth freedom flights carry passengers between two third-party countries as part of a mission starting or finishing in either France or Netherlands. These are counted as half a flight. Seats on these planes are counted as half a seat.

Q1. To what extent are today’s airlines serving their home country’s former colonial possessions?

Here are the results:

Just under a third of Air France flights are scheduled to head off to the former French empire and current overseas departments. Roughly a sixth of KLM flights serve the former Dutch empire.

Interestingly, when capacity is considered the two airlines move in opposite directions.

32.5% of Air France flights serve overseas France and former French colonies, but only 28.8% of seats and 20.3% of Available Seat Kilometres (ASK) serve the same markets.

Meanwhile at KLM, 16.2% of flights are to the former Dutch empire, but 17.5% of seats and 17.1% of ASK serve the same markets.

Q2. Is Air France shaped by glitz and glamour and KLM shaped by finance and trade?

I classified the following destinations as glamorous: Dubai, Buenos Aires, Bridgetown Barbados, Rio de Janeiro, Shanghai, Singapore, New York (Newark and JFK), Las Vegas, Los Angeles, San Diego, San Francisco

I thought that these were the sort of places that influencers might hang out. Do you agree? Answers on a postcard…

The other destinations were all classified as ‘trade’. Note that all the glamorous routes are going to have a large amount of trade traffic associated with them as well.

Now here is the same table as before, but with glamorous and trade destinations split out.

Read more

Last week AirAsia X launched a fifth-freedom route. The nimble Malaysia-based airline will operate a flight from Kuala Lumpur to London Gatwick via the Gulf island nation of Bahrain, pictured.

But at the same time as announcing the flights, they said what at first glance seems to be a lot more too.

Advertisement:

This article was written using data from OAG Schedules Analyser: visit oag.com. Thanks OAG!

The introduction to the press release said…:

Commencing on 26 June 2026, these strategic routes establish Bahrain as AAX’s first global hub, strengthening its role as a key gateway between Southeast Asia, the Middle East and Europe

…and AirAsia’s side said:

Bahrain as our strategic aviation hub allows us to connect Asia with the Middle East and Europe more effectively while creating a scalable platform for future growth

It sounds to me like AirAsia is looking at turning Bahrain into a hub, not just a connecting point for so-called fifth freedom flights which originate or terminate in Malaysia and proceed to a third country via Bahrain.

The press release went on to say:

This partnership [enables] Capital A businesses such as cargo and MRO services to scale alongside AAX. Today, Teleport marks an important milestone as its first flight arrives in Bahrain, following its recent USD50 million capital raise

Capital A is the company which incorporates all of AirAsia’s businesses. Teleport is a logistics provider that calls itself “asset light”. It sounds like AirAsia is making a commitment to Bahrain, but quite how big or small this commitment will be remains to be seen.

The Bahraini side of the press release also says:

The investment … will create direct employment across aviation operations and generate wider opportunities across tourism, logistics, hospitality, and related services [and] support skills development and long-term career pathways for Bahrainis

It seems that Bahrain expects a lot of benefits.

The press release leaves three interesting questions unanswered.

First is whether Bahrain will have it’s own AirAsia airline brand – AirAsia Bahrain – under a Bahraini Air Operator Certificate (AOC).

The launch date is imminent, only four months away. The first flight is operated by AirAsia X under it’s Malaysian AOC.

Perhaps AirAsia is doing a trial run before making the big commitment that basing a fleet of planes and a maintenance operation in Bahrain would involve. On the other hand, maybe they just want to get started quickly and the AOC will follow.

Bahrain’s remarks certainly seem to indicate that they expect further investment and an AOC could be part of that. Bahrain is in Asia if not the far east, so AirAsia Bahrain would be slightly further to the west but still on-brand for the airline group.

On the other hand, Teleport’s investment is likely to be “asset light” and perhaps the airline’s will be too.

If AirAsia Bahrain was set up as it’s own operation, it would be easy for passengers to transit between flights. Transit certainly seems to be part of the plan. The press release also says:

Leveraging Bahrain’s strategic location and aviation structure, the KUL-BAH-LGW service anchors a multi-sector route anchored on transit and partnership opportunities.

Which leads to a second question: if AirAsia does not set up it’s own AOC in Bahrain, will it transfer passengers between alternative fifth freedom flights?

Such an arrangement would certainly be unusual. Transiting from one flight to another under the sixth freedom of the air is practiced by almost every airline.

Transiting from a third freedom flight that terminates in one country onto an onward fifth freedom is rarer but practiced. An example would be taking British Airways flight BA11 from London to Singapore and then transferring onto British Airways flight BA15 which flies from London to Sydney with a stop in Singapore.

I do not see why a third to fifth freedom transfer would not be viable. In the “olden days” of aviation with airlines like BOAC operating multi-stop flights it would have happened all the time. But I have never personally come across arrangements like this in my time in the business so far.

This suggests to me that AirAsia may be considering a Bahrain AOC and an AirAsia Bahrain sub-brand.

The third and final interesting question left unanswered by the press release is where else AirAsia might be looking at flying to and from via Bahrain.

Bahrain has a more limited air travel market than Doha, Abu Dhabi and Bahrain across the water.

In tomorrow’s schedule 46 destinations are served from Bahrain. Only five of these – Qassim in Saudi Arabia, Goa in India, Baku in Azerbaijan, Mangalore in India and Najaf in Iraq – are not served by either Emirates, Etihad or Qatar Airways on the same day.

Meanwhile 135 out of 176 destinations served by the “big three” middle eastern airlines are not available from Bahrain tomorrow. One of these is Kuala Lumpur, which is taken care of by the new entry.

94 of these 135 are routes that might be plausibly served by AirAsia Bahrain to passengers originating in Malaysia.

Read more

The minimum number of Cabin Crew required to operate an aircraft is determined by each airline’s national regulator. While one Cabin Crew member per door is not a hard and fast rule exactly, it is probably a pretty good approximation.

For the maximum number of Cabin Crew, count the number of Cabin Attendant Seats, also known as jump seats.

I wonder if in theory there is nothing to stop an airline boarding additional crew for an extra-attentive service and having them occupy passenger seats? Answers on a post card…

Advertisement:

A solid airline Cabin Crew management process has seven ingredients:

1. Planning – figuring out how many crew are needed at a given point in time

2. Rostering – assigning specific people to specific flights

3. Crew tracking – so the airline can understand where crew are in real time

4. Reporting – notifying operations managers, dispatchers and gate agents when crew arrive ready for duty

5. Logbook – helping crew record all their duty hours

6. Change management – for example, operating a flight when a rostered crew member reports sick

7. Route bidding – helping crew fly where they want to fly, subject to priority rules covering seniority and operational requirements

The systems need to focus on legality – ensuring that Cabin Crew do not exceed their permitted hours and that all flights are appropriately staffed. They also need an element of flexibility to facilitate recovery in times of disruption.

But there is also a cost management element – a crew management system needs to incorporate the cost of overstays, deadheading and crew staying in hotels vs crew returning to their home base into it’s decisions.

I think there is an opportunity for a startup in the crew management software field to integrate Revenue Management (RM) demand forecasts in real-time into the planning and rostering stages.

The idea is that airlines who are able to use RM forecasts can allocate crew further in advance and more efficiently to flights that are lightly or more heavily loaded, and flights that generate higher or lower revenue.

For example, flights that are expected to be lightly loaded may need fewer crew. Flights that are expected to be higher revenue may be an opportunity for airlines to deploy their most highly skilled and Cabin Crew to deliver the best service possible.

It would have to be the type of startup that is looking to create something to sell on rather than transform the industry. The current vendors may not be too far off.

Do current Cabin Crew management systems really need to integrate an RM forecast?

Existing Cabin Crew applications help crew understand real-time passenger information. Think the iPads that you see crew holding on an increasingly large number of airlines.

These show loyalty status (used for meal priority), plus anything the airline knows about a passenger’s preferences or typical service requests.

Such information is made available to Cabin Crew in “real-time”, but are in many ways a digital daughter of the old Passenger Information List that you can see printed out and pinned to wall somewhere in the galley. They are defined at the time of the flight, when things are controlled by airport Departure Control Systems (DCS).

Integrating an RM forecast into crew planning and management software would take place at an earlier stage, when things elsewhere in the airline are handled by the Computer Reservation System (CRS).

It is unlikely that a crew planning and management system would interface with CRS directly to roster according to forecast load and revenue. It would interface directly with RM.

Why might a real-time RM forecast be helpful for Cabin Crew planning?

I think there are seven potential benefits of incorporating RM forecasts in Cabin Crew planning and management. Together they are a pathway to dynamic and variable crew rostering that flex staffing around demand and regulatory requirements, in the same way that RM flexes availability around a capacity constraint.

First-up is forward-looking demand-driven shaping of rosters. Far in advance, RM can flag flights or periods expected to be above normal load. This might be driven by seasonality, for example flights over Christmas being busier, but it is likely that crew planners already account for this.

More exciting is the possibility of RM telling crew planners that certain points in the year or particular flights are expected to be busier or quieter than usual. This would allow the planners to adjust their roster accordingly, potentially saving money.

Read more

In an industry characterised by operational complexity, intensive optimisation and attention to every conceivable detail, the numbers can get very complex very quickly.

But sometimes, simple numbers provide powerful results. Today’s article is about three of these:

1. Why is daily service so much better than five-days-a-week?

2. Why is 3-3-3 abreast (pictured above) a more common aircraft configuration than 2-5-2 abreast?

3. Why is carry-on baggage a bigger issue for continent- and region-sized low-cost carriers than long-range full-service airline brands?

Now read on…

Advertisement:

1. Why is daily service so much better than five-days-a-week?

Imagine an airline flying a route once a week. Say out on Monday and back on Monday. Passengers have one choice – out Mon-back Mon.

Suppose they double the frequency to twice-weekly. Say out Mon and Wed, back Mon and Wed. Now passengers have four choices – out Mon-back Mon, out Mon-back Wed, out Wed-back Mon, out Wed-back Wed.

Three frequencies give nine combinations, four frequencies 16. Five frequencies give 25 choices and seven, or daily frequency, gives 49.

Moving from one weekly frequency to two doubles the cost commitment to a route but quadruples customer choice. A fine return in many cases.

In the same way, increasing frequency from five-weekly to daily doubles consumer choice for a 40% cost penalty. This is why daily service is so much better than five-days a week.

Even if flights are not so busy on two or three of those days, airlines might be able to charge more on the flights that are, covering the cost and maybe even earning a surplus.

Eventually this strategy runs out of puff in all but a few longhaul markets. But there are plenty of short-range services where high frequency is prized. on short business trips for example people will value the chance to come back on the same day they arrive.

Double-daily service gives 196 combinations, four-times daily. Triple-daily gives 441 and quadrule-daily a whopping 784.

The marginal extra flight a day runs into returns that are increasing but diminishing though. Double-daily services offer four-times the choice of daily. But triple-daily offer 2.25-times as much choice as double-daily. Quadruple-daily offers 1.78-times the choice of triple-daily.

On the day this article is published BA is scheduled to fly nine times a day from London Heathrow to New York. Add three codeshares with American Airlines and the frequency is an enormous twelve flights a day, counting both JFK and Newark airports as New York.

Yankee competitor United Airlines is scheduled to operate six flights on the route and groovy-funky Virgin Atlantic offers four flights. That is a considerable choice.

Emirates out of Dubai to London is another example of a high-frequency long-ish range route.

On London to Los Angeles, another prime route, the BA-AA pair offers four flights a day. When I flew to Dallas there was a choice of four AA and one BA flight.

On some routes, choice looks high at first glance but in practice is a bit more limited.

BA offer an early evening and late night departure from London to Singapore, for instance. But both of the return flights are late night departures arriving at zero dark hundred hours.

This type of overnight scheduling was considered desirable in a business travel heavy world. Long-range daylight flights are sometimes known as the Chairman’s Special because only the Chairman dare take them.

As the airline industry moves to on-board connectivity I think we will eventually, over decades, start to see more daylight services. Many workers and their bosses will consider a short day of good work better than a long day of poor work following a red-eye.

2. Why is 3-3-3 abreast a more common aircraft configuration than 2-5-2 abreast?

Read more

Today I am interested in the relationship between an airline’s higher capacity routes and their lower capacity routes.

Advertisement:

I took a week of schedule data for ten large United States-based airlines and calculated the number of seats on each route. These are bi-directional, so for example Chicago to Los Angeles and Los Angeles to Chicago are counted together.

This article was prepared using data from OAG Schedules Analyser: visit oag.com. Thanks OAG!

The ten airlines, arranged by IATA code, are:

AA – American Airlines | AS – Alaska Airlines | B6 – jetBlue | DL – Delta | G4 – Allegiant Air | MX – Breeze | NK – Spirit | SY – Sun Country Airlines | UA – United Airlines | WN – Southwest Airlines

The date I chose was 4-May-26 to 10-May-26. This is not a public holiday and should be roughly “normal” times for business travel. Airlines are unlikely to have a special focus on sunny leisure-focused routes.

First of all, let’s take a look at US domestic routes. There are a whopping 20.6 million seats a week across 4,234 routes. American Airlines has the broadest network – 826 routes and the most capacity too – just under 4.8 million seats..

The chart below shows the rank of each airline’s domestic routes on the horizontal axis, the largest at the left. The vertical axis shows the number of seats on each route in a logarithmic scale as a proportion of the number of seats on the largest route.

The purple line shows “Zipf’s Law” – the second highest ranked is 50% of the first, the third highest 33%, the ninth 11% and so on.

While I appreciate that at first glance this charts seems a bit abstract, I think it shows nicely the following:

1. Southwest, America, Delta and United are much “flatter” curves than the others (except Allegiant Air).

It is no coincidence that these are the largest airlines.

I think this is an interesting way of showing how the depth of an airline’s schedule (frequency and connectivity) trumps breadth.

They have chosen to have high frequency and high capacity on even their relatively less important routes.

This is partly because after a certain point there are not many domestic destinations left to fly to. But it is mainly because consumer choice with flights at lots of different times of the day adds a lot of value that the airlines can monetise.

2. All airlines are above the Zipf’s Law line. I would expect airlines below the Zipf’s Law line to focus exclusively on a few routes to the detriment of others, which is clearly not something we see airlines doing because passengers value connectivity and optionality.

3. Sun Country Airlines is the closest to the Zipf’s Law line. A high proportion of this airline’s routes are seasonal. They choose to focus on a core network of major cities from their Minneapolis hub.

4. Almost all the airlines experience a “cliff edge” as their least well-served routes might only be daily or less, so total capacity is a fraction of the more frequently served routes.

Next up, let’s examine international routes to and from the US, operated by single-aisle aircraft. I excluded sixth freedom routes that serve two third-party countries as part of a journey originating or terminating in the United States.

The scale of single-aisle international operations is small compared with domestic – only 1.99 million seats a week (domestic 20.6 million) across 720 routes (domestic 4,234).

Read more

Poland is slap bang in the middle of continental Europe. It’s capital Warsaw lies 1,438 miles from Ufa in the Ural mountains, only a few miles from the eastern edge of continental Europe. Warsaw also lies 1,249 miles from Cork in Ireland, close to Queenstown, Titanic’s last port of call, and a few miles from the great Atlantic ocean.

A circle with radius 1,438 miles or 1,249 miles around Warsaw includes almost everything European.

So it is no surprise that Poland is well served by European airlines.

But recently I heard an interesting claim. A colleague said that because the Polish air travel market is not concentrated, LOT is not really the national airline of Poland.

They are of course the Polish flag carrier. But are they the main Polish airline? I wanted to find out. As part of this analysis I also wanted to figure out whether or not LOT Polish Airlines might ever become part of Greater Lufthansa. Spoiler: I doubt they will.

So I took a week of schedule data from OAG for Mon-10-Nov-2025 to Sun-16-Nov-2025. This includes the 11-Nov Independence Day holiday.

Is the claim true? Read on…

This article was written using data from OAG Schedules Analyser: visit oag.com. Thanks OAG!

Poland’s domestic market

Unsurprisingly, the domestic market is dominated by LOT. They operate:

1. 378 domestic flights & 35,851 flights a week

2. 84 (22.2%) of these flights are between LOT’s two hubs at Warsaw & Krakow

3. 290 of the flights are to or from Warsaw.

There are also six Ryanair Polish domestics a week – three from Krawkow to Gdansk and three back.

It is certainly not the case that any non-Polish airline is anywhere close to being a significant airline on Polish domestic.

But then Poland is quite a small place, geographically speaking. It is only 388 miles from Szczecin in the north west to Rzeszow in the south east. I have not taken the trains in Poland but a browse of Reddit suggests they are not so bad.

The Warsaw & Krakow international markets

Warsaw and Krakow are the two international hubs of LOT Polish Airlines.

Together the two airports have 4,300 (70.4%) of Poland’s 6,109 weekly international flights. Their share of the 1.09 million seats a week to and from Poland is 68.4%. So the average plane flying internationally to or from Warsaw and Krakow is smaller than the average plane flying elsewhere.

LOT has 1,513 (35.2%) of these flights and 204,679 (27.6%) of the seats. Ryanair and Wizz together have slightly more flights than LOT (1,598, 5.6% more) but a higher share of the seats (342,124, 67.2% more).

The top three airlines serve a diverse range of routes with little overlap. Ryanair flies to only 17 of LOT’s 75 international destinations. Only five of LOT’s top 20 routes by frequency* are in the Ryanair network at all. This is counting both Ryanair’s core Ireland Air Operator Certificate (AOC) and Ryanair’s UK AOC.

I am counting the various London destinations – Stansted, Heathrow, Luton, Gatwick as different routes because although they all serve central London, they have quite different catchment areas overall.

Wizz competes with LOT a bit more intensively. 30 of the pink-un’s routes are shared with LOT and they fly on 13 of LOT’s top 20. This is counting all the various Wizz AOCs together.

* LOT operate twin-aisle planes on long-range routes New York JFK & Newark, Chicago, Toronto, Seoul, Los Angeles, Delhi, Miami and Tokyo, which are not fairly comparable with Ryanair or Wizz.

The Polish airline’s longest single-aisle routes sound exotic to western European ears but are considerably (roughly a thousand miles) closer to Poland than they are to London, Paris or Amsterdam – Dubai, Tashkent, Riyadh and Nur-Sultan.

Lufthansa comes next. They have 352 international flights a week (23.3% of LOT’s) and 51,596 seats (25.2%).

Unlike Ryanair and Wizz, which fly to 66 and 54 destinations from Warsaw and Krakow respectively, Lufty is more concentrated.

Each airline in the Greater Lufty group flies to their own hub. Munich and Frankfurt for mainline Lufty (188 flights a week), Vienna for Austrian (72 flights), Zurich for Swiss (54), Brussels for Brussels Airlines (26) and Dusseldorf for Eurowings (12).

Air France-KLM comes next with two routes, Amsterdam and Parid. Then easyJet, with seven routes in the UK and four in the Schengen Area (Paris, Basle, Geneva and Amsterdam).

LOT is clearly the major airline of Warsaw and Krakow. Ryanair and Wizz are also prominent.

The international markets from second-tier cities

Read more

When it comes to aviation, macroeconomics matters.

A prosperous home market allows Qatar Airways is able to offer one of the world’s most popular business class seats.

In Europe, Ryanair’s ebullient CEO Michael O’Leary is often on the record speaking about how important economic growth is for his airline.

The money supply matters too. When travellers found their trips cancelled due to COVID, airlines handed out vouchers. Consumers used these vouchers and their cash income to compete for seats a few years later.

Unsurprisingly with more money chasing the same seats, fares jumped by roughly the value of the vouchers.

But one thing I have not yet looked at is how macroeconomic variables have different impacts on full-service vs low-cost carriers. A recent paper from Jonathan Felix and Eduardo Saucedo from EGADE Business School in Mexico* covers exactly this point.

* Effects of macroeconomic variables on domestic passenger airlines: Evidence from low-cost and full-service carriers in Mexico

The paper’s detailed econometric analysis offers a neat and tidy indication that low-cost airlines are more sensitive to macroeconomic changes than full-service carriers. I have read it and explain it’s key findings here so you do not have to work through all the maths.

First of all I take a quick look at how Mexico’s domestic air travel market is structured. I identify the key airlines, routes and markets.

Next I summarise how the Felix & Saucedo paper considers each of the main macroeconomic variables that influence airlines, including a few points in their model that I disagree with.

Finally I examine the two economist’s results and offer some extensions to their policy recommendations.

The first section of this article was written using data from OAG Schedules Analyser: visit oag.com. Thanks OAG!

I have not used correct Spanish accents on the names of the Mexican airlines and cities mentioned in this article. Being British, accents are not something I normally use and are not available on my keyboard. I decided it would be easier to use no accents at all rather than probably get them wrong – sorry.

Air travel in Mexico: a brief market portrait

Read more

The government of Somalia has apparently purchased two Airbus A320 single-aisle planes. They will be put to use by the end of September as part of a revival for national carrier Somali Airlines, which suspended operations in 1991.

Somalia needs these planes. The country of 19.3 million people has only 74 domestic and 94 international departing services a week. Of these, 63 domestic and 75 international departures are to or from Mogadishu, the capital.

The country is nearly 1,000 miles long, with the longest coastline in Africa. Getting from place to place by land or sea takes time and may be dangerous.

How should Somalia use their new planes? I wanted to answer three questions:

Q1. Should Somali Airlines prioritise international or domestic services?

Q2. Is joining an alliance a good idea for Somali Airlines?

Q3. What might the long-term future of Somali Airlines look like?

Let’s take a look…

This article was written using data from OAG Schedules Analyser: visit oag.com. Thanks OAG!

Q1. Should Somali Airlines prioritise international or domestic services?

The maps linked below shows the domestic and international departures from Somali airports during the week of 8-Sep-2025 to 14-Sep-2025. There are 5,598 domestic and 10,104 international seats a week. It also shows a selection of nine airports which do not appear to have any flights filed in OAG.

Click here for the map of 74 domestic flights

Click here for the map of 94 international flights

Click here for the map of airports that I do not see any data for

All the maps were generated using the excellent Great Circle Mapper: visit gcmap.com.

The table below shows how these are spread over the week.

Read more

The three global airline alliances are huge. All together they serve 208 countries with 308,199 flights and 49.4 million seats a week. 17,956 origin-destination pairs are served, counting a return trip as two routes.

And that is only counting the 58 carriers that are full members of one of the three alliances – oneworld, Skyteam and Star Alliance. Add the 43 affiliate members and the scale would be even more enormous.

Airline alliances span the globe. Each of their members commands significant attention. But there is surprisingly little analysis of how the alliances are structured, how they compare with each other and how they fit into the global market as a whole.

In this article I wanted to investigate the three alliances from a global macroeconomic perspective. I first outline why airlines participate in alliances and then answer six macroeconomic questions.

Q1. How do the alliances compare with one another?

Q2. To what extent do the alliances compete on the same routes?

Q3. How focused in each alliance on longhaul premium travel?

Q4. Does any alliance offer more comprehensive access to emerging markets?

Q5. Is one alliance more exposed to inflation risk?

Q6. How exposed is each alliance to geopolitical risk?

For one airline based in one country and flying to their established markets these questions are all important. But at the alliance level where groups of airlines are working together they become even more relevant.

This article was written with data from OAG Schedules Analyser: visit oag.com. Thanks OAG!

Why do airlines participate in alliances?

When airlines want to expand the reach of their networks there are two options.

They can launch new flights, which offers a direct stake in a route’s success but is costly. Alternatively airlines can agree to support each other by offering their seats for other airlines to sell and market on their behalf.

This type of co-operation can take four forms:

1. Under an interline agreement, two airlines agree to sell services on each other, either in accordance with standard IATA proration procedures or under a Special Prorate Agreement.

2. With a Codeshare Agreement, airlines have all the benefits of interline plus the ability to sell seats on another airline under their own flight number, which is supposed to make the option more palatable to passengers.

3. Some airlines have Joint Ventures in place, where they sell seats on each other under each other’s codes and divvy up all the revenue.

British Airways have a Joint Venture with Qatar Airways for heading east from London and a Joint Venture with American Airlines for flights heading west.

4. Entering an alliance allows airlines to sell and market flights together, and also to offer loyalty benefits to programme members from all of their partners,

The idea is that passengers may, for example, be more likely to fly British Airways and indeed pay a premium to do so if they know that their Cathay Pacific diamond card will get them whizzing through priority check-in, into the lounge and then into one of the best seats onboard.

The opportunity to earn miles and points in a passenger’s preferred loyalty programme so all flights count for an aspirational redemption will be a critical part of choosing an airline for many travellers.

Q1. How do the alliances compare with one another?

The chart below shows that Star Alliance is the biggest alliance.

Star Alliance has a 40.9% share of the flights, a 42.6% share of the seats and a 43.2% share of the ASKs. 58 airlines are full members of one of the three alliances. 25 of these (43.1%) are Star Alliance members.

Note: In this analysis I am concentrating on full alliance members only. All three alliances have associate members too. Associate members are only included if they operate under the same IATA code as the full member. That would not change Star Allliance’s position though. They have 32 associates while oneworld has ten and Skyteam has two.

The next chart shows that across all three alliances domestic flights represent a significant proportion of the total – from 57.9% of flights on Star Alliance to 69.5% on oneworld.

A large domestic focus makes sense. Airlines want to join alliances to feed all their flights. Especially flights with seats that would otherwise go unsold or be subject to competitive pressures on yield.

Since domestic flights are short they represent a low share of capacity in terms of Available Seat Kilometres (ASK) though. Domestics are only 26.1% of capacity on Star Alliance, 30.5% on oneworld and 35.4% on Skyteam.

Read more

The way that airlines numbers their flights suggests something about how their operations are organised.

On British Airways, four-digit numbers starting with a number 2 indicate the airline’s network to and from London Gatwick rather than Fortress Heathrow. At Emirates, flight numbers EK300 to EK399 are all flights from Dubai to the Far East while flights EK400 to EK499 are all Australian flights.

There are also some special flights numbers that airlines put on certain routes. And even some they avoid. Not many airlines operate flight number unlucky 13. No Chinese carrier operates flight number unlucky 4.

I looked at a selection of prestigious flight numbers for 100 airlines on every continent. I considered number one, 10/11 and 100/101 prestigious for everybody. And for flights to and from Asia I looked for numbers 8, 88 and 888.

This article was written using data from OAG Schedules Analyser: visit oag.com. Thanks OAG!

Now read on…

Flight number one

25 of the 100 airlines had flight number one. Four of these go to London – Emirates, Qatar Airways, Qantas and Delta. Another goes home from London – the evening Malaysian Airlines departure for Kuala Lumpur.

At London-based British Airways, flight number one is not currently operated. It was the morning Concorde departure bound for New York JFK back in the day.

More recently it was a London City to New York single-aisle service stopping at Shannon, carrying the accounting and finance set and guaranteed revenue-wise by the big money institutions.

These days, BA’s London City service is no longer required by the market as the Elizabeth Line whooshes the bankers to Heathrow in a little over 45 minutes. So London City Airport does not have the attraction it once did.

Two number one flights go to Singapore – United from San Francisco and Qantas from Sydney, the latter heading on to London.

Los Angeles, Tokyo, New York JFK and Doha also have two prestigious flight number ones arriving on their runways.

It is interesting to see Virgin Australia sending flight number one to Doha. Qatar Airways holds a 25% minority stake in the Aussie carrier.

Qatar-Australia bi-lateral agreements allow airlines based in both countries to operate a certain number of flights. So to up the frequency, Qatar Airways wet leases it’s own planes to Virgin Australia, who then connect Sydney to the Qatari capital and beyond with more services than Qatar Airways would be permitted on their own. Clever.

Read more

AirAsia is a remarkable airline success story. At first it was a loss-making government-owned carrier. Then in 2001 magnetic CEO Tony Fernandes bought the Air Operator Certificate (AOC) and set about turning the business into an aviation behemoth.

This summer the airline is operating 8,481 flights a week carrying 1.6 million seats and producing 2,163 million Available Seat Kilometres. Each and every week.

The group holds seven AOCs spread over five countries.

Malaysia is the largest, with AirAsia and AirAsiaX delivering 58.5% of capacity. Thailand comes next – Thai AirAsia and Thai AirAsia X deliver 26.7% of capacity. Indonesia (10.0%), the Philippines (4.1%) and Cambodia (0.6%) make up the rest.

AirAsia Japan, AirAsiaX Indonesia and AirAsia India were once part of the family but are not any more.

All the airlines in the group together have grown by 20.4% capacity-wise and by 23.7% by number of seats since 2023.

The group has become a leader in retailing too – with the AirAsia super-app consumers can buy all their travel but also a wide range of consumer products and services from finance to healthcare.

AirAsia is successful because like any well run airline they exercise cost control and practice revenue management. But you know that already. I do not plan to waste the headline.

This article was written using data from OAG Schedules Analyser: visit oag.com. Thanks OAG!

In this article I want to look beyond those points and figure out the secret sauce that has helped AirAsia grow so far and so fast. I have tried to answer three questions.

Q1. How have the AirAsia Group’s corporate structure and the different brands they own in aviation and beyond driven the airline’s growth?

Q2. What macroeconomic conditions in east Asia have created the demand which AirAsia carries?

Q3. Is it possible for other airlines to emulate AirAsia’s success?

Now read on…

Q1. How have the AirAsia Group’s corporate structure and the different brands they own in aviation and beyond driven the airline’s growth?

Many airlines have a complex ownership structure. A holiday company might own the actual planes, if they are not leased. The airline brand might be part of a bigger group.

Read more

Island airlines are fascinating.

Bangkok Airways, possibly the world’s greatest party airline, sends many of it’s planes through the beautiful island of Koh Samui.

Nauru Airlines planes hop around the Pacific. Air Tahiti, Air Tahiti Nui and Fiji Airways link their countries to the world. Air Vanuatu had good intentions but their economy was simply too small to keep them out of bankruptcy.

Another type of island airline, Rhein Neckar Air is based in mainland Germany but flies only to islands.

Closer to my home in the United Kingdom, most of our island airports are in Scotland. This article outlines the major air routes in the region and how the summer season offers a great deal more capacity than the winter..

This article was written using data from OAG Schedules Analyser: visit oag.com. Thanks OAG!

Remote yet beautiful, the Scottish Highlands and Islands have many distinctive yet small communities that require public-service air links to the rest of the country.

The Highlands and Islands also have well-travelled tourist routes and great attractions such as the Callanish Standing Stones (pictured), which I once visited on a misty summer’s day in 2002.

Routes of the Highlands & Islands

The maps below show all Highland & Island routes, which are exempt from Air Passenger Duty. Inverness is the gateway of the Highlands, with flights to London, Amsterdam and other significant cities. Inverness is also a connecting point for Loganair flights from Sumburgh, Kirkwall and Stornoway which continue down to London, Manchester and Belfast.

[Maps generated by the Great Circle Mapper: visit gcmap.com]

Kirkwall on Orkney (KOI) and Sumburgh on Shetland (LSI) are something of the hubs of the north, with 54 and 50 flights a week respectively. Stornoway in the Outer Hebrides (SYY), where I went to see the Callanish Stones, has 43. The others are much smaller.

One of the routes, Westray to Papa Westray (WRY to PPW), is the shortest scheduled route in the world at just two miles and takes just two minutes. I was able to find fares for eight Pounds on Loganair’s website. Many flights appeared to be sold out.

Major airlines of the Highlands & Islands

Loganair, Directflight and Hebridean Air Services are the main airlines. Inverness has services from British Airways, easyJet and others.

Read more

Riyadh Air (RX) is one of the two mega-projects in aviation today. The other is Air India, which I will be writing about soon.

When I last wrote about RX they had 60 Airbus A321neos and 39 Boeing 787s on order, plus another 787 which will apparently be used for training and spare parts. On 16-Jun-25 RX placed another order, this time for 25 data generating A350-1000s.

But where will Riyadh Air fly? Today I wanted to try and figure it out. I will also have some remarks on whether or not RX should join an alliance, and how they should calibrate their network for day vs. night-time flying.

No schedules have been published yet, but there are a number of clues that might give us an idea. These are:

1. The global cities that could be good launch route candidates when the first aircraft is delivered

2. The ten airlines that have signed partnership agreements with RX so far

3. Opportunities for Saudia to consolidate it’s operations that are currently split between Jeddah and Riyadh

4. Sports that RX is sponsoring

5. RX’s other strategic partnerships.

Spoiler alert! My top “magic seven” route shortlist for RX is:

1. Madrid

2. Frankfurt

3. Singapore

4. San Francisco

5. Seattle

6. Washington DC

7. Mumbai

For reasons that I will explain they may not be the international launch routes. But they are likely to be strong candidates. Find out why by reading on…

Unfortunately many of the good candidates are subject to slot constraints, so take everything I have analysed here subject to that rather big caveat.

This article was written using data from OAG Schedules Analyser: visit oag.com. Thanks OAG!

This is the third article in my Saudi aviation series. The others are:

29-May-25: Saudi Arabia's Hidden Airport Champion: Cluster2 Airports is set for growth, but will not take the Kingdom far to meeting it's ambitious aviation goals

18-Jun-25: Riyadh Air: Four prescriptions for success: The Saudi startup needs to do more than most airlines to achieve their shareholders' goals

In the fourth and likely final part of the seriesI will be writing about what I think RX means when they say they want to be the most digital airline in the world. I expect that this will appear in a month of so. Stay tuned!

The launch routes when the first aircraft is delivered

RX’s CEO Tony Douglas reckons that his airline will eventually fly to more than 100 destinations. But it has to start somewhere.

It is quite probable that like many airlines they will put their new planes on a short route first to get the service in tune. The first route could well be Riyadh to Jeddah or Dammam.

What happens next will depend on the pace of aircraft delivery and the order in which aircraft types enter into service. The A321neos were ordered in October 2024. But the 787s were booked in March 2023. Most likely the Boeing twin-aisles will come first.

If two or more twin-aisle 787s arrive in quick succession, RX could fly almost anywhere daily.

A more interesting case is how they would deploy only one aircraft. There are three possibilities.

Cairo is Saudi Arabia’s most frequently served international market, with 532 flights a week in September this year. Istanbul and Dubai come second and third with 249 and 210 flights a week respectively.

American markets are too far away to serve with just one plane.

My guess is that RX will fly to at least one charismatic world city, plus Cairo. The candidate that everybody will be screaming is of course London.

But as the tables below show, London is slightly too far away to have comfortable turnaround times. As is Madrid, and RX is sponsor of football giant Atletico de Madrid, which must surely be a launch route candidate.

Paris and Frankfurt might work a bit better as they are a little closer.

Another possibility is heading east.

Far eastern routes like Kuala Lumpur and Jakarta have strong Islamic heritage. KL is the closest to Riyadh, but as the tables show it is just slightly too far even to tag on a domestic rotation. Jakarta would occupy one aircraft for a day.

A third option is to offer two routes that are either at the long end of short-haul or the low end of mid-haul, depending on your perspective. A rotation pair like Istanbul and Mumbai might work. The two routes each take four to five hours, which gives time for turn-around.

The ten airlines that have signed partnership agreements with RX so far

RX have put together a handy web page summarising all their partnerships. It makes interesting reading. Ten airlines are listed: Air China, Air France, China Eastern, Delta, Egypt Air, KLM, Saudia, Singapore Airlines, Turkish Airlines and Virgin Atlantic.

The maps below show all the routes to and from Saudi Arabia that are currently operated by any of these airlines, except incumbent flag carrier Saudia which I will cover in the next section.

Altogether there are 546 such flights a week, providing 252.6 million Available Seat Kilometres. Egyptair operates 332 (60.8%) of these flights, but they are all relatively short. In ASK terms they only operate 38.3% of ASKs ex-Saudi Arabia.

RX has signed a partnership agreement with two airlines that do not currently fly into KSA: Delta and Singapore Airlines.

Read more

According to the First Epistle of John in the King James Bible “there are three that bear record in heaven, the Father, the Word, and the Holy Ghost: and these three are one.” (1 John 5:7).

Meanwhile down here on Earth three European airlines bear most impressive records.

According to their 2…

Read more

Air travel makes travel to and from islands so much easier. Hopping on a plane and jumping over the water is much easier and faster than sailing across it.

Many of the world’s great airlines serve island countries. Singapore Airlines and Gulf Air of Bahrain are perhaps the two most famous examples. British Airways and Japan Airlines would count as island…

Read more

Love is in the air on Valentine’s Day. Quite literally for Air India, Airbus and Boeing. On Valentine’s Day 2023, the Maharajah announced an order for 470 aircraft. 250 Airbus and 220 Boeing. Where will they use them? I decided to find out…

The aircraft order is mainly single-aisles, with 210 A320neos and 190 737 MAX in the books. The wide-bodies are 40 …

Read more

Europe to South Africa is one of the world’s ultimate weekend breaks. You can leave the office at five o’clock on a Friday, fly nearly 12 thousand miles on a pair of overnight flights and be back at work by nine am on Monday. For travellers originating in London, that is an average air speed of 187 miles per hour.

Heading due south(ish), there is no jetl…

Read more