4 ms·
That needs to be sourced from somewhere; the classic understanding of airlines is they usually overbook [0] and rely on people not turning up. In fact, the who
by roenxi 5y ago
That needs to be sourced from somewhere; the classic understanding of airlines is they usually overbook [0] and rely on people not turning up.
In fact, the whole "lets not sell our inventory, we'll make higher profits!" idea here is suspect. The idea that having unsold inventory in the model is a good idea is just false; having inventory left over is always undesirable. If it was sold at the last minute then they would have more profits. The scalpers are incentivised to minimise leftover inventory to the point where it isn't going to be a big deal.
[0] https://www.travelperk.com/blog/overbooking-airlines-quick-guide-business-travel-processes/ https://www.travelperk.com/blog/overbooking-airlines-quick-g...
- Retric 5y agoIt isn’t a question of what happens on specific flights, it’s a question of what happens generally. Global load factors are ~85% which is a long way from constantly overbooked. Pre pandemic to get a normal picture: https://www.airlineratings.com/news/packed-planes-new-normal-airlines-fill-seats/ https://www.airlineratings.com/news/packed-planes-new-normal... Low cost US carriers have a load factor of 90% that’s a trade off between fixed aircraft size and variable demand. However, US carriers only average a load factor of 86.5% meaning legacy carriers like American Airlines have a below 86.5% capacity factor and that difference between 90% and ~85% is exactly the optimization I am talking about.
- roenxi 5y agoI'm guessing that is a statistical artefact. They'd be running routes that are completely packed in one direction and not so full in the other due to seasonal effects or due to day/night cycles. So even if they were trying to overbook they'd still have quite empty planes much of the time. 85% capacity use of an asset is really quite high. If they managed to achieve mid-high 90s that would be startling given how good humans are at missing flights and changing plans. To repeat myself, if the plan is to leave your capital unutilised for 15% of the time, on purpose, when there is an alternative, then the planner is a lousy capitalist and doomed to soon be a broke one.
- Retric 5y ago> I'm guessing that is a statistical artifact. No, it’s a difference in business models. Low cost carriers approach things very differently. https://youtu.be/1-uNMj57Y4c https://youtu.be/1-uNMj57Y4c It’s not that their wasting this 15% of their unused seats airlines are simply more profitable with this pricing model. “While you might expect the airlines to lower their prices a few days before the departure day to occupy the last seats, the opposite is true: Selling 20% of the remaining seats for $1,500 is more profitable than selling half of them for the regular fare of $550.” https://flightfox.com/tradecraft/how-do-airlines-set-prices https://flightfox.com/tradecraft/how-do-airlines-set-prices Further that’s a statistical average, they might not have any last minute travelers on flight X or they might sell out flight Y 1 hour before departure. As airlines can’t both raise and lower prices at the last minute they A reserve seats for the possibility and B frequency lose the gamble.
- roenxi 5y agoWell, sure, but that is assuming that the market is exactly 85 people. If we hold the number of people fixed and it is impossible to sell all of the inventory then holding back inventory is much more effective - because we have assumed it is impossible to sell. The hypothetical in that article would make more money if they sold 80 seats at $495, 15 seats at 900 and then 5 at $1500 (that makes $60,600). There are strong incentives here to fill the seats. And they are doing the scalpers trick where someone desperate to get on the flight can. As opposed to when the good would be overbought and someone willing to pay $1,500 has to start paying off passengers for their booking instead of giving more money to the people who put planes in the sky. Plus, if we're both running this strategy, I have huge incentives to offer $1,495 seats to steal your high paying customers off you. It just doesn't work as a strategy unless it is already impossible to fill the plane - which is quite likely the case.
- Retric 5y agoNo, the math works fine. Ignoring baggage fees and first class vs catch etc. If you have a flight with 300 identical seats and you could sell everyone single one of them for say 100$ and that’s your maximum possible revenue from a single price point. Now the question becomes how do you capture value from people willing to pay more than 100$? You can play some games with round trip tickets and minimum stays, but again let’s ignore that and just focus on the ramp up. If you designate say 75% of your ticks at a lower price of X then some people willing to pay more will just buy at that price. Essentially the trick is to minimize people who are willing to pay more who end up buying cheaper tickets. And it turns out people booking close to the flight are generally willing to pay more. Sure six months out they still sell off up to 75% of tickets at ~100$ raising or lowering prices based on demand so their losses are limited if nobody else buys the ticket the flight is still mostly full. But at that 75% they start the ramp up and they also ramp things up as the flight gets close. 2 weeks out they say all tickets are 200$ even if some of those 75% are unsold. At the last day bump it to 400$. If over 1 month we can ignore seats at 100$ because that’s unchanged. Sell 1,000 seats at 200$, and 500 seats at 400$ and fly with say 1500 empty seats that’s 400,000$ where you would only make 300,000$ if all 3,000 seats where sold for 100$. But what if they dropped prices at the last hour? Well most of the time they have empty seats so many would avoid the 400$ ticket and would tend to wait around for that window especially for flights from A to B leaving every hour. Peak prices are a game of chicken and they can’t blink or they lose billions in revenue. This is of course heavily simplified but it does demonstrate the basic trick I am talking about where they extract value from unsold inventory.