wjw

Random question which probably has an obvious answer that I’m somehow not realizing…

What is the purpose/utility for Air Canada to codeshare a flight operated by, for example, LH?

I get that code-sharing allows Air Canada to sell tickets on routes they don’t fly and couldn’t otherwise market, but this isn’t the case for alliance partners, right?

I’ve noticed recently when booking some TATL flights that the Air Canada website gives the option of booking LH 471 or AC 9104 (its codeshare). Same price.

Is there a reason that Air Canada markets AC9104 when they could just sell seats on LH471, as they already do? Is there any advantage as a passenger to being booked on the native flight # vs the AC codeshare? As I understand it, LH metal means the experience will be dictated by LH regardless, and the 014 ticket stock means I’ll get SQD regardless.

Stranger

This is a good question.

About offering both options, I believe it used not to be like that, appeared somewhat recently. It seems they have allowed for more non-AC flight numbers to be quoted on their web site.

Incidentally, I believe it's not always the case that both options are priced the same, at least in some fare categories.

zrh2yvr

Not to get too technical but it helps with fare calculations sometimes. It also may allow them to approve certain booking classes separately from the LH availabilty. In complex itineraries you may need to book a particular carrier code to allow the fare to price and this allows it to happen. Then there are different types of code shares. I think you can just book the AC code share on TATL LH flights without the need to have any other AC segments. However, on the intra-european flights - the code share cannot be booked without the linked TATL flight. That being said - on the intra-european flights - the AC fares will price the same regardless of which carrier code you book within Europe (OS/LH/LX). In one example (for an LX ticket) we needed to get to BOG out of ZRH and the fare required LX code - we got the LH flight FRA-BOG on LX code - without that code being there - it would not have priced.

Changeup2000

From my recent rebooking experiences, it is quite interesting: I had a confirmed upgrade to business class. If I rebooked to AC codeshare flights, it only allowed me to rebook into economy cabin, but it allowed me to rebook into the business cabin on the partner airlines numbered flights, even though they were exactly the same flights.

Stranger

That might be because on the codeshare, AC had exhausted its P quota.

montrealer

I wonder if it's just P quota, or glitches in the fare rules and booking engine?

A few times, I've had to pay $ to a travel agent to book me a P-class itinerary with AC on the intercontinental legs, and a *A partner airline on the other end. The AC web engine would only offer me the AC codeshare flight number, in economy. I needed an agent's help to book the leg under the operating carrier's flight number, in P (or equivalent). This was even when Expertflyer reported P>0 on the AC codeshare itself.

I've definitely had this with NH as the *A partner, and also at least one European airline -- can't remember if it was TP, OS, or LH.

canadiancow

But that's all because of rules written into the fares... by the airlines.

I don't think "you need a codeshare because the airline says you need a codeshare" really answers the question.

montrealer

The following is a reasonably well-informed opinion based on working with some non-*A airlines on strategy and revenue management some years ago; however, I do not know the specifics of AC and *A, and how practices may have evolved.

In the good'ol'days, fares were published by IATA, a cartel of all the major airlines. There were no code-shares, no or very little fare discounting, and the price of a ticket was allocated to the operating carriers according to a formula, I believe largely dependent on mileage. Interestingly, it was considered too complex to do the calculations for all tickets, so a certain amount of random sampling was done to calculate for some tickets and extrapolate for the rest.

In the good'not'so'ol'days, after airlines were doing their own revenue management and competing on price, two types of code-sharing agreements developed. In a blocked-space agreement, the operating carrier offered up X seats to the marketing carrier, and the marketing carrier did their own revenue management on that block (fares, booking classes, etc.) In a free-sale agreement, the operating carrier controlled the booking classes and up-to-date availability, and the marketing carrier merely re-sold individual seats from that as available. The specifics of each bilateral agreement controlled the revenue-sharing. By and large the operating carrier got the lion's share of the revenue.

Layered on top of this, individual airlines began to enter JVs on certain routes or markets, where they agreed on how to split revenues and costs, or profits, in that market. Generally, these were designed so as to make the economics metal-independent, i.e. so the airlines didn't care if a given passenger flew on a ticket issued by airline X or Y, and on a flight operated by X or Y. These arrangements were then implemented via codeshares and routing/application rules in the pulished fares, plus a lot of wizardry on the accounting back end.

Finally, alliance agreements included various frame agreements on code-sharing and JVs between members. These would provide frameworks where no specific pairwise agreements existed, but (usually? not sure) could be overriden by specific JV or code-sharing agreements where negotiated.

All this has generated a messy dog's breakfast. What I think it means is as follows:

• If a travel agent seeks to book a mixed ticket, different airlines' revenue management will be pinged depending on whether all-AC or mixed-*A flight numbers are being queried. In the mixed-*A case, each airline's RM is queried using the appropriate fare basis, with a mapping if needed. In the all-AC case, only AC's RM is queried, though on many (most? nearly all?) routes it will be interconnected on the back end and reflect the operating airline's data.
• If you use the AC web engine or call AC, it will preferentially serve up AC codeshares, and may or may not block non-codeshare flights. That probably varies depending on the phase of the moon and whatever quantum superposition principle applies inside AC's IT infrastructure at that moment
• FF benefits to the passenger depend (within *A) only on operating carrier, so it won't matter. Fare eligibility may, by design or by accident, vary, due to the interaction of complex fare rules and fare basis mapping between airlines. The most frequent source of inconsistencies are deep-discount premium class bookings, and mixed-class bookings with premier economy on some legs, since that is where the revenue management of different airlines differ the most.
• On the back end, the $ will be the same to AC where a JV exists, but may be different where that is not the case. There are a lot of unexpected surprises in sweating old IT systems like this and in a given situation, it's quite possible no-one has noticed.

Personally, on most of my tickets, the price and availability have been the same for using AC-codeshare flight numbers versus operating airline native flight numbers, though there have been a number of situations where using the native operating airline flight number, and doing a bit of a song and dance to get it, have paid off.

canadiancow

Thank you for the post.

But I still don't see any reason for codeshares to exist. "They allow specific agreements" could apply equally well between any interline partners without codeshares.

Stranger

Here is a surprising one, at least to me: looking at TLS-YYC return, with first leg on Sept 8 at 10:15 am, TLS-FRA-YYC, business low, the AC codeshare shows one seat available, with TLS-FRA in J.

With the actual LH flight number, it gets: "Mixed Cabin: Between Toulouse (TLS) and Frankfurt (FRA) you will be seated in Economy Class."

When I would normally expect the opposite.

RangerNS

I'm sure it all boils down to, at some point, either evidenced or not, the airlines concluded that PAX and/or TAs would trust particular itineraries apparently on a single carrier more than those split across carriers. Or would interpret a codeshare as a high level of trust; to OEM a flight, so to speak.

Going way back, a codeshare might have been easier to book than splitting a trip across multiple call centers (telex centers?).

And because its still a feature of the system... it's still a feature of the system.

TravellingChris

What I find equally fascinating is "unfaithfulness"--where an alliance member pursues a codeshare relationship with a non-alliance carrier in order to serve destinations that the former doesn't. The problem arises when those same destinations are already served by a fellow member of the first airline's alliance. The whole idea behind international alliances is to partner with carriers across the globe in order to serve most destinations that a traveler could want to fly to, and to feed traffic to and from alliance partners for mutual benefit, rather than that of a third party.

A good example was the codeshare agreement with Cathay Pacific that Air Canada reached a few years ago--AC placed its code on onward CX flights from Hong Kong to destinations such as Manila, Bangkok, Saigon and Kuala Lumpur. However, each of these cities was already served by a Star Alliance carrier. For example, AC should have been feeding Bangkok-bound passengers to THAI at either HKG or NRT rather than Cathay Pacific. (The agreement was reached before AC started non-stop flights to BKK itself.) Manila-bound passengers should have been placed on ANA flights from NRT rather than CX ones via HKG. Indeed, the whole idea of an alliance is to give your fellow member airlines preference when routing your passengers to destinations you don't serve directly.

We see this again with AC's relationship with Emirates. There are plenty of onward destinations that Emirates serves (such as in the Middle East and Subcontinent) that AC doesn't. However, most of those cities are already served by Star Alliance carriers such as Turkish Airlines and Air India. Shouldn't AC be directing onward traffic onto their flights instead of those of non-member Emirates?

YOWgary

ANA has Star Alliance partnerships with both United and Air Canada that would enable a passenger in Tokyo to fly to Ottawa, and a passenger in Vancouver can choose to reach Kuala Lumpur on AC plus any of Air China, ANA, Asiana, EVA, Thai, Turkish or Singapore, but the alliance isn't broken by this.

Agreements of this nature are generally intended to provide a substantially new capacity to each airline; for example, Emirates and Cathay are never going to launch service to Winnipeg or Thunder Bay, and Air Canada's never going to launch service to Conakry or Surabaya.

canadiancow

I disagree with that: A map from the Great Circle Mapper - Great Circle Mapper

I'd much rather the HKG connection. More options are always better.

AC fares to India (okay, I only checked a couple for YYZ-GOI) do allow routing on AI. Or UK. Or EK.

But this is a totally separate discussion from codeshares. All of this could be done (or is done) with native flight numbers.

rjburns

Codeshares rarely result in any savings. But they nearly always result in major pain during IRROPS.