Until recently, China Airlines flew two relatively successful flights from Los Angeles to Taipei every day. But on March 25, the airline took the risky step of moving one of those flights to Ontario, an hour east (in no traffic) of LAX, the region’s primary airport. Though I came around to thinking a flight like this could actually work, the way China Airlines set it up initially seemed, shall we say, far less than optimal. Now, a few months in, the airline is making big changes by shifting times and swapping aircraft to try to make the service viable.
On the surface, this kind of service seems crazy. Ontario, an airport that had no intercontinental flights at all, suddenly found itself with a massive widebody jetting off to Taipei every day. I and many others figured it was too early for service like this. Eventually, it’s inevitable this kind of route will work since LAX is pushing toward its capacity with no hope for additional pavement, but I didn’t think we were there yet. After talking to a whole host of people at the airport and the airline, however, I started to see some of the rationale develop. There is a large nearby Taiwanese community, not to mention Vietnamese. Those people would gladly flock to Ontario and avoid LAX if given the option. There are also growing economic and manufacturing ties pushed by political relationships that would undoubtedly help the flight succeed. I started to warm to the idea when I wrote about it last year.
Once details about the implementation began to leak out, I started getting nervous once again. China Airlines was originally going to start this as a sub-daily service, but it decided to just jump right in with a daily flight. The right airplane probably would be a 787, but China Airlines doesn’t have any of those. Instead it would use the much larger 777-300ER. Sure, the airline cut a frequency at LAX hoping to nudge people to Ontario, but that could also have had the problematic effect of simply pushing people to one of EVA’s 3 daily nonstops. This had to be too much capacity in Ontario.
To make things worse, the schedule didn’t make much sense. The flight from Taipei was ok, leaving at 4:10pm and arriving Ontario at 1:50pm, but it was the westbound that was problematic. That flight left Ontario at 4:15pm and didn’t arrive Taipei until 8:55pm. That did allow for a handful of connections in China Airlines’ late Taipei bank, but it didn’t allow the flight to benefit from the full power of the Taipei hub.
The end result of all this has been less than stellar performance. We don’t know a ton of detail, but Ontario puts out a monthly traffic report. That report breaks down the number of passengers per airline for each month. Since China Airlines has only one flight at the airport, we can calculate load factor for that flight. The last report release was for May, so here’s what we know up through that point:
Remember, it only operated for 7 days in March, so that first month doesn’t really count. April and May are more representative of what’s been going on. There are a couple potential issues with the data here. First, I assumed a flight operated every day. If there were any cancellations, then that would positively impact the load factor. Second, this is for both directions. If there is an imbalance for some reason, then it may be that one way does well and the other doesn’t. I don’t have that information. Third, this isn’t exactly high season, so it may not represent what an annualized load factor would be. Hopefully June and July were better.
Regardless, it’s not good. So what is China Airlines going to do about it?
Earlier this month, it announced it was going to tweak the timing of the flight. Starting September 17, the westbound flight will switch to a redeye leaving near midnight and arriving Taipei in the early morning. Even though I personally hate westbound redeyes, that will open up a ton of new connecting opportunities beyond Taipei. The return will shift a couple hours later as well, but that will only help in allowing more inbound connections into Taipei. At the same time China Airlines cut the flight to operate only 5 days a week from September 17 through October 27, a slower time period.
Now, China Airlines has quietly updated its winter schedule operation which begins on October 28. From that day, the airline will use a smaller A350-900 with 32 business (compared to 40 on the 777-300ER), 31 premium economy (vs. 62) and 243 in coach (vs. 256). It looks like the flight will go back to operating daily at that point. Will it help to have 306 seats on every flight instead of 358? You bet it will. But will it be enough? If the A350 had flown in April and it May, it still wouldn’t have cracked a 70 percent load factor. Maybe with better connections that will change, but this is far from a guaranteed success story.
I don’t have any sense: how do these numbers compare to typical startup routes? Obviously they’d be terrible for the annualized load factor on an established route, but doesn’t it take quite a long time for customers to start to use a new route?
I have heard long-haul international flights take about two years to become profitable. The lower operating costs of the A350-900 should help.
Alex – I think this one is hard to pin down. It’s not really a completely new route. It’s just been shifted from LAX, so it should perform better in the beginning than a truly brand new route would do. Obviously China Airlines isn’t happy with the performance since it’s already making big tweaks only a couple months in.
It’s been shifted from LAX but then on the other hand it’s also trying to effectively establish an entirely new international airport, so I would think those two things might balance each other out or maybe even weigh in favor of this being harder than a typical new route. Even if you can persuade LA-area residents to use Ontario, there’s a whole other marketing challenge with explaining to Asia-origin passengers what the hell Ontario is.
Another reason this might even be harder to establish than a brand new route: if someone wants to go (say) between eastern Washington/northern Idaho and TPE, they’ll search for flights between Spokane and TPE (almost certainly expecting a connecting itinerary). If an airline started a new nonstop on that route, it would turn up. But if someone wants to go between the Ontario area and TPE, many probably just go straight to searching for LAX-TPE and would never see this option (as grichard notes below). So flight search engines won’t do even that basic marketing for the airline on this route, whereas it would on a truly new route.
(Went away for a week and didn’t see replies until now.)
Interesting to note, that CI ONT service regularly prices $70-150 cheaper than their LAX service, while business class is $700 cheaper consistently. So besides loads, yields are worse at ONT. Not good combination.
Also seems like they also shot themselves in the foot at LAX being left with single flight option against EVA’s 3x daily service.
These low fares are the typical introductory fares in a new market… I would posit that if they are that deeply discounted a year from now, it is a significant issue, but for now, it is to be expected (good marketing of a new service).
CI ends up competing against itself across town at LAX. Even with the lower pricing its clearly not driving the traffic to use the ONT flight.
In the short term, yes… but if you consider the fact that 70% of all outbound SoCal ticket purchases to Taiwan originate from zip codes closer to ONT than LAX (as well as 70% of ticket purchases to Vietnam, 65% to China PRC, 50% to the Philippines, and 46% to South Korea), CI could also be deemed to be making a smart strategic decision in the medium and long-term… customers that can fly from an airport closer to home — particularly in the Los Angeles, the most heavily-congested city in the world since 2016 — pay a premium for the convenience. Just take a look at domestic fares from SNA, ONT, and even BUR; it is reasonable to project that international demand could exhibit similar trends over time as international capacity at ONT grows. And when you adjust for freeway traffic congestion during peak periods, 10 million of the 19 million people in Southern California/Greater Los Angeles live closer to ONT than to LAX. Those are the only two airports in the region that have the infrastructure for trans-oceanic flying, and if the alternative is 35-mile-in-3-hours trip to LAX or a 35-mile-in-30-minutes trip to ONT (or far worse as those distances increase in what is a megalopolis that roughly runs 80 miles by 80 miles); ONT could arguably be worth the premium for the business traveler that values his/her time.
I’m surprised at the point-of-sale stat for Vietnam considering the demographic cluster at the western edge of Orange County. Depending on precise ZIP data, the driving mileage LAX vs. ONT can split nearly evenly, which might help explain how ONT could be classified as “closer” than LAX.
I’ll agree with the potential of ‘first-mover’ advantage for CI (and, to some extent, HA at LGB.) Both of these attempts represent opportunistic marketing strategies to capture further upstream proven sources of demand.
Also, in the immediate to medium term, commuting to/from LAX will only worsen with traffic growth and landside construction. It will be interesting to observe if the near-term access challenges at LAX contribute to air passenger growth at the L.A. area satellites.
Since this will largely be LA O&D traffic, I’ll bet they have trouble with searches missing them. Lots of people search by airport code, and if I were searching for a flight from Taipei to Los Angeles, I might well search on “LAX” and never notice this flight.
Makes sense, and as someone who has visited SoCal but isn’t familiar with the area, I didn’t even realize that ONT could do international flights of this length. I’m not an advertising guy, but as a layperson I would presume they are paying for web-based ads when people search for flights from LAX to Taipei, in addition to hitting the local ethnic media.
A few additional facts to consider:
1. CI’s load factor for June rose to 71.8% — not world-beating, but not terrible for a third month of a brand new international service. (And, yes, April and May are the lowest-demand months of the year in the U.S.-TPE market).
2. The imbalance of inbound and outbound load factors that marked the ONT-TPE service during April (64.3% inbound/50.4% outbound) and May (64.5%/52.3%) actually flipped during June, to 69.5%/74.1%). This doesn’t necessarily address the question of point of sale, but I would suggest that this change might have been influenced, at least in part, by significant marketing/advertising support that the Ontario Intl. Airport Authority (OIAA) brought to bear in Taiwan, in coordination with CI’s marketing team. Prior to June, the OIAA’s marketing/advertising support had been focused on the Southern California market. Such support in both sides of the market will continue for the foreseeable future… which will arguably address the clear “Where is Ontario?” issue.
3. While the shift to an A350 from a B777-300ER is certainly a welcome change — for fuel economy as well as market calibration reasons (ONT is more appropriately a B787/A350 at this stage in its maturation, given the lack of connectivity beyond ONT at this point), the change in the departing flight schedule is very significant.
Today, CI’s ONT-TPE flight arrives in Taiwan at 2055. At that time, there are only logical connections (i.e., within 3 hours) to just 4 Asia/Pacific markets (HKG, BKK, MEL and SYD). In addition, VN adds an alliance codeshare flight to SGN. (Obviously, non connections to China PRC by CI, MU, or CZ are included in these totals).
With the revised winter schedule, outbound passengers from ONT will be able to make connections within three hours of arrival at TPE to 20 destinations in the Asia/Pacific (FUK, BKK, RGN, TAK, HKG,PUS, SGN, NGO, MNL, PNH, DEL, SYD, ICN, TOY, GUM, KOJ, KIX, OKA, SIN, and HAN). During that same three-hour window, VN will operate its own codeshare connections to SGN and HAN. If you expand the window for connections to four hours, CI connections to KUL, CTS, and DPS will be available.
Looking at the inbound side of the market today, CI connects ONT to 14 Asia/Pacific markets within 3 hours of its 1610 TPE-ONT departure: TAK, BKK, PUS, HKG, KOJ, ICN, NGO, ISG, TOY, SIN, HAN, KIX, SGN, and PNH. If one expands the window for connections to four hours, MNL, FUK, OKA, and NRT also come on line.
With the changed winter schedule, that number of connections within three hours of departure will drop to 10 (NRT, GUM, KIX, RGN, HKG, TOY, BKK, PNH, HAN, and SGN). Within a four-hour window, TAO, YTI, SIN, ICN, and NGO would also become viable connections.
Obviously, the Asia/Pacific-Los Angeles market for connecting traffic is the most competitive of any U.S.-Asia/Pacific market, and additional connecting markets won’t necessarily (a) bring higher-yield traffic, or (b) guarantee the viability of the service. However, with more connections over TPE to strong markets in Southeast Asia will limited or no nonstop service to the Los Angeles region today, CI’s ONT-TPE service can reasonably be expected to benefit from significant additional passengers — particularly in Southern California, where more outbound O&D passengers to that region live and work closer to ONT than LAX, and are arguably open to the value proposition of spending a few hours connecting over TPE, in lieu of facing the Russian roulette that is freeway traffic to drive to LAX for a flight.
I presume they were hoping for more interest from the Taiwanese community in Monterey Park/Alhambra/San Gabriel area or the Rowland Heights/Hacienda Heights area where you live within the same driving time between ONT and LAX (depending on traffic) and ONT is certainly easier to get in and out of than LAX.
CI also offers a complimentary shuttle bus to Monterey Park or Rowland Heights for pax on their ONT flight.
The problem is their reputation in said community is a bit meh. Most prefer EVA over CI to TPE unless a steep price difference comes into play. Not to mention in the event of OSO you then have to get from ONT to LAX for any alternate flights.
I’m curious about sales & discounting. What percentage of tickets are published full tariff, discounted corporate and consolidator.
PF – That information isn’t public.
Not that this factors into the economics, or many people’s purchase decisions, but I reckon the A350 is the best widebody currently flying. I’ve done a few sectors recently and it seems to deliver on the promise of the Dreamliner to enable you to arrive more refreshed.
It looks good too, sitting in MEL last week waiting for my CX A350 there were CI and TG A350s parked and it looks very sleek.
I noticed CI began announcing their change at ONT through their website. Knowing it’s the early days since the announcement I wasn’t expecting a big change though I went to expert flyer to see if there has been a change. I checked their economy section.
Noticed a slight rise on ONT though comparable to LAX. Let’s see what happens in the long term.
A few weeks back I had noticed a slight change regarding CI ONT Night time flights. Though, didn’t want to come off as novice since it could have been a fluke.
Now its obvious. I went back to check again Expert Flyer (checked their economy/premium economy sections for CI23(ONT)/CI7 (LAX)), and checked the flights starting on Monday Sept 17 (ONT)/Tuesday 18 (LAX) and Thursday Sept (20) ONT/Friday Sept 21 (LAX). From what I see, the ONT flights are quickly filling up while LAX is struggling to get passengers. Tuesdays are toss up between LAX and ONT. Though LAX has the upper hand during the weekends (odd due to the traffic nightmares especially Friday evening). It’s a similar pattern every week afterwards. Up to late Oct/Early Nov.
Moral of the story, from what I’m seeing, CI might of made the right changes for the survival of the route, and perhaps just perhaps saved the embarrassment from officials at ONT who pushed for this flight.