Monday, November 25, 2019

Thanksgiving Week Break

‪No post this week! This week, the only bird I’ll be talking about is turkey. Maybe pheasant. ‬

To all the crews, mechanics, ground handlers, customer service reps, airport ops, airport volunteers, air traffic controllers, ARFF personnel, etc, etc, etc, thank you for your service to all passengers, both rookies and experts. If you didn’t spend time away from your family this holiday season, much of the public would not be able to spend time with theirs arguing about football and politics.

Thank you to all those serving this holiday season. ‪Have a Happy Thanksgiving!‬

Wednesday, November 20, 2019

Placing the CRJ-550

A few weeks ago, United rolled out its latest new-ish aircraft: The Bombardier (soon to be Mitsubishi)  CRJ-700 CRJ-550. Unnecessarily confusing? Ya, we are just getting started.

If you are in one of the Midwest or the East Coast cities which has the new CRJ-550 service, I can almost guarantee your local news agency has reported about this "new" aircraft. I have to hand it to United's press team, I personally have not seen this much positive coverage about a reconfigured aircraft. Granted, they are selling it as a "new" aircraft. But is it really?


Even though the aircraft is a variant of the CRJ-700, technically, yes. In the FAA's speak, the aircraft is technically a CL-600-2C11. The FAA classifies the CRJ-700 as CL-600-2D15. So if the aircraft is not the same, what changed?

While all systems and the build of the CRJ-550 and CRJ-700 are the same, the CRJ-550 has lowered a certified max gross takeoff weight of 65,000lbs from the 75,000lbs available to the CRJ-700. In addition to a max certified weight reduction, the aircraft gauge is also certified at a max of 50 seats. The certification of the weights and gauge as well as the new aircraft type all have to do with United's scope agreement. As defined in United's pilot agreement signed in 2013, a 50-seat aircraft is any aircraft with 50 seats or fewer AND a max gross takeoff weight of 65,000 or less. It simply was not possible for Bombardier to offer the 75,000lbs CRJ-700 with fewer seats to United. Both the weights and gauge had to be certified lower.

In terms of deployment, to understand how I believe United decided which markets to launch the aircraft with, we have to understand exactly what parameters the aircraft can operate. First, the consideration of distance has to be addressed. To best understand what type of mission distance the CRJ-550 to fly, it is best to compare the against the CRJ-700 mission profile.


Bombardier is currently marketing the CRJ-550 to have a maximum distance of 1,000nm, roughly a 30% mission distance reduction compared to the CRJ-700. This can be traced back to the max takeoff weight. If the aircraft were to offload 20 passengers at standard weights and with their associated baggage, you can expect to offload 4,000lbs in weight. However, the other 6,000lbs has to come from somewhere and it appears to be the fuel that took the hit.  With less fuel, less mission mileage. For the November routes, the CRJ-550 maximum mission distance appears to be in the 700-mile range or roughly 70 percentile of the CRJ-700. Future schedules continue to show this trend.

Next, the United has announced the aircraft would be within the Chicago hub and later expand to EWR and IAD. We can also assume the CRJ-550 will only replace other express flying, which given they are in the middle of contract negotiations, I'd say this is a safe bet. This allows us to start to build a list of possible cities and their pairings which may be eligible for the aircraft. Based on these conditions, 200 markets appear to be eligible targets to receive the new routes.


When we start to dig deep into the initial launch routes, we see a pretty clear pattern at first. The routes that were selected to launch the aircraft are typically markets that are performing double-digit above system RASM performance. On average, these markets outperform the 2018 system average by 18% in RASM performance. This makes sense as higher-performing markets may be indicative of higher customer loyalty or higher yieldable demand.


Now, call me a cynic, but I do have to wonder if these markets were selected due to their higher performance to help with messaging. What do I mean? Let's assume for a moment that the new flights do not stimulate incremental revenue performance as expected. These higher-performing markets would still allow United to claim the new fleet type is profitable, even with the up to 10% CASM penalty they will incur.

If we dig a little deeper into these markets, clearly they are being yielded pretty high. On the load factor front, each of these markets has plenty of room to still take additional demand. The softer loads make for one great argument why United would not want to swap service out to a larger gauge aircraft to put higher service classes in the market. Another similar theme with the routes, most of them are missing a significant amount of higher class service.


The CRJ-550 has 20% of its seats dedicated to first class with a configuration of 10 first-class seats, 20 Economy Plus, and 20 economy seats. This is a significant upgrade compared to other regional jets operating within the markets.

With the premium first-class product on the regional aircraft, United does have the opportunity to further increase the yields within the markets. In these markets, United has seen the potential to increase fares in the locally as high as 50% compared to economy fares. That said, the sample size of first class tickets on the routes is limited, given the limited ability to book first class seats. I do think one of the most unspoken revenue generators on the flight could actually be the Economy Plus product. Depending on how the product is managed, each Economy Plus fare could generate more ancillary revenue for the carrier.


As United continues to grow its CRJ-550 fleet, currently at 10 aircraft to 54 aircraft by fall 2020, other high potential markets can be quickly identified. Assuming no operational and scheduling constraints, I suspect we will continue to see United target expansion markets with low premium penetration cabins and high revenue performance.

It is impossible to use public information to gather other likely drivers for United's selection process.  I strongly believe United would have access to data on its customer segmentation and behavior data to help them identify routes with the highest upgrade potential. Additionally, network planners and corporate sales teams would have access to corporate sales information which might also help them target potential upgrade opportunities as well.

I would be remiss if I did give my two cents on the CRJ-550 approach. I have to hand it to United, given their pilot contract constraints, this does appear to be an aircraft which would deliver superior service versus a traditional regional jet product. However, this feels like a bandaid to a larger issue with United and their scope clause. Currently, United is locked in contract negotiations with their pilots and at least publicly it does not appear like they have made much progress recently as their scope clause continues to come into focus.


Taking a closer look at scope, when it comes to fleet deployment, United is at a significant disadvantage compared to other legacy carriers. Within the domestic network, United dispatches 35% of their domestic flights on 50 seat aircraft. This is a full 15pts more of small regional jet deployment compared to American. In the 60-70 seat category, United has 5-10pts fewer regional jet deployments versus American and Delta. This is the scope "limitations" which United continues to focus on.

Increasing CASM to swap out the ERJ-145 might be the right call in a lot of markets, however, with the jet only planned to operate out of ORD, IAD, and EWR with a limited 700-mile range, United will likely hit a diminishing rate of return on the aircraft relatively quickly. What United needs, but likely will not get in their next pilot contract, is to increase gauge within their regional operation, regardless of which pilot group operates it.

I do have to wonder what the fate of the CRJ-550 would be if tomorrow United received their must desired scope relief. Would United keep receiving the CRJ-550, or would they convert all orders to the CRJ-700? What would be the fate of the CRJ-550s that already have been delivered? Could a supplemental type certificate raise the max takeoff weight and recertify the max passenger configuration?

These are all questions for individuals with higher paygrades than me, but I suspect someone already has most of these answers worked out in the unlikely event United's scope world changes tomorrow.


Edit: Online it was correctly pointed out to me that the current airframes are in fact old CRJ-700 airframes. I had assumed, incorrectly, these were new deliveries from Bombardier to their regional partner. If someone has an actual fleet plan with born on dates of the current and future CRJ-550s, I will edit and update the post.

Wednesday, November 13, 2019

What's Fueling Delta's Georgia Increases?

On November 4th, I was befuddled. For those that know me, this is nothing new. But on that day Delta put out a press release announcing the carrier would be increasing frequencies from Atlanta to four relatively small Georgia communities: Albany; Brunswick; Columbus; and Valdosta. The total population of the cities totaled just over 343,000 living within the cities.

While I was not surprised by Delta making simple capacity decisions in these cities, it did shock me to see a national press release for roughly 200 daily seats (400 round trip seats). This equates to 0.0002% in increased daily seats in and out all Georgia cities on Delta. However, Delta did not just put out a press release, the carrier coordinated with the Governor's office for a quote. I would understand including statements from local mayoral offices about the 30% daily seat increase in the local community, but the Governor? It did not really make sense to me.  

Here's the governor's statement: 
"With roots in Georgia dating back to 1924, Delta Air Lines has helped put our state on the map as a gateway to the global economy. Delta serves 80 percent of key U.S. destinations within a two-hour flight from Atlanta, and as these new flights begin operating, they will open new doors for economic growth in every corner of our state. I am grateful for Delta's partnership and their continued investment in Georgia."
Delta's statements were also extremely pro-Georgia as well:
"The airline employs tens of thousands of Georgians – it's among the state's top private employers – and contributes millions of dollars and countless volunteer hours to charities and organizations throughout the metro area."
Now, I am not a public relations professional nor a professional writer (clearly), but I can think I can tease out Delta’s messaging with these routes. If I were to restatement the quotes, "Delta drives a massive amount of the economy in Georgia". But honestly, how do these four routes tie to Delta's economic drive within the state? I suspect, very little. But let's take a look at each of the route's performance and see if they warranted the extra capacity or if something else helped fuel these increases over the line.


A quick RASM curve review shows that each of the markets appears slightly above Delta's YE2Q2019 RASM curve. Generally, each route shows revenue performance is roughly 10-20% above the system curve. On is the surface, each of the routes does appear to be performing well. This does ignore any possible impacts of small city inefficiencies (cost spread) and CASM impact associated with a smaller gauge aircraft.

When we trend the service over time, each of the routes has shown improved revenue performance over the last three years. The graph below is annualized to remove volatility in seasonal performance.



Much of the increased route performance appears to come from the load factor side rather than prorated fares on the routes, however, what strikes me is none of the routes are particularly full.

While we do see loads peaking around the mid-80s for ATL-VLD, the other routes typically are performing below 80%. Anything below an 80% load factor should be able to pick up additional passengers should Delta want to carry them. It is possible the routes may be filling on a peak day, however, Delta should have the flexibility to add a flight or increase gauge on those specific days, rather than another full daily flight to each of the markets.


While the routes are performing slightly above the system revenue performance, given there is plenty of room to carry additional passengers, I suspect there might be more to the story here than just performance-based additions to Georgian cities. Why not swapping to a larger gauge aircraft during peak performance? To me, it is is too clean of a decision for all four cities to see flight increases all at the same time without another underlying theme.

What exactly do I mean? I suspect, but will never be able to prove, the state of Georgia's on-again, off-again battle with its aviation fuel tax may be the true focus. After Delta brokes ties with the NRA, Georgia senators rejected a bill which was backed by the governor and passed by the house. The bill would have suspended the aviation fuel tax for 20 years. I wonder if these capacity investments are part of a larger campaign to show Georgia legislators how Delta is deeply invested within the state even though the flights are really immaterial to the network.

Delta's press release took great pains to show the investments the carrier was made by adding roughly 200 seats into the state of Georgia. However, Delta frequently makes similar changes across their network which are never discussed, let alone with a press release. Further, the press release quotes the governor who has been actively championing the suspension of the jet fuel tax is really the cherry on top for me.

While I understand how important these additional flights were to the individual cities, they largely are immaterial to the state. These cities generated roughly $42M in O&D revenue into the Delta network, which ironically, is the estimated tax saving amount for Delta if the fuel tax was suspended.

I do want to be clear on these route additions. Each of the routes were performing above system unit revenue averages, however, a 30% increase in seats on a mid-70% load factor route does not always pan out. It is completely possible the new flights could create unique connections that the other flights did not generate. These new flights might be the best thing since sliced bread, however, they do not feel like route performance-based additions, rather it feels part of a campaign to get their fuel tax suspended.

Wednesday, November 6, 2019

CVG-SFO: One Complicated Route


Thank you, everyone, for sticking with me for another week. This week is a much longer analysis of CVG and United's CVG-SFO market. This review is much more in-depth than previous posts. Please let me know if you like the more in-depth reporting or a shorter read. Thanks! 

Cincinnati has been an interesting city to watch over the last few years. In 2017, Delta announced it no longer considered it to be one of their hubs but still considered CVG as a focus city. The declaration that CVG was no longer a Delta hub came after over a decade of near-continuous trip and seat reductions in the city which stabilized around 2015.



While Delta went into a status quo type state with its capacity set in CVG market, other carriers were growing and started to expand significantly starting in 2013. Allegiant and Frontier have all set up large operations within the city and believe it or not, Allegiant is the second-largest O&D passenger carrier in CVG. In addition, Southwest launched service in 2017 and some legacy carriers have been adding flights to previously unserved hubs. Let’s also not forget that Amazon’s air operation has developed a significant logistics operation on the field with a significant expansion planned as well. After a decade of continuous cuts, in the past few years, it has been good to be the CVG airport. 



While the Cincinnati has lost 70% of its seats since 2004, the local O&D market has never been stronger. Cincinnati had roughly 10,000 one-way PDEW in 2019, up 61% versus 2004 when seats peaked (in the data currently available). This is due to how airlines are operating within the airport. Back in 2004, only 20% of the Delta passengers were originating or terminating their travel into the city. Today, however, Delta has minimal connections over the city.



Please allow me a minor soapbox moment. While the number of flights and seats have been greatly reduced, the recent story of CVG is actually a great one. The airport is serving the local market better than it ever has. Often airport executives or board members in general focus on seats and flights as their core metrics, rather than originating or terminating passengers (PDEWs). These passengers are customers living in or visiting the city/surrounding area which helps drive the local economy. Flow passengers, however, provide a limited amount of local contribution other than extra support jobs. Sustainable growth is often rooted in the local market. With relative ease, an airline can displace flow passengers between hub cities, however, an originating or terminating passenger can only originate or terminate at your airport (ignoring leakage).

When I was developing capacity sets, many times our capacity sets would produce fewer trips but more seats and O&D passengers, however, we would often be raked over the coals by consultants, board members, and local community officials even though our capacity set would actually produce a better economic outcome for the airline and airport. But I digress.

I would like to point out this was not the case with CVG. They were a great group of professionals to work with. Okay, soapbox moment over. Back to normal programming.

So, why all the interest in CVG? In 2017, United was one of the legacy carriers that expanded their service into CVG from one of their unserved hubs. However, this route appears to be short-lived. A few weeks ago, United announced they would discontinue their once-daily Cincinnati to San Francisco service.

The route cut piqued my interest. Why? A couple of reasons. First, CVG was an airport that I analyzed regularly during my airline days, so I continue to keep tabs on capacity moves in the market. Second, given the dynamic ULCC activity in CVG, I thought this might be one of the first time a ULCC pushed a legacy carrier out of a major hub market (this does not appear to be the case). CVG, however, is an incredibly complex city and CVG-SFO an interesting market.

In June 2017, United launched new daily service between CVG and SFO as part of a larger new market announcement. The new service directly competed with existing daily service on Delta and seasonal service on Frontier.


In response to United's announcement, Delta increased its less-than-daily service to near-daily. At the same time, Frontier reduced their Bay Area service from seasonal less-than-daily and summer daily to seasonal less-than daily to San Jose. I can only assume this was an attempt to differentiate their Bay service from others. After a year, Frontier returned their seasonal service back to SFO. Frontier has yet to extend their summer schedule, so it is yet to be determined if they return to the market in 2020.

For those surprised that Delta is still in the SFO market, writing the route off as a reduction to Delta's de-hubbing, Delta still has a significant nonstop network out of CVG. With over 68 average daily flights to 32 destinations in March 2020, Delta still has a significant operation in CVG with nearly three times the amount of seats of any other carrier. I am not going to lie, I completely forgot how large Delta is still within the CVG market.


Turning to United's route performance, year-ending 2Q2019 performance shows the route performed below system expectations. While both directions were soft, there appears to be a pretty significant divide in directional performance as well. The route as a whole appears to be performing 27% below system RASM in YE20192Q. While neither direction appears to hit system goals, overall performance appears to be hindered by the CVG-SFO directional flight.



Digging deeper into the schedule, the CVG-SFO directional flight would generally depart CVG in the later evening, roughly 7pm. This late CVG directional departure would put the flight into SFO around a little past 9pm. This connected the flight to a late evening bank in SFO largely to the west coast, backhauls, and limited long-haul international Pacific service. The return SFO-CVG flight would generally depart SFO between 10:00am and 11:00am local. Again, this would allow the flight to connect to the west coast, backhauls, and long-haul international Pacific service.


With these timings, the route offered few unique connections via SFO. Only HKG, SIN, and AKL would provide unique round trip connections not offered by other UA hubs. This means the route would only offer a unique nonstop between CVG-SFO as well as connection that complimented other hubs.

It should, however, be pointed out that the market timings were not consistent. When the market started, the route departed CVG in the evening, however, in the winter of 2019, the route was adjusted to a morning departure from CVG. Both of these departure times are consistent with market departure timings from IND and CMH to SFO. Indianapolis has a morning and evening departure to SFO. Columbus’ new service departs in the morning. When CVG-SFO was retimed, the route saw a significant increase in load factor. Throughout the first part of 2019, the load factor was up double digits. 


It is, however, difficult to tease out the impact of the market retiming. Generally, originating CVG passengers was a much larger population set than originating SFO. So a market retiming favoring CVG would likely increase route performance, assuming no significant changes to the flow timings. However, as the route was retimed, United’s fares declined 20-30% in the first and second quarters. This type of fare decline should significantly stimulate the market.

Digging deeper into the route makeup, I was interested to find that CVG, as well as nearby SFO markets, were very local. In fact, CVG-SFO local makeup was perfectly average compared to SFO to CLE, DTW, and IND. 



However, compared to other UA operated markets, CVG struggled with its segment fare. After CVG-SFO was retimed in 2019, it appears the United's revenue management team restructured the fares within the market in an effort to stimulate passengers in the market.


While I believe Frontier's nonstop service in the market had an impact on the fares during peak season, I believe the biggest impact towards United's route performance has to do with the originating makeup of the CVG to SFO. Of passengers originating and terminating between CVG and SFO, 60% of the market live in the CVG area.


This type of market makeup naturally favors Delta on the route. In the city of CVG, Delta still captures roughly 44% of all originating traffic. While down from 83% of the originating market in 2004, it still is a large lead versus any other carrier. This can naturally cause Delta to outperform in routes when the market favors CVG point-of-origin.

Finally, it is important to discuss the opportunity cost of the route. Yes, while the route did underperform with revenue production, it is also important to discuss the cost of the route in terms of aircraft time. On average, CVG-SFO cost the network 9.7 hours of total block time to operate the route. This equates to two average roundtrip markets on the United network. Even if the route was producing system-level returns, network planners would have to weigh the value of the route in terms of tactical importance versus flying two shorter haul routes. If we briefly ignore network and hub design, it could be possible for United to serve twice as many passengers in their embattled California hub with the CVG-SFO time. And let's not forget, United is having to harvest aircraft out of their schedule with the grounding of the MAX fleet. Cutting a longer haul, underperforming route is much easier than pulling out multiple shorter haul flights. 

While I am a little surprised United did not allow the retiming and what appears to be a new fare structure to fly a little longer, with all the considerations above, I can understand why they are exiting CVG-SFO. I would, however, not count them out of the market forever. While Delta is stable within CVG, if Delta were to exit CVG-SFO, I could see United quickly jumping back in. At least until then, United will be discontinuing the route in January 2020.


Wednesday, October 23, 2019

Alaska's Shifting California Love

Over the past few years, Alaska as attempted to make good on their desire to be the West Coast’s go-to airline. After spending $2.6B to purchase Virgin America, Alaska's route network received a significant shot in the arm, boosting their already growing California network.


While many remember the 2017 SJC and SAN investments, Alaska's interest in California has been growing for quite some time especially in San Deigo. Starting around 2010, Alaska started increasing its San Diego service, both in terms of destinations and flights. The first wave started with vacation destination gaps, Hawaii and Mexico, which was followed up with a wave of Intra-California flights to Fresno, Monterey, and Santa Rosa.


San Diego saw modest growth between 2013 and 2017 with seasonal Mammoth Lakes service launched in 2013 (discontinued in 2018) and Kona service launched in 2015. Finally, in the latter part of 2016 thru 2018, Alaska really refocused its growth in the market and launched 16 new destinations. This massive expansion accounted for 14% of Alaska's 2018 year-over-year ASM growth and 26% of their departure count growth.


Recently, Alaska announced new destinations were coming to San Diego (SBP and RDM) as well as depth to MCO, BOI, STS, SJC, and BOS. Within the same RDM announcement, LAX and SFO also saw a healthy dose of growth.


In the past, the combined AS/VX saw rapid growth. From 2010 to 2019, the combined company had an average ASM CAGR of 6.5%, however, after weaker performance, the company's growth hit the brakes in 2019 with an anemic 2% ASM growth. In 2020, growth is projected to increase, however, a meager 3-4%. Taken in a vacuum, Alaska's August 28th and September 4th announcements would consume roughly 50% of its 2020 growth rate.


However, Alaska implemented adjustments within its California network. In the first and second quarter, Alaska exited or seasonally reduced a significant amount of transcontinent and midcontinent routes. These markets that were exited starting in 2020 will free up roughly 1.7B ASMs compared to the recent announcements of 1.1B ASMs added to the network. The gap of roughly 0.6B ASMs will open the equivalent of another 1% of year-over-year ASM growth that has yet to be announced. Seasonal reductions would open another modest amount of ASMs which also could be reallocated.

Reds = exits; Blue = seasonal reductions

To understand why Alaska made these seasonal reductions and market exits, we again go back to the RASM curve. Examining Alaska’s RASM curve, most of the markets are below, some pretty significantly below the 1Q2019 RASM curve. This tends to point towards some performance issues with many of the routes which were reduced. This should not be too terribly surprising. Most of these routes were launched within the last 1-2 years. It often takes time for the market to produce system-level returns.


I really do not believe performance was the entire story here. Alaska's midcontinent expansion out of San Diego never really made sense to me. If you are going to be California's go-to airline, why would a carrier introduce routes that were not really that important to the originating California market? The markets that Alaska has removed from their schedule were largely markets with customers that fly to California. Rather than concentrating on a singular California city, for Alaska to successfully grow the exited markets, it would have required growing a customer base across seven different cities where Alaska was not necessarily a major player in the market.


Digging deeper into the San Diego market, Alaska has significant gaps within their San Diego portfolio. Of the eleven markets with over 300 passengers originating from SAN, Alaska only has service to five of the markets. Building a San Diego customer base would require Alaska to offer meaningful service other top originating markets to grow a base within the city.


While Alaska did withdraw from many non-originating markets, we can see them reallocating their service in other key markets (SJC, BOS, and MCO). To me, this does not point to a withdraw in the SAN market, rather a recalibration. If I was a betting individual, I suspect we will see a redeployment of additional E175s with medium frequency to many of the top short to medium-haul markets. If the aircraft and capacity are available, this type of deployment would increase relevancy in SAN while balancing new ASM exposure compared to longer-haul and larger-gauge markets.

If we are to see more redeployment within San Diego by Alaska, my best bet would be DEN, OAK, PHX, and maybe LAS, all top demand markets from SAN. These routes would likely be funded in part from the drawdown from the midcontinent and transcontinental that we saw announced within the last month. Further, I would suspect if these routes are going to operate within 2020, they would be announced within weeks to a couple of months to hit the ideal 2020 booking curve.

Given the recent moves by Alaska, it appears to me that Alaska is not withdrawing from their San Diego investment rather time will show their recent withdraws to be adjustments in a much longer-term strategy.

Wednesday, October 16, 2019

Why is JetBlue leaving Hobby?

In just a couple weeks, JetBlue will be packing its bags and leaving Houston Hobby (HOU) for Houston Intercontinental (IAH). The move, which was announced over the summer, may have come as a surprise to many. But this is not the first time a carrier picked up and left Hobby for Intercontinental. Remember, Frontier in 2012 leaving due to "customer feedback"? 

In JetBlue's press release they specifically state: 
"JetBlue regularly evaluates network performance, demand and customer feedback to strengthen its focus city strategy. The move to Bush Intercontinental is aimed at strengthening JetBlue’s relevance in New York and Boston, while also growing the carrier’s customer base in Houston where travelers love the airline’s award-winning service and competitive fares to and from the Northeast." 
Clearly, JetBlue was not happy with their performance at HOU and believe their Boston and New York customers would prefer IAH. So, we will evaluate JetBlue's statements regarding market performance as well as examining some of the Houston economic factors which likely drove the change.

It is important to understand JetBlue's history in Houston. JetBlue launched JFK to Houston service in 2006 with three daily A320 flights. Within a year, the A320 service was swapped for mostly E190 service. In the middle of 2008, the service was further downgraded from three flights to two. For the next five years, flights out of Houston remained largely unchanged outside modest fleet adjustments here and there.


As JetBlue started to build up its Boston franchise, Boston to Houston was launched in 2013. Initially, the route operated with two daily flights to Boston, however, lagging performance forced JetBlue to change the Houston set up. In the latter part of  2014, both New York and Boston were trimmed to one daily A320 flight in each market.

Taking a look at JetBlue's 2013 RASM curve, Houston to Boston service revenue production was significantly below JetBlue's RASM curve. JFK appears to be inline with system unit production. However, for both of these routes, it is important to remember that being below a RASM curve does not guarantee a route is losing money nor a route on the RASM curve to be making money. The greater the deviation from the RASM curve, the high the probability a route is over or under-performing to expectations. Given the large Boston deviation, Boston appears to be significantly underperforming.


For the Boston to Houston to close the gap between the RASM curve and our estimated RASM, the route would have needed to produce an additional $5,000 per flight. I want to reemphasize what this gap actually means. A gap above or below the RASM curve does not mean a carrier in making or losing money. Rather, it shows a perceived revenue opportunity cost of not operating a flight producing system-level revenue results.

While the performance on the route has significantly improved since the route was launched, the revenue performance still appears to be below where a route operating at this distance should be. Further, the route has largely remained stagnant since 2015 in improving the unit revenue.


One of the largest drivers in underperformance on JetBlue's BOS-HOU was its load factor. The carrier started the route incredibly soft, filling just over 50% of their seats. This significantly lagged its competitor's Southwest and United. Even with downgauging aircraft and cutting flights from two to one, the route still lags competitors' load factor to this day.


We can break down each carriers' load factor. In doing so, we quickly see that JetBlue's load factor softness is due to network design, not lack of demand. In fact, JetBlue outperforms other carriers' local load factor (load factor contributed only to passengers flying between Boston and Houston as their origin and final destination). This outperformance has only increased since the route was launched in 2013. However, without a network in Houston nor Boston for passengers to connect, JetBlue cannot flow additional passengers on the BOS-HOU leg.


With revenue gains stagnant and an inability to flow passengers via BOS-HOU, why look north? Hint: It is due to the local Houston economics. The IRS and Census Bureau provides a significant amount of data to understand and target areas of wealth in Houston.

Below, we pulled the 2016 IRS 1040 tax filings by zip code. Within the data, we can see how many tax returns were filed, the number of dependents, and a range of income metrics. Below, we mapped this data to show the population size and the average adjusted gross income by zip code. The two black dots are the location of Hobby (south side) and Intercontinental (northside).


Quickly, we notice the larger and wealthier population clusters are on the north and west side of Houston. While there are smaller, wealthier zip codes closer to downtown, Hobby is not necessarily that much more convenient. According to Google Maps, a traveler departing early in the morning from the west side of downtown (zip code 77027) would experience a 25-minute ride to Hobby vs 30-minute ride to Intercontinental. Mileage is incredibly deceiving. The access and egress from Hobby is largely city streets compared to the highway infrastructure around Intercontinental.

Further, Hobby does not have convenient access to some of the largest headquarters within the area. We were able to find a collection of mapped Fortune 1000 headquarters. When mapped the data shows the southside of Houston is largely a Fortune 1000 headquarter desert. Intercontinental, however, does have access to nearby headquarters as well as equally competitive access to headquarters located downtown and on the Houston west side.


It appears based on the information above, JetBlue really only had a few choices. They could stay with the status quo and keep a subpar route, remove the seventeenth-largest Boston demanded market from their network, or move airports and hope for better performance. With JetBlue's ambition to grow Boston to over 200 daily departures, removing one of the top O&D markets is really not an ideal plan. Neither is keeping the status quo on an underperforming market. So JetBlue is doing their next best option and moving to Intercontinental on October 27th.

Do you believe I missed something or think I am incorrect? Let's have the discussion below! You can also suggest future topics for me to review.

Wednesday, October 9, 2019

Mistakes were made: How I failed to understand the basics

After getting the call that US Airways was withdrawing from Colorado Springs, I went into an immediate reaction to help prepare my leadership with the best possible explanation for US Airways' decision. I started to pull what data I knew. It was inevitable the press would be calling the following morning. Following my (flawed) analysis, we took the stance that the route failure was likely due to the cost inefficiencies of the 50 seater aircraft in the market. If US Airways had operated the right aircraft within the market, the route would have continued to be successful. (Press coverage)

Like most US airports at the time, we heard the 50 seaters were dead men walking and airlines upgauging to drive unit costs lower was the way of the future. Based on my analysis, our revenue and load factor performance were on par with other US Airways flights. It was only logical to believe the route failure was based solely on aircraft costs.

In reality, route level trip cost modeling based on public data is incredibly complex, unlike ticket revenues prorates. Ticket revenue prorate methodology generally follows either the square root of mileage or IATA predetermined prorates. These two methodologies deliver result in very similar results. Since IATA prorates are not public, revenue modeling used by most large data suppliers, including my models, typically follow the square root of mileage methodology.

For trip cost modeling, there is no industry standard for an airline to prorate indirect costs (aircraft ownership, maintenance, headquarters overhead, etc). Each airline's finance department determines their individual methodology. Without an industry-standard cost model, any cost model based on public data pushed to the route level will have material gaps. So much so, I would argue such models should not be used when analyzing route level performance nor presenting business cases to airlines. More on that another time.

Now, it's cringe time. When joining the industry ten years ago, I believed that the airline profitability formula was simple RASM - CASM = Profitability. The term stage length adjustments or understanding the decaying relationship between RASM and distance never crossed my mind. In my mind, all RASM was equal. No adjustments needed.

After US Airways informed us they were leaving, I produced a chart similar to the below. The below chart attempts to show COS-PHX RASM was significantly above US Airways mainline CASM. Therefore, we assumed if US Airways would just operate mainline aircraft, COS would be profitable, hence the statement "... the carrier could be losing money in the Springs because it was flying 50-seat regional jets on the route that are more expensive to operate than larger aircraft." If you didn't realize it from the graph, this is wrong. 



Further analysis backed up my hypothesis. According to DOT data, COS station outperformed the US Airways network across three core metrics we were tracking: RASM, load factor, and passengers flying nonstop on the route. Segment fares were soft but with the threeway battle in Denver, it was not surprising.



If you are new analyzing airlines, the presentations above likely did not throw red flags. At best, the RASM chart shows the bell shape curve that skews right towards higher performance. So, what is wrong? Well, it's all in the curves.

One of the first steps airport professionals do when analyzing route performance is graph a scatter plot of RASM vs distance. US Airways' COS-PHX route was right in line with the RASM curve for the US Airways' network (note: different data providers may show RASM differences). RASM curves demonstrate what an airline typically demands it's unit revenue performance over a unit of distance to be.



US Airways' 2008 RASM curve looks similar to a typical RASM curve. The shorter the stage length, the higher the RASM. Because of this, taking a simple market-level RASM and measuring to see if it is above or below the system CASM really was meaningless. Further, the bell-shaped curve that I produced for my leadership was simply a system representation of US Airways' network build rather than performance. 

Digging deeper into the performance of COS-PHX, on the surface, the route may not have caused significant concerns. The 2008 RASM appeared to be inline with system averages. However, it is important to trend the route performance over time and consider the entire environment in which the route was operating. However, it is important to note, while we do not know exactly what the route's CASM was, generally the smaller the aircraft the higher the CASM. While Colorado Springs was operating in line with system stage length RASM, however, it was produced on some of US Airways' highest CASM producing aircraft. 

Taking a look at the route's history, in the mid-2000s, the route lagged significantly in load factor (65%) with a $122 average segment fare. The US Airways' Revenue Management team clearly was not happy with this performance. Segment fares appear to be sacrificed drive up leg load factors throughout 2006 - 2007. At the same time, the Network Planning team decreased the gauge of the aircraft. All of these actions did increase load factor, but it did not actually stimulate additional passengers until 2007 when the segment fares were down 20% vs 2005. 


Further, RASM really did not respond until 2007 and stayed flat in 2008 as the team tried to raise fares back to historical levels, likely in response to significant fuel cost pressures. However, as the fares increased, demand would drop off.

In the final year of service, US Airways seemed to be heading back to decreased performance. RASM was near a five-year low, segment fares were there to match, and US was operating some of their smallest (and highest CASM) fleet within the market. At the same time US Airways was also working on upgauging and removing smaller aircraft markets from its Phoenix network. In 2008, US Airways discontinued EUG, CLD, ASE, and OKC, all of which were operated on aircraft similar to their COS-PHX route. 

Clearly, US Airways was having significant issues with fares and passengers in the market, but why? In 2006, Southwest entered Denver with an aggressive growth and fare strategy that pressured the entire Colorado to Phoenix market. Before Southwest's arrival into Denver, Denver to Phoenix fares averaged $120, however, one year later, the fares were averaging below $80. Colorado Springs fares did follow with a premium. 


US Airways did not only see pressure within the local Phoenix market but their California via Phoenix, which was the top flow markets, also saw pressure. As the intensity of the competition in Denver increased, fares Denver to California began to decline. Over the course of five years, fares Denver to California were down 25%. At the same time, fares within the Colorado Springs to California markets were relatively flat, however, passengers declined as much as 20% as passengers decided to forgo Colorado Springs for Denver's fares and service offerings. 


US Airways clearly faced a no-win situation. The growing Denver competition placed significant pressures on Colorado Springs. Any attempts by the US team to increase fares resulted in losing passengers to Denver competitive carriers. The Network Planning team was already operating some their smallest Phoenix aircraft into the market, so the ability to downgauge further was off the table. Finally, other similar-sized markets were being removed from the Phoenix network. Really, US Airways had only one choice, remove Colorado Springs from their network. And that's exactly what they did.