Ashes, Ashes, We All Fall Down

IATA released its international traffic results a couple of weeks back and after a very positive rebound in international passenger traffic over the last nine months, April's results showed a considerable decline thanks in large part to Volcanic Ash that affected traffic across the North Atlantic.  Overall, passenger traffic was down by 2.4% for April vs. April of a year ago.  More importantly, the results were down from the 10% increase recorded in March.

European traffic suffered the most, declining by almost 12% while North American traffic declined by 2%, Asia-Pacific showed a 3.5% increase, Middle East airlines recorded a 13% increase, and Latin American carriers showed a 1% increase.

The 'ash' was one of those unexpected external elements that tend to damage traffic trends.  Moreover, other external events can damage the rebound as strikes at British Airways and currency instability may further impact traffic returns putting into jeopardy a nice upturn.

Separately, IATA is now forecasting that the world's airlines will post a profit of $2.5Billion in 2010.  It was only March where the same organization said that the airlines would lose $2.8Billion.  This apparent 'miss' (hey...what's a $5.3Billion error between friends) reminds of the expression: "...people shouldn't forecast, especially about the future...."

The Parking Lot is Full

Stored aircraft is a key metric when assessing the health of the commercial airline industry.  The number of aircraft that are parked (I use parked and stored interchangeably) are related to seat capacity changes, aircraft values, and airline aircraft fleet plans.  The nearby time series graph shows the peaks and troughs of parked aircraft between 1990 and June 2010 and are divided by aircraft class groupings--RJ (regional jets, both modern and older RJ types), NB (single aisle aircraft), and WB (twin aisle aircraft).  The data is from Ascend's aircraft database.  Note that Ascend's definition of stored aircraft is for aircraft that have been reported as parked for at least 30-days.  Consequently, aircraft that are down for heavy maintenance checks may be included in the data but are likely to be a small sub-set of the totals.  Also, these are absolute unit numbers and don't show the relative nature of the parked inventory, i.e., number of parked aircraft vs. the number of aircraft in the total inventory.  I will get into that in another post.  Finally, there are numerous elements that come into play that affect parked airplanes.  This analysis is basically a high level assessment.

In the graph, I highlighted the large events that have created turmoil in the stored aircraft market and you can obviously see the rather steep effects that these events have had on the parked inventory.  The single aisle aircraft class shows the most volatility, growing by more than 100% in the Gulf War of 1991 with another two-fold increase as a result of 9/11. Parked single aisle aircraft during the current recession rose from about 800 in 2007 to more than 1200 by 2009.  From the trend line data, it is also apparent that while the parked inventory can increase quickly (almost overnight it seems) it takes years to reconcile itself during better economic times.

Twin aisle aircraft and regional jets show a different dynamic with parked wide-body aircraft rising consistently between 1990 and 2003, but then falling for for about four years before rising slightly during the current recession. Parked RJs have been increasing for most of the decade and are now near 500 units--although many of those are older RJs (like BAe 146s and Avros).  Of the modern RJs (like the Embraer 145 and CRJ 100/200), about 250 are parked (the Ascend data indicates that 44% of the ERJ135s in the total inventory are parked---that clearly is a function of today's fuel prices).  From the data, it is clear that narrow-body aircraft are more susceptible to economic downturns than either WBs or RJs at least on an absolute basis (of course, there are many more NBs in inventory than either WBs or RJs---which makes the case for looking at this data on a relative basis a necessity in order to analyze these trends fully).

A final observation.  Parked aircraft tend to follow physical age and technology age: 18% of 747 aircraft are currently stored, while less than 1% and less than 3% of 777 and A330 aircraft respectively are parked.  Less than 1% of 737NGs are parked while more than 50% of JT8D powered 737s are parked and are not likely to return to service.

Wither Regional Aircraft Deliveries?

Regional jet aircraft deliveries have continued their year over year decline from their peak that occurred in 2001. RJ deliveries in 2010 will likely be less than 125 units, the lowest since 1997.  50-seat RJ deliveries distorted the market and were simply a result of pilot scope restrictions which artificially drove the entire market.  Long gone are those days and they are not likely to return.  Today, larger and more efficient RJs are being delivered but you can't say that they are delivering in large numbers.  Scope provisions are also driving--and restricting--these deliveries. 

The larger RJ aircraft have considerably better unit costs but even at 70 to 90 seats, their unit cost performance may not be sufficient in a world where yields are being driven down by low-cost providers.  There clearly is a need for larger RJs, but just how many is debatable--as is the number of manufactures who are producing these types.  The sweet spot for aircraft size is in the area of 120 to 160 seat aircraft.  These types tend to do well in hub flow networks and in a low-cost carrier networks.  No surprise then that Bombardier launched the CSeries and it is my belief that Embraer will follow.  The market is too fertile for mid-size single aisle aircraft to pass up particularly when the market for 70-90 seat aircraft is---and keeping with my garden metaphor---getting a little bit dry.

Wither regional airlines and aircraft?  I know, I know...this is blasphemous to some, but consider: today's RJ backlog for Embraer and Bombardier is down 35% over levels at the beginning of 2005; today's backlog for Airbus and Boeing narrow body aircraft is up 131% over that same period.

Size Matters

With the Regional Airline Association (RAA) annual convention happening next week, I thought I would take this opportunity to provide a small set of analysis on what has been occurring over the last decade as to the trends of aircraft size and the trend of how far passengers travel on regional size aircraft.

The nearby chart (click on it to enlarge) shows an x/y scatter plot for average passenger journey vs. the average number of seats per aircraft over the last ten years.  This data is taken from the RAA website.  I simply did some clever analysis of the data and laid it out in the chart (the latest data available on the RAA website is through 2008; I suspect they will be releasing 2009 shortly).  As you can see, there has been a steady upward progression in both the average passenger journey and the average number of seats on regional aircraft.  The dots represent the intersection of both of the data elements and I labeled them by the corresponding year.  I then plotted a trend line through the points to provide a perspective as to where the data would track if history is any indication of the future.

There are a couple takeaways from the chart:  while the upward progression is apparent, there is some 'bunching' of the data for the years 2005 through 2008.  In my opinion this suggests that regionals are now at close to their limits in terms of the metrics on the chart with both of these dimensions being a function of aircraft size limits as determined by pilot scope agreements, i.e., the regionals have bumped up against numbers of regional aircraft permitted and the size of those aircraft which in turn limits the average passenger journey.  This is rather important to the other takeaway from the chart: the forecast trend line will only be realized if there is an open market not limited by scope or at the least, there is additional liberalization of aircraft size that regionals can deploy.

This is a major issue that the regional airline industry can't control but significantly effects their growth prospects going forward. Will there be scope relief in the future that can fulfill the forecast trend line?  How will that manifest itself and how will it occur? 

There are some hints to this issue out there now, but that will have to wait for another time. 

Biz Jets

As many of you know, the business jet market is a pretty exciting space.  At the end of 2009, the number of in-service business jet aircraft was 60% higher than at the end of 2000 (according to Airclaims, there were more than 16,000 business jet aircraft in-service at the end of 2009 compared to about 10,000 in 2000).  Contrast those figures with commercial jet airliners which had an increase of about 30% (for in-service aircraft) over that same period (19,600 aircraft in 2009 vs. 14,800 in 2000).   Thus, it is an important segment to watch and with this post, I will--from time to time--comment on the business jet sector.

GAMA (General Aviation Manufacturers Association) 1st quarter 2010 shipment report was published recently.  The report indicated 164 business jet deliveries for the March quarter, which was down 14% from the first quarter of last year.  Business jet shipments can be quite volatile as they tend to track changes in stock market indices like the Wilshire 5000 and the S&P 500 (I will produce that relationship in a subsequent post).  There are six sizable competitors in the business jet sector--Bombardier, Cessna, Dassault, Embraer, Gulfstream, and Hawker/Beechcraft.  Also, Boeing and Airbus are players in this sector but generally deliver a small amount of business aircraft which are based on their narrow body commercial aircraft products.

Of note in the recent GAMA shipment report was the great disparity by manufacturer in shipments from 2010 compared to the same period in 2009.  For example, Cessna had a whopping decline of 55% between the two quarters, while Gulfstream went down only 10%, Bombardier down 13%, and Dassault was actually up by 6 aircraft units (a 55% increase overall). Cessna is the largest business jet producer (in terms of units) with about 30% of the total for all biz jet manufacturers. Thus, it's performance has a heavy weight on industry totals and percentage changes.  The Cessna jet airplane models that showed the largest declines (1Q10 vs. 1Q09) were the Mustang with eight less units being shipped, the CJ3 also declined by eight units, and the XLS model(s) had a reduction of 12 units.

Also of note, were the deliveries of large aircraft vs. smaller aircraft.  Large aircraft like the Gulfstream 350/450/500/550, the Bombardier Global 5000/Express, and the Falcon 7X all did fairly well despite a down quarter.  All together, there were about 60 large aircraft shipments for the quarter which was a 26% increase over the 2009 1st quarter.  Generally, large aircraft shipments tend to be around 25% of total shipments, but in the 1Q2010, large aircraft were about 35% of the total.  This buoyed the results for Dassault and Gulfstream (plus to a lesser extent Boeing and Airbus).  The large aircraft shipments may be a result of an 'overhang' that existed after the larger aircraft were ordered when economic times were better and are just now being shipped.  Nevertheless, it is a trend worth watching.

Finally, very light jets, i.e., the Cessna Mustang and Embraer's Phenom 100 have had a pretty good run over the last year.  Please note that some in the industry may not consider the Mustang and Phenom 100 as very light jets, but I do (and it's my blog).  In any event, these two aircraft have now shipped over 200 units in the year ending with 1Q2010.  Not bad for a newly formed sector within general aviation and more specifically, business jets.  So, both ends of the business jet landscape, i.e., the very small and the very large appear to be doing well.  Keep watching.

More on the CO and UA Merger

I am revisiting the United and Continental merger analysis to provide an additional assessment of the market strengths of the merger potential.   The first chart illustrates the market share for U.S. airlines by operating division.  Note that this analysis is only for the seven largest U.S. airlines and I have added together CO and UA.  The data is for the year ending with the 3Qtr 2009 and is taken from the airline's D.o.T. filings.

The United and Continental market shares are highlighted to demonstrate their market clout by operating region.  As such, CO+UA had a 25% share of domestic passenger traffic (again this is for just seven airlines and doesn't include JetBlue, AirTran, Frontier, Spirit, regional airlines, etc, so that share is lower than indicated here); 35% of the Atlantic region, 27% of the Latin American region, and 52% of the Pacific region.  The shares for the international sectors do not include foreign carriers.  Of note, Delta (including NW) also has significant share standing with 23% domestic, 39% Atlantic (the highest share in that region), 27% for Latin America, and 38% for the Pacific.  The other so-called Legacy carriers--American and US Airways--have significantly lower shares particularly on the international traffic sectors which may, over the long term, cause competitive disadvantages for AA and US.  Indeed, in the Pacific region, Delta and UA+CO captured 90% of passenger traffic.

The second chart is really the 'money shot'.  (Before I get into it, I should mention for those that don't know me that I have a reputation for creating 'busy' charts, i.e., charts with lots of data and words on them.  The second chart is in that category; but stick with it as it is quite meaningful and in the end will be easy to understand, except maybe for those of you in Colorado....just kidding). 

Anyway, the chart indicates growth RPKs (revenue passenger kilometers) on the horizontal axis for the 20-year period between 2008 and 2028 for major passenger traffic flows as taken from Boeing's 2009 Current Market Outlook.  The market share for the CO+UA share of passenger traffic for the year ended 3Q2009 is on the vertical axis.  The size of the circle represents the relative size of the growth RPKs for each of the major regions.  The RPKs are for traffic flows in both directions, i.e., North America to/from Asia, etc.  Note, North America is used as a proxy for  the US as Boeing does not break-out US data independently (does Canada matter?).  

From the Boeing data, I analyzed the traffic growth flows for the major regions for North American traffic.  Thus, traffic forecasts are given for North America to/from Europe, intra North America, North America to/from Asia (including China), and North America to/from South and Central America. Those regions are placed on the chart with the center of each circle being the intersection of CO+UA's current market share with the size of  forecast growth.  To give you a sense of proportion, intra-North America traffic (according to Boeing) totaled 976 billion RPKs in 2008 and will grow to 1.6 trillion RPKs in 2028, an increase of about 600 billion RPKs (the size of the circle on the chart).  This growth amounts to about 38% over 20 years which averages to just under 2% growth per year.  Note that all airlines who can deploy capacity in these areas will be vying for a share of the market RPKs for the respective regions as indicated on the graph. 

So how is the CO+UA merger positioned in each of these growth areas?  As shown in the first chart, their shares in these important regions is quite competitive or even better.  And in the context of the second chart, their current market share position bodes well for the capture of the growth RPKs over the long term.  Their weakest market share is in the North America to/from South/Central America region.  But those traffic flows have the lowest growth prospects of any of the traffic totals indicated on the chart.

From this analysis, it is clear that more consolidation in the U.S. will be inevitable.  Both the Delta Northwest merger and now the Continental United merger raises the strategic stakes.  Others, particularly American and US Airways, in their current forms, will be considerably weaker when they compete for future market shares in the growth markets against the merged entities.  

What Goes Up (Blood, Sweat & Tears, circa 1967)

The passenger and yield picture for the US major airlines in recent months is quite the roller coaster. In the span of just a few months, passenger yield (system wide yield as measured by revenues divided by total revenue passenger miles, which computes to cents per mile) moved from gains of over 10% (by month, year over year) to a decrease of 20%. Similarly, passenger traffic (domestic and international) moved from fairly good gains in early 2008 to declines of more than 10% in less than six months. The combination of these two trends made for a devastating decline in total passenger revenues.

Today, as with the IATA traffic data reported last week, a rebound appears to be well underway as we see continuing strength in both yield and traffic for the US airlines.

A noteworthy element in the passenger traffic and yield trend is that passenger traffic tends to have a fairly long decline (or increase) taking the better part of a year to move from trough to peak and vice-versa (at least in this data set which is for a little more than four years). On the other hand, yield has a much sharper decline (or increase) taking less than a year to move from trough to peak to trough. This stands to reason as airlines can't react quickly enough to decreases or increases in traffic relative to their passenger fare yield. As indicated in 2008, it took the airlines about eight months to lower fares despite deteriorating traffic conditions for pretty much that entire period. Likewise, it took about six months for the airlines to react to bettering traffic conditions before they could attain better yields. Of course, hindsight is always perfect vision and yield management for an airline is more of an art than a science.


The bottom line for the dynamics of passenger traffic and fare yields is how it changes total revenues. As the second chart indicates (click on the chart to enlarge), 2006 and 2007 and well into 2008 were pretty good boom years with revenue growth averaging 7% across the months. Even though the recession started officially in December of 2007, total revenue still grew because yields were growing despite traffic falling. But a perfect storm ultimately occurred as the airlines encountered the depths of the recession with a combination of poor traffic and poor yields. This produced a very impressive reduction in total passenger revenues. In some months, revenue production declined by 25% from a year earlier. For the entire period of 2009, total revenues were down about 20% relative to 2008. This amounted to $16B of lost revenue spread among a handful of airlines. That's real money. The airline industry is indeed a penny driven business--a small fractional change in passenger yield can create huge swings in revenues and the fortunes of the airlines in the process.