Biography for William Swan Chief Economist Seabury-Airline Planning

Biography for William Swan Chief Economist, Seabury-Airline Planning Group. AGIFORS Senior Fellow. ATRG Senior Fellow. Retired Chief Economist for Boeing Commercial Aircraft 1996 -2005 Previous to Boeing, worked at American Airlines in Operations Research and Strategic Planning and United Airlines in Research and Development. Areas of work included Yield Management, Fleet Planning, Aircraft Routing, and Crew Scheduling. Also worked for Hull Trading, a major market maker in stock index options, and on the staff at MIT’s Flight Transportation Lab. Education: Master’s, Engineer’s Degree, and Ph. D. at MIT. Bachelor of Science in Aeronautical Engineering at Princeton. Likes dogs and dark beer. (bill. swan@cyberswans. com) © Scott Adams

Value of a Nonstop William Swan Chief Economist Boeing Commercial Airplanes Marketing April 2004

Five Viewpoints 1. 2. 3. 4. 5. Value --- What is it worth? Price --- What does the Market Pay? Demand --- Does it increase travel? Market Share --- Does the airline benefit? Cost --- Is there extra cost?

Nonstop Value • “Everyone knows” a nonstop is “worth more” – More than “standard service” – “Standard Service” is an online connect • Less travel time – About 1. 8 hours lower travel time • 1 less “up-and-down” cycle (0. 5 hours) • Zero connect time (1. 0 hours) • No circuity (0. 3 hours) – Business value = hourly wage + overhead ? – Leisure value = hourly wage – taxes ? • Less travel worry – – No missed connections Fewer lost bags Unbroken onboard rest One less set of take-off and safety announcements

Utility Value Contained in Soars • Utility value for a trip includes – – Elapsed time Price Suitability of the time of day Nuisance of connecting process • Market share model has Utility Function – Determines flight choices based on values • Utility function expresses consumer surplus – This is precisely the value beyond price paid

Distinguish between Value and Payment Marketplace economics: – – Create Value Get people to pay for it Part of value appears as “consumer surplus” – Surplus is value beyond fare paid Nonstop value looks like about 2 hours of time – – – Market failure to “collect” is not proof Value is still created But marginal value of time might lower

Operative Statement of Value • Nonstop service is “perceptibly superior” – – Compared to online connection Dominates at least a 4 -hour window of time-of-day At least “equal value” for adjacent hours Not superior at wrong time of day • Perceptibly superior criterion is important – Makes nonstop first choice for demand • First dibs on high-fare traffic in peak days • First dibs on high load factors in off-peak days • Based on matching fares

Important Implicit Assumption • Airline tactics for “superior” (nonstop) product – Match fares – Garner all high-fare demand – Garner high load factor in off-peaks • Greater revenues from this – Compared to charging higher fares – Compared to giving up “first choice” advantage • Intermediate case is tempting – Charge slightly higher fares, particularly in peaks – Not so high as to give up “first choice” advantage – Difficult balance to achieve

Perverse Data Possible • Data will show – Majority of local O&D traffic will be on nonstop • 40 -75% of market, for one-a-day • Nonstop tickets dominate average fare – Perverse outcomes for local O&D traffic • First choice for all low-fare traffic (flexible departure times) • First choice for part of high-fare traffic (partial-day time windows) • Nonstop mix could be heavy on low-fare • Scheduling incentives also increase discount loads – Local O&D discount revenues good, even discounted – Relative to pro-rated connecting traffic using this leg – Lower costs increase willingness to accept low fares • Save connecting up-and-down cycle airplane costs • Save cost of connecting passengers and bags

Summary of Value • Nonstop creates value for sure – Roughly 2 hours trip time saved • Airline unlikely to raise prices – Gets higher revenues from being “first choice” • Dominate high-fare demand for part of day • Obtain high load factor on off-peak days • Average local fare may be low – Dominates low-fare market for all of day – Lower cost increases willingness to carry lower fares – Local revenues not diluted with pro-rate to second leg

Nonstop Prices • Simple looks show no fare advantage • Do we believe adjusted market trends? • How about side-by-side in same market?

Simple Look Shows No Fare Advantage for Nonstops • Regression fare formula: – Fare = $166 + $0. 046 * Distance – Trend for US domestic fares – US ticket price data by O&D available • Dummy for nonstops is no help – Fare = $169 + $0. 045 * Distance - $14 * NON – Where NON=1 if served nonstop; 0 otherwise – Wrong sign, poor significance, for NON

Nonstop Markets Show Lower Fare Trend Before Adjusting for Market Size

Deeper Analysis Shows Nonstop Value is Captured • Fares decline in larger O&D markets – – Fare = $191+0. 048*Distance-11*Ln(Pax) term means lower fares for big markets Statistical Significance of Ln(Pax) term is big Reverse causality (low fares cause high Pax) is not the explanation—because Log form attenuates • Full form with nonstops: – Fare = $197+0. 051*Distance-16*Ln(Pax)+$36*NON – $36 premium for nonstop in a market – Shows value is at least partially captured

Bigger Markets Have Lower Fares

Viewing Just One Market Size Nonstop Fares are Higher

Nonstop Markets Trend $36 Higher, After Adjusting for Market Size

$36 Premium for Nonstops • • In comparable-sized markets, US domestic Shows some value is captured in prices $36 for average fare Lower ($14) for discounts (25%ile fare) Values coherent with value of time This for US domestic O&D airport pairs Same market fares confirm value

Nonstop Premium is $20 -$50 Nonstop vs. Connect in Same US O-D Market

$79 Premium for Atlantic Markets • • Premium higher on long trips Time savings also higher Coherent with utility values Data not as certain – Less consistent, year to year – Fewer points, lower quality on market size – Pacific markets even worse

Atlantic Nonstop Premium is $75 Markets Served Nonstop vs. Connecting Markets

Atlantic Premium is $75 -$125 Fares Nonstop and Connecting, Same O-D Market

Before-and-After Studies Confuse • Look at relative changes – Markets served nonstop 1995 and 2000 (2 q) – Markets gaining nonstop services by 2000 • • Revenue growth is 23% higher Passengers growth is 50% higher Fares are 19% lower!! These values without size adjustments

Recent Nonstops Added by Low-Fare Carriers • • Comparing size-adjusted fare trends Markets gaining nonstop service by 2000 Compared to markets already nonstop (in 1995) Fare trend lower in markets about to be added Fare with new nonstop barely changed Fare in older nonstop markets up 10% REVENUEs are still up 25% Different expression of demand curve

Conclusions about Prices • Simple look says no nonstop premium – Fare trend for nonstop markets not higher • Correcting for market size “fixes” paradox – Strong trend of lower fares in larger markets – For a size, nonstops get higher fare • Results confirmed – By reasonable values of time – Nonstop carrier gets premium in same market – But not by latest LCC before-and-after studies • But recent low-fare carriers are giving value – Nonstop additions creating more revenue, not higher fares

Demand Increases With Nonstop • Nonstop service has higher value • Some of value may be captured in price • Remainder of value seen as demand increase – Better deal means higher demand • Increased value means higher demand curve • New demand point – Different number of passengers – Different fare – Lies on higher demand curve

Demand Is A Curve Not a Point

Adding Nonstop Increases Demand About 25% • • • Before and after studies “Demand” curve measured as total revenue Most of increase is more passengers Fare rise can be small 25% is above nonstop fare differentials This is US domestic travel data

Nonstop Market Share • What is “Market” – Demand curve with current nonstop service – Old Demand curve plus one nonstop service • What is “Share” – – Reported share of O&D passengers Share of market total revenue Share of high-fare (business) passengers Share of low-fare (leisure) passengers • What is “Nonstop” – One flight, with standard fare premium? – Flight with all fares matched and capacity open – Is observed average fare premium a mix or price result?