Effects of Airline Industry Changes on Small- and Non-Hub Airports (2015)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

35 C H A P T E R 3 3.1 Population The study population includes airports classified as small- or non-hub airports at any time dur- ing the study period (2001 to 2013). Only airports in the continental United States were included in the study population. Airports in Alaska, Hawaii, and U.S. territories were excluded because they have unique qualities that make comparisons to airports in the continental United States difficult. The Bureau of Transportation Statistics (BTS), a division of the U.S. DOT, publishes a list of air traffic hubs annually. The classifications are based on an airport’s share of all enplaned (boarded) passengers in the United States. Small-hub airports enplane between 0.05% and 0.24% of all enplaned passengers. In 2013, small-hub airports enplaned approximately 1,000 to 5,000 passengers per day. Non-hub airports enplane more than 10,000 passengers per year, but less than 0.05% of all enplaned passengers. In 2013, non-hub airports enplaned approximately 27 to 1,013 passengers per day. Given that airport classifications are published each year, an airport in the study population might have been classified as a small- or non-hub airport for only 1 year of the study period, all of the years of the study period, or some number of years between these extremes. In any case, all data for each airport were collected for every year of the study period. Exhibit 3-1 shows the number of small- and non-hub airports in the continental United States from 2001 through 2013. All of the small-hub airports received scheduled service in each year. The number of small-hub airports ranged from a low of 60 in 2002 and 2003 to a high of 69 in 2012. A few non-hub airports did not receive any scheduled service in each year. The number of non-hub airports with scheduled service ranged from a low of 193 in 2009 to a high of 224 in 2001. The combined number of small- and non-hub airports ranged from a low of 262 in 2003 and 2009 to a high of 290 in 2001. For reference, small- and non-hub airports in the continental United States in 2013 were catego- rized by FAA region, as shown in Exhibit 3-2. The number of small-hub airports ranged from a low of 3 in the Central and New England Regions to a high of 19 in the Southern Region. The number of non-hub airports ranged from a low of 12 in the Central and New England Regions to a high of 48 in the Great Lakes Region. The combined number of small- and non-hub airports ranged from a low of 15 in the Central and New England Regions to a high of 56 in the Great Lakes Region. 3.2 Data Elements A database of the airports in the study population containing a number of data elements was created, including information on scheduled commercial service, community demographics, market data, and airport financial data. Data Analysis, Airline Industry Changes, and Case Study Selection

36 Effects of Airline Industry Changes on Small- and Non-Hub Airports Not all data elements were available for each year of the study period for each airport in the study population. For instance, some airports did not have scheduled commercial service in some years of the study period. Similarly, not all airports reported financial data for all years. The database contains all data available when this research was being conducted. A spreadsheet containing many of the airport-specific and year-specific data items discussed in the following sections is available for download from the TRB website by searching for ACRP Report 142. This file also includes the QSI metric discussed separately in Chapter 8. Exhibit 3-1. Number of small- and non-hub airports, 2001–2013. Exhibit 3-2. Number of small- and non-hub airports by FAA region, 2013.

Data Analysis, Airline Industry Changes, and Case Study Selection 37 3.2.1 Scheduled Commercial Service Scheduled commercial service data were retrieved from the OAG. The OAG contains flight schedule information for all scheduled commercial service. Data for each airport in the study were retrieved for October of every year in the study period. The month of October was selected because it is considered a “shoulder” month for airline traffic; on average, it is neither the busiest nor the slowest month of traffic for U.S. airlines. 3.2.1.1 Metrics The number of departures, seats, and available seat miles (ASMs) were the metrics selected for inclusion in the database from the OAG scheduled commercial service data. Exhibit 3-3 shows the average daily flight departures per small- and non-hub airport during the research period. The number of average daily flights per small-hub airport declined from 48 in 2001 to 37 in 2013. The number of average daily flights per non-hub airport declined from 11 in 2001 to 7 in 2013. Despite the substantial decline in flights at small-hub and non-hub airports over the research period, the number of seat departures did not decline as substantially because the average num- ber of seats per aircraft (hereafter referred to as average aircraft seat size) increased, as shown in Exhibit 3-4. From 2001 to 2013, the average aircraft seat size at small-hub airports increased from 73 to 82 seats and the average aircraft seat size at non-hub airports increased from 40 to 52 seats. Changes in aircraft size serving small-hub and non-hub airports are analyzed in more detail later in this section. Exhibit 3-5 shows the average daily seat departures per small- and non-hub airport during the research period. The number of average daily seat departures per small-hub airport declined from 3,537 in 2001 to 3,022 in 2013. The number of average daily seat departures per non-hub airport declined from 446 in 2001 to 383 in 2013. These declines were less dramatic than the declines in average daily flights per airport. To explore the effect of seasonality on small-hub and non-hub airports, the number of seats available in February and July were compared to the number of flights available in October to estimate winter and summer seasonality, respectively. Seats for the entire 2001 to 2013 period were summed for each of the 3 months for this analysis. Only airports classified as small- or non-hub in 2013 were included in this analysis. Exhibit 3-3. Average daily flights per small- and non-hub airport, 2001–2013.

38 Effects of Airline Industry Changes on Small- and Non-Hub Airports Exhibit 3-6 shows winter seasonality for small-hub and non-hub airports. Airports with 85% or fewer flights in February relative to October are in red and airports with 115% or more flights in February relative to October are in green. Airports with more flights in the winter include airports near ski areas (e.g., EGE, HDN, ASE, and MMH) and airports in warm-weather vacation areas (e.g., DAB, PIE, SRQ, and AZA). Airports with fewer flights in the winter include airports in northeast vacation areas (e.g., ACK, MVY, and BGR) and Gulf Coast vacation areas (VPS and ECP). For ease of discussion, this section refers to airports by their FAA location identifiers. Appendix A contains a table with descriptive information for each airport in the data set used for this report. Exhibit 3-7 shows summer seasonality for small-hub and non-hub airports. Airports with 85% or fewer flights in July relative to October are in red and airports with 115% or more flights in July relative to October are in green. Airports with more flights in the summer include airports in northeast vacation areas (e.g., ACK, MVY, PVC, and BGR), airports in the northern Great Lakes Region (e.g., TVC, PLN, and Exhibit 3-5. Average daily seat departures per small- and non-hub airport, 2001–2013. Exhibit 3-4. Average aircraft seat size per departure for small- and non-hub airports, 2001–2013.

Data Analysis, Airline Industry Changes, and Case Study Selection 39 Exhibit 3-6. Small- and non-hub airports with substantial winter seasonality, 2001–2013. Exhibit 3-7. Small- and non-hub airports with substantial summer seasonality, 2001–2013.

40 Effects of Airline Industry Changes on Small- and Non-Hub Airports RHI), airports in warm-weather vacation areas (e.g., DAB, SFB, PIE, and PGD) and airports in the Rocky Mountain range (e.g., MSO, BZN, JAC, HDN, and ASE). Only a few airports had fewer flights in the summer (BKG, BLV, IFP, OGD, PSP, and PUW). 3.2.1.2 Details The OAG data can be analyzed at several levels of detail associated with the carrier and ser- vice type. Given that carrier identity was known, carriers could be grouped for analysis. For this study, carriers were assigned to mainline, regional, low-cost (LCC) and “other” carrier groups. Over the 2001–2013 period, the LCC group included the following carriers: AirTran Airways, Allegiant Air, America West Airlines, ATA Airlines, Frontier Airlines, JetBlue Airways, Midwest Airlines, Southwest Airlines, Spirit Airlines, Sun Country Airlines, USA3000 Airlines, and Virgin America. The service type characteristics included whether the service was domestic or interna- tional and the equipment type that provided the service. The top chart in Exhibit 3-8 shows the number of October seat departures offered by car- rier type at small-hub airports for each year of the study period. The aggregate number of seats offered at small-hub airports declined from 6.7 million in 2001 to 6.0 million in 2013, a reduc- tion of 10%. However, there were substantial differences in the change in seats by carrier type, as shown in the bottom chart of Exhibit 3-8. The number of seats offered by LCCs increased by 73%, while the number of seats offered by mainline carriers (and their regional partners) declined by 24% and the number of seats offered by other carriers declined by 57% (this latter group had few flights). Exhibit 3-8. Small-hub seat departures by carrier type, 2001–2013.

Data Analysis, Airline Industry Changes, and Case Study Selection 41 The pattern of seats offered by carrier type at non-hub airports followed the same pattern as that of small-hub airports, although at a greater degree. As shown in the top half of Exhibit 3-9, the aggregate number of seats offered at non-hub airports declined from 3.0 million in 2001 to 2.5 million in 2013, a reduction of 19%. The bottom half of Exhibit 3-9 shows the percent change in seat departures relative to 2001 for each of the carrier groups. By 2013, the number of seats offered by LCCs had increased by 283%, while the number of seats offered by mainline carriers (and their regional partners) declined by 21% and the number of seats offered by other carriers declined by 57%. Exhibits 3-8 and 3-9 do not fully capture the changing service by carrier type at small- and non-hub airports during the research period because they combine the types of service offered by mainline carriers. Mainline carriers either offer service on their own aircraft or on aircraft operated by regional partners. The aircraft operated by regional partners are smaller aircraft (regional jets and turbo-prop aircraft) relative to the aircraft operated by mainline carriers (typi- cally narrowbody jet aircraft) at small- and non-hub airports. Exhibit 3-10 shows the share of combined mainline and regional seats offered on mainline and regional partner aircraft during the research period. At small-hub airports, the share of seats on regional partner aircraft increased from 42% in 2001 to 66% in 2013. At non-hub airports, the share of seats on regional partner aircraft increased from 81% in 2001 to 90% in 2013. The increase in the average number of seats illustrated in Exhibit 3-4 can be explained by changes in the types of aircraft serving small- and non-hub airports over the research period. The most Exhibit 3-9. Non-hub seat departures by carrier type, 2001–2013.

42 Effects of Airline Industry Changes on Small- and Non-Hub Airports substantial effects have been the reduced service on turbo-prop aircraft and the increased service on large regional jet (RJ) aircraft. Small RJs are defined as regional jet aircraft with 50 seats or fewer and large RJs are defined as regional jet aircraft with more than 50 seats. The top half of Exhibit 3-11 shows the number of flights by aircraft type from 2001 to 2013 for small-hub airports. The number of large RJ flights increased from approximately 3,000 in 2001 to 13,000 in 2013, while the number of turbo-prop flights decreased from approximately 29,000 to 6,000. The bottom half of Exhibit 3-11 shows the share of flights by aircraft type from 2001 to 2013. The share of flights by turbo-prop aircraft decreased substantially, while the large RJ and small RJ shares increased substantially. Most of the changes occurred from 2001 through 2004, after which the shares of flights by aircraft type remained relatively stable. Exhibit 3-12 presents similar flight by aircraft type data for non-hub airports. The top half of Exhibit 3-12 shows the number of flights by aircraft type from 2001 to 2013. The number of small RJ flights increased from approximately 10,000 in 2001 to 27,000 in 2013, while the number of turbo- prop flights decreased from approximately 59,000 to 15,000. The bottom half of Exhibit 3-12 shows the share of flights by aircraft type from 2001 to 2013. The share of flights by turbo-prop aircraft decreased substantially, while the small RJ share increased substantially. The shift from turbo-prop aircraft to small RJs continued throughout the 2001 to 2013 period for non-hub airports, unlike the aircraft type share changes for small-hub airports which mostly occurred from 2001 to 2004. 3.2.2 Community Demographics Demographic data for the community in which each airport was located were retrieved from U.S. Census Bureau data sources. The Core Based Statistical Area (CBSA) in which the airport was located was identified. CBSAs include both metropolitan (core urban area of 50,000 or more pop- ulation) and micropolitan (between 10,000 and 50,000 population) areas. Using the identified CBSA, population and per capita income data were retrieved for each year in the research period. Exhibit 3-13 shows 2013 per capita income for the CBSA in which each airport is located, for the top ten and bottom ten small-hub and non-hub airports when ranked by per capita income. Exhibit 3-10. Mainline/regional seat share by operating carrier type at small- and non-hub airports, 2001–2013.

Data Analysis, Airline Industry Changes, and Case Study Selection 43 The median per capita income for each hub group is also shown, to demonstrate the ranges in per capita income. The median per capita income for CBSAs with small-hub airports in 2013 was $41,627 and the median per capita income for CBSAs with non-hub airports was $39,381. 3.2.3 Market Data 3.2.3.1 Traffic Traffic data for each airport were calculated using BTS Airline Origin and Destination Survey (DB1B) data. The DB1B is a 10% sample of airline tickets from reporting carriers collected by BTS. The number of O&D passengers per day and average yield for each airport were calculated for the third quarter of the year of the research period. Yield is a measure of the average fare paid by all passengers per mile flown. Two versions of the average yield were calculated: raw and stage-length-adjusted (SLA). Raw yield may not be com- parable among different airports because the average flight distance may differ. Given that airline costs consist of both fixed and variable costs, yield decreases as flight distance increases because fixed costs are spread over increasingly larger flight distances. By establishing a common assumed flight distance and adjusting yields appropriately, stage-length-adjusted yield permits yield com- parisons among airports. The stage length adjustment uses the following formula: Stage-Length- Adjusted Yield = Raw Yield * sqrt(observed length of haul / industry avg length of haul). Exhibit 3-11. Small-hub flights by aircraft type, 2001–2013.

44 Effects of Airline Industry Changes on Small- and Non-Hub Airports Small Hub Non-Hub Exhibit 3-13. Per capita income for top 10 and bottom 10 ranked small- and non-hub airport core based statistical areas, 2013. Exhibit 3-12. Non-hub flights by aircraft type, 2001–2013.

Data Analysis, Airline Industry Changes, and Case Study Selection 45 Airports were ranked by 2013 O&D passengers; the top ten and bottom ten small-hub and non-hub airports are shown in Exhibit 3-14. The median number of O&D passengers departing each way (PDEWs) in 2013 at small-hub airports was 2,126 and the median number of PDEWs at non-hub airports was 201. Airports were also ranked by 2013 stage-length-adjusted yield. The top ten and bottom ten small-hub and non-hub airports are shown in Exhibit 3-15 (airports without DB1B traffic data are excluded from this exhibit). The median SLA yield in 2013 at small-hub airports was 20.2 cents per mile and the median SLA yield at non-hub airports was 22.7 cents per mile. Small Hub Non-Hub Exhibit 3-14. Daily O&D passengers for top 10 and bottom 10 ranked small- and non-hub airports, 2013. Small Hub Non-Hub Exhibit 3-15. SLA yield for top 10 and bottom 10 ranked small- and non-hub airports, 2013.

46 Effects of Airline Industry Changes on Small- and Non-Hub Airports Small Hub Non-Hub Exhibit 3-16. Uniqueness score for top 10 and bottom 10 ranked small- and non-hub airports, 2011. 3.2.3.2 Airport Uniqueness An airport uniqueness metric was calculated for each airport using the variation in user types operating from the airport and airport catchment areas. The user types consisted of commercial, fractional ownership program, general aviation (GA), rotor, and freight operators. The airport catchment areas were determined using the full price of travel, which includes the cost of a flight and the access, egress, and flight time costs. A uniqueness value was derived for each airport that is meant to represent the economic surplus value associated with all current flight activity (excluding freight). The primary driver of the analysis is based on the idea of opportunity cost—the value of the next best alternative. The metric assesses value by estimating the economic loss (in dollars) that would be incurred by current airport users if the airport were to close entirely, the basic idea being that users’ next best alternative would be to use a suitable nearby airport. This option of course will depend on multiple factors, including distance from the current airport, associated increased travel time and cost to an alternative airport, service characteristics of the alternative airport, and the users’ value of time and sensitivity to increased costs. Exhibit 3-16 shows the top ten and bottom ten small-hub and non-hub airports in terms of uniqueness values for 2011. Uniqueness values were not able to be estimated for a few very small non-hub airports, which are not shown in Exhibit 3-16. These values were only available for 2011. The median value for small-hub airports in 2011 was $52.6 million and the median value for non-hub airports was $4.6 million. These values in essence represent the value of an airport to its users, relative to nearby substitute airports. 3.2.4 Airport Financial Data Airport financial data were retrieved from a database of airport responses to FAA Form 5100-127 (Data retrieved October 15, 2014). This form is used for reporting airport revenues, expenses, and other financial information. The airport financial data are not available for every

Data Analysis, Airline Industry Changes, and Case Study Selection 47 0.0M 0.5M 1.0M 1.5M Enplanements ($20M) ($10M) $0M $10M $20M $30M N et M ar gi n Net Margin Small Hub Non-Hub 0.0M 0.5M 1.0M 1.5M Enplanements ($20M) ($10M) $0M $10M $20M $30M O pe ra tin g M ar gi n Operating Margin Exhibit 3-17. Operating and net margin by number of enplanements for small- and non-hub airports, FY 2013. airport in the study database because some airports do not file these reports or do not file the reports in full. Operating margin is equal to operating revenue less operating expenses. The operating mar- gins for small- and non-hub airports in FY 2013 are shown in the left chart of Exhibit 3-17 arrayed against annual enplanements. As seen there, few of these airports show a positive operat- ing margin, regardless of the number of passengers handled. However, airports of all sizes receive significant non-operating funds. The operating mar- gins shown on the left of Exhibit 3-17 do not include the effects of non-operating revenues or expenses, which consist of the following: • Interest income • Interest expense • Grant receipts (primarily Airport Improvement Program [AIP] funds) • Passenger facility charges (PFCs) • Capital contributions • Other non-operating revenue • Special items If all non-operating income and expense categories are included, this yields the “net margin” for each airport; these figures are shown in the right chart of Exhibit 3-17. Now most small- and non-hub airports show a positive net margin. However, for this analysis, we are primarily interested in whether these airports have sufficient funds to engage in meaningful air service development. As discussed in Section 4.3 below, the FAA places significant restrictions on how AIP grants may be used, and as a practical matter they cannot be used for common ASD efforts (e.g., destination or tourism marketing, direct subsidies to carriers, revenue guarantees, and specific carrier targeting).

48 Effects of Airline Industry Changes on Small- and Non-Hub Airports 0.0M 0.5M 1.0M 1.5M Enplanements N et M ar gi n E xc lu di ng A IP , P FC & C ap ita l C on tri bu tio ns ($25M) ($20M) ($15M) ($10M) ($5M) $0M $5M $10M $15M Adjusted Operating Margin Small Hub Non-Hub Exhibit 3-18. Adjusted operating margin by number of enplanements for small- and non-hub airports, FY 2013. Similarly, the use of PFCs collected from boarded passengers are restricted to projects that enhance safety, security, or capacity; reduce noise; or increase carrier competition. Although such projects may have a positive influence on a carrier’s decision to provide service at a given airport, these are not the sorts of primary ASD efforts that are the focus of this analysis. Finally, capital contribution revenues are also funds that cannot be used for direct air service development. Exhibit 3-18 presents a third view of the financial performance of small- and non-hub air- ports by including operating revenues and expenses, plus non-operating funds except for those involving grant receipts, PFCs, or capital contributions. The research team believes this provides a relevant picture of the potential of these airports to fund primary ASD efforts. Although the above financial results should be viewed cautiously because they are self-reported by airports and often include incomplete information, the numbers suggest that small- and non- hub airports struggle financially from an operational standpoint and may have few resources available for air service development. 3.3 Case Studies 3.3.1 Selection Process The research team used a data-driven approach to develop the list of case study airports. Specifically, using the data on changes in air service at small- and non-hub airports, the research team identified small- and non-hub airports with successful and unsuccessful recent air service histories. The researchers looked at variables such as the percent change in available seats from 2001–2012, the percent change in flights from 2001–2012, and the percent change in the number of air carriers from 2001–2012. At the time of case study selection, 2012 data were the most recently available data. Using the literature review of incentives as a point of departure, the research team

Data Analysis, Airline Industry Changes, and Case Study Selection 49 evaluated the most successful and least successful airports in retaining or attracting service for evidence of the use of innovative incentive programs. This allowed the research team to follow best practices in case study research (see Yin 2003) by identifying both unique and representative case sites. In response to ACRP Project Panel comments, the research team also considered geographical diversity and the presence of Allegiant Air when developing the list of case study airports. The research team also included consideration of statewide programs designed to provide incentives for air service development at multiple airports. 3.3.2 Case Study Airports Based on the described selection process, the research team developed the following list of case study airports: • Small-hub airports – Burlington International Airport (BTV) – Akron-Canton Airport (CAK) – Northwest Florida Beaches International Airport (ECP) – Phoenix-Mesa Gateway Airport (AZA) – Bozeman Yellowstone International Airport (BZN) – Hector International Airport-Fargo (FAR) • Non-hub airports – Toledo Express Airport (TOL) – Redding Municipal Airport (RDD) – Augusta Regional Airport (AGS) – Charles Schulz-Sonoma County Airport (STS) – Monterey Peninsula Airport (MRY) – Asheville Regional Airport (AVL) • Statewide programs – Kansas Affordable Airfares Program (Wichita-ICT) The case study airports have the following characteristics: • 7 with increased scheduled service from 2001–2012; 5 with decreased service • Geographic distribution—1 Northeast, 3 Southeast, 2 Midwest, 1 Central, 1 Mountain West, 1 Southwest, 3 West • 6 with Allegiant service, 6 without Allegiant service The selected airports are identified in Exhibit 3-19, with indicators for 2012 hub status and the change in seats from 2001 to 2012. 3.3.3 Data Analysis of Case Study Airports In-depth data analysis of case study airports was performed to describe the airports, identify trends in airline service, and reveal factors influencing the changes in airline service experienced by the airports. Communities are often interested in both the number of carriers serving an airport and the types of carriers serving an airport. For this analysis, airlines were classified as one of four types: LCC, mainline, other, or regional. LCCs generally have lower fares than their competitors and offer point-to-point service. Mainline carriers offer extensive hub-and-spoke networks and ser- vice on narrowbody or widebody aircraft. Regional carriers operate smaller regional jet or turbo- prop aircraft on behalf of mainline carriers to provide service to markets without sufficient

50 Effects of Airline Industry Changes on Small- and Non-Hub Airports Exhibit 3-19. Case study airports. Mainline Regional LCC Other Exhibit 3-20. Number and types of carriers serving small-hub case study airports, 2001–2013. demand for service on narrowbody or widebody aircraft or to provide increased frequency in markets. Other carriers include those that do not fit any of the other three categories, such as Alaska Airlines and foreign carriers such as Air Canada. In this analysis, if a mainline carrier serves an airport also served by one of the mainline carrier’s regional partners then the statistics reflect service by two carriers (one mainline and one regional). Exhibit 3-20 shows the number and types of carriers serving small-hub case study airports from 2001 to 2013. All of the small-hub case study airports had scheduled service in every year of the period except for ECP (which opened for scheduled service in 2010) and AZA (which did not have scheduled service from 2001 to 2006). The airport with the largest increase in the number of carriers from 2001 to 2013 was FAR (+3) and the airport with the largest decrease was BTV (-4).

Data Analysis, Airline Industry Changes, and Case Study Selection 51 Mainline Regional LCC Other Exhibit 3-21. Number and types of carriers serving non-hub case study airports, 2001–2013. In 2013, all of the airports except AZA received service from at least one low-cost, one main- line, and one regional carrier. AZA received service only from LCCs in 2013. Exhibit 3-21 shows the number and types of carriers serving non-hub case study airports from 2001 to 2013. All of the non-hub case study airports had scheduled service in every year of the period except for STS, which did not have scheduled service from 2002 to 2006. The airport with the largest increase in the number of carriers from 2001 to 2013 was MRY (+2) and the airport with the largest decrease was TOL (-3). All of the non-hub case study airports had service from at least one regional carrier in 2013, except STS. It was very rare for non-hub case study airports to be served by mainline carriers— only AGS and AVL were served by mainline carriers at any point during the study period. Exhibit 3-22 shows the number of seats offered by carrier type at small-hub case study airports from 2001 to 2013. Often, the change in the number of available seats was not as great as would be expected from the change in the number of carriers shown in Exhibit 3-22. This can be partially explained by the consolidation that occurred in the airline industry during the study period. For instance, BTV had regional service from six carriers in 2001: American, Continental, Delta, Northwest, United, and US Airways. In 2013, after the Continental-United and Delta- Northwest mergers, BTV had regional service from three carriers: Delta, United, and US Air- ways. The number of seats offered by United regional carriers in 2013 was more (21,088) than the number of seats offered by Continental and United regional carriers in 2001 (16,047). Exhibit 3-23 shows the number of seats offered by carrier type at non-hub case study airports from 2001 to 2013. Although the number of seats at most non-hub case study airports was fairly steady throughout the study period, TOL experienced a dramatic decline in available seats. The number of available seats at TOL declined from approximately 47,000 in 2001 to 7,000 in 2013, a decrease of 85%. Exhibit 3-24 depicts average daily commercial passenger traffic measured as passengers per day each way (PDEWs) for the small-hub case study airports. The lines labeled Maximum, Aver- age, and Minimum refer to PDEWs for all small-hub airports in the applicable year. FAR and AZA had PDEWs less than the minimum in some years—this is because FAR and AZA were clas- sified as small-hub airports in 2012, but were classified as different hub sizes in previous years.

52 Effects of Airline Industry Changes on Small- and Non-Hub Airports Mainline Regional LCC Other Exhibit 3-22. Seats by types of carrier for small-hub case study airports, 2001–2013. The PDEWs for each of the small-hub case study airports were below the small-hub average for nearly every year, except in 2011 to 2013 when CAK had more PDEWs than the average. Exhibit 3-25 depicts similar information for the case study airports classified as non-hubs. AVL, AGS, and MRY all had above-average PDEWs throughout the research period, and their PDEW figures generally increased over the research period. Although TOL had many more PDEWs than the average in 2001, the PDEWs for the airport declined steadily throughout the research period until it ended below the average in 2013. After beginning service in 2007, the PDEW figure at STS closely tracked the average. Finally, RDD PDEWs were below-average throughout the research period. Exhibit 3-26 shows SLA yield for small-hub case study airports from 2001 to 2013. Average refers to SLA yield for all airports classified as small hubs in the applicable year. SLA yield at FAR, ICT, BZN, and ECP was above-average in most years and SLA yield was below-average in most years at BTV, CAK, and AZA. Exhibit 3-27 depicts similar information for the case study airports classified as non-hubs. SLA yield was below-average at all non-hub case study airports in most years except for AGS in most years and RDD in 2013. Exhibit 3-28 shows year-over-year (YOY) population and PDEW change for small-hub case study airports from 2001 to 2013. The population change is represented as bars and is graphed on the left axis. The PDEW change is represented as the lines and is graphed on the right axis. Nearly all of the communities with small-hub case study airports had positive population growth

Data Analysis, Airline Industry Changes, and Case Study Selection 53 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 0K 1K 2K 3K 4K 5K 6K P D E W FAR ECP BTV AZA Average Minimum Maximum ICT CAK BZN Average AZA BTV BZN CAK ECP FAR ICT Maximum Minimum Exhibit 3-24. PDEWs for small-hub case study airports, 2001–2013. Mainline Regional LCC Other Exhibit 3-23. Seats by types of carrier for non-hub case study airports, 2001–2013.

54 Effects of Airline Industry Changes on Small- and Non-Hub Airports 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 $0.05 $0.10 $0.15 $0.20 $0.25 S LA Y ie ld ICT FAR CAK BZN BTV Average ECP AZA Average AZA BTV BZN CAK ECP FAR ICT Exhibit 3-26. SLA yield for small-hub case study airports, 2001–2013. 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 0 200 400 600 800 1,000 1,200 1,400 P D E W Average TOL STS RDD MRY Minimum Maximum AVL AGS AGS Average AVL Maximum Minimum MRY RDD STS TOL Exhibit 3-25. PDEWs for non-hub case study airports, 2001–2013.

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 $0.16 $0.18 $0.20 $0.22 $0.24 $0.26 S LA Y ie ld TOL STS MRY Average RDD AVL AGS AGS Average AVL MRY RDD STS TOL Exhibit 3-27. SLA yield for non-hub case study airports, 2001–2013. Exhibit 3-28. Population and PDEW change for small-hub airports, 2002–2013.

56 Effects of Airline Industry Changes on Small- and Non-Hub Airports Exhibit 3-29. Population and PDEW change for non-hub airports, 2002–2013. in every year of the research period. Most small-hub case study airports also had positive PDEW growth in most years of the research period. The PDEW change axis is truncated at a maximum of 50% for presentation purposes, so two data points are not shown, both for AZA: 117% in 2009 and 63% in 2010. These high growth rates are a result of AZA receiving scheduled service for the first time in 2008, as noted in the exhibit. ECP is excluded from the exhibits on PDEW change because it only began receiving scheduled service in 2011. Exhibit 3-29 shows year-over-year population and PDEW change for non-hub case study airports from 2001 to 2013. The population change is represented as bars and is graphed on the left axis. The PDEW change is represented as the lines and is graphed on the right axis. Nearly all of the communities with non-hub case study airports had positive population growth in most years of the research period, except for TOL. Most non-hub case study airports also had positive PDEW growth in most years of the research period, with the exception of TOL. Exhibit 3-30 shows the relationship between year-over-year change in CBSA per capita income and PDEW for small-hub airports. Average refers to the average per capita income and PDEW

Data Analysis, Airline Industry Changes, and Case Study Selection 57 Exhibit 3-30. Income and PDEW change for small-hub airports, 2002–2013. for all airports classified as small hubs in the applicable year. The annual change in CBSA per capita income is shown in green bars (positive change in income) and red bars (negative change in income) and is charted on the left axis. The annual change in PDEWs is shown as black lines and is charted on the right axis. The PDEW change axis is truncated at a maximum of 50% for presentation purposes, so two data points are not shown, both for AZA: 117% in 2009 and 63% in 2010. These high growth rates are a result of AZA receiving scheduled service for the first time in 2008, as noted in the exhibit. The effect of the 2008–2009 recession is seen in the sharp decline in per capita income in 2009. Other than that year, year-over-year per capita income increased in nearly all years for most small-hub case study airports, similar to the PDEW trends. Exhibit 3-31 shows the relationship between year-over-year change in CBSA per capita income and PDEWs for non-hub airports. Average refers to per capita income and PDEWs for all air- ports classified as non-hubs in the applicable year. The annual change in CBSA per capita income is shown in green bars (positive change in income) and red bars (negative change in income) and

58 Effects of Airline Industry Changes on Small- and Non-Hub Airports Exhibit 3-31. Income and PDEW change for non-hub airports, 2002–2013. is charted on the left axis. The annual change in PDEW is shown as black lines and is charted on the right axis. STS is excluded from the chart due to data limitations (2002 to 2005 data are missing for STS). The effect of the 2008–2009 recession is seen in the sharp decline in per capita income in 2009. Other than that year, year-over-year per capita income increased in nearly all years for most non-hub case study airports. In many cases, the PDEW changes did not follow the same trends as the income changes for the non-hub case study airports. The primary takeaway from these comparisons of the case study airports is that, despite some similarities with each other, there is a wide variety of airport activity, local demographics, and economic variables among the group. Thus, the research team is confident that the case study airports represent a good cross-section of small- and non-hub airports.