Regional Airport Planning

Aviation Demand Forecasts (February 2000): Executive Summary

Overview

The aviation forecasts are essential to the updating of the Regional Airport System Plan (RASP) as they provide the underlying estimates of future air passenger and air cargo "demand" for airport capacity, indicating when and where new runways may be needed. The RASP serves as the overall policy document for identifying regional airport capacity needs and alternatives for addressing these needs which can then be considered in the plans of the individual airports. While a number of aviation forecasts were reviewed in the process of updating the RASP forecasts, the new forecasts presented here differ in terms of the comprehensiveness of the analysis, the regional scope, and the methodologies employed.

In developing our forecast methodology, we examined the factors and trends that have influenced the growth of Bay Area air passenger and air cargo demand, both regionally and at San Francisco, Oakland, and San Jose International Airports. Our analysis considers economic and pricing factors which affect domestic and international travel as well as the way the airlines will most likely respond to this growth in terms of the markets and Bay Area airports served.

Using the air passenger and air cargo projections, we then forecast commercial passenger and air cargo flights at each major airport, which when combined with general aviation and military flights, gives a picture of the potential demand for use of the Bay Area airport runways. The forecast of aircraft operations (takeoffs and landings) is derived from trends in airline load factors (the percentage of seats filled), trends in the size of aircraft airlines are purchasing, and the frequency of service provided in competitive markets. For air cargo we attempt to estimate the likely division of air cargo tonnage transported in all cargo aircraft ("freighters") versus that which will be carried in the "bellies" of passenger aircraft. Finally, because airport runways are open to the whole flying public, we expect continued use of all air carrier runways by general aviation (small planes), air taxi companies (small planes for hire), and some military aircraft.

In the following summary of the forecast results we will attempt to draw connections between the forecasts and their implications for the airport capacity assessment in Phase 2 of the RASP update.

Air Passenger and Airport Forecasts

Air Travel Markets
In looking at air passenger forecasts we are concerned both with the rate of growth (when the demand will materialize) as well as how much demand there will be. We are also concerned with where passengers will be flying, which suggests which airlines and airports are likely to provide expanded or new nonstop flights to various domestic and international destinations. Also, the level of demand in these air travel markets will determine the frequency of airline service and the types of aircraft most suited to providing the service. The number of commercial flights relates directly to the demand for runway capacity at each airport.

For air passenger travel, an econometric model was used to analyze air travel growth in 75 Domestic US markets, resulting in a high and low passenger forecast. The model considers socioeconomic factors such as population and income, as well as passenger fares and distance between city pairs. Because the downward trend in passenger fares has been a major stimulus to air travel in the Bay Area and nationally, we have evaluated trends in "real" air fares. We are recommending the mid-range of the high and low forecast for use in Phase 2-- the detailed evaluation of alternative airport system scenarios. Our projections for international air travel also reflect the mid-range of a high and low forecast. But in this case we have relied heavily on the forecasts of other industry players (the two major airplane manufacturers) because of the extensive amount of analysis they have conducted to determine worldwide growth in passenger travel. The characteristics of the most significant markets can be summarized below:

California. California (defined here as including all cities within the state) is the largest air corridor in the world in terms of passenger volumes. Frequent service to major California airports is currently available from all three Bay Area airports. Airlines provide high frequency low fare service, generally with smaller sized jet aircraft that can be turned around quickly at the airport terminal. In 1998, 26% of the local Bay Area passengers traveled in California and generated about the same percentage of total aircraft operations at the three major airports.
Domestic. Domestic air travel markets are subdivided into several regions, such as Western US, Central US, Eastern US, Alaska and Hawaii. The characteristics of many Western US markets (including Seattle and Portland) are similar to the California corridor in terms of service, and they are served with high frequency, smaller jet aircraft. The Central and Eastern US markets have less demand, therefore fewer flights, and are served with larger aircraft that can be operated economically on the longer distance routes.
International. International travel is the fastest growing market, but in 1998 constituted only about 15% of local Bay Area air passengers. It includes flights across our borders to Mexico and Canada, but also direct, long distance flights to Europe and Asia, which are made possible by using larger aircraft with nonstop flying capability. Aircraft technology, which enables long distance travel, has also changed the role of various US cities as "gateways" to foreign countries, as aircraft no longer have to stop at intermediate airports to refuel.
Commuters. Commuter markets are a subset of the California and domestic air markets mentioned above, but are of importance because of the effect of commuter aircraft on airport capacity. Examples of commuter aircraft destinations include out of state cities (Eugene, Or) as well as numerous in state cities (Sacramento, Fresno, Santa Barbara, Eureka, etc.) Commuter aircraft are small aircraft (less than 100 seats) that feed air passengers into an airline hub to connect with a large selection of domestic and international flights. Most commuter flights serve San Francisco Airport, which is a hub for United Airlines. While commuters enable air service to be operated economically in small markets, they also use airport runway capacity that would otherwise be available for larger aircraft.
Connecting passengers. These are passengers that are not "local" to the Bay Area and only use an airport as a transfer point from one flight to another on an extended journey. Connecting passenger demand is included in calculating the number of flights at each airport in the future, but these passengers do not impact the ground transportation system needed to serve the airports. Again, San Francisco International Airport has a relatively high level of connecting passengers due to United Airline's hub operation. About 18% of Bay Area air passengers are connecting passengers.

Our analysis of 75 Domestic Air Service Areas, international, and connecting volumes indicates that the Bay Area air travel market as a whole will be maturing, but punctuated by some still fast-growing markets. Total Bay Area air passenger volume in 2020 will be 111.1 million annual air passengers (MAP), which on an annualized basis represents a growth rate of about 3.1% per year. On the other hand, international air travel will approach growth rates of 4.9% per year. The Bay Area's largest air travel market, the California corridor, will have a "mature" growth rate of about 2.2% per year. As a result of the aggregation of our forecasts for specific markets we expect a net increase of 54 million annual passengers between 1998 and 2020. The largest increases in air travel between 1998 and 2020, ranked from highest to lowest will be:

International, another 12.8 MAP
Connecting, another 10 MAP, served within the airport terminals
Eastern US, another 8.8 MAP
Western US, another 8.6 MAP
California, another 7.6 MAP; and finally
Central US, Alaska, and Hawaii, generating another 5.7 MAP

Looking at the Commuter market to small cities (which must be extracted from the figures above), there will be a total of 5.2 MAP on commuter flights in 2020, or about 4.7% of the Bay Area passenger total.

Airport Distribution of Passengers
Given the regional forecasts above, we next developed a "basecase" passenger forecast for each airport by evaluating airline service strategies and the number of future air passengers within each airport's "catchment" area (the latter analysis relies heavily on information from MTC's 1995 Air Passenger Survey). The basecase estimate represents our starting point for analyzing airport system capacity, but is not our final forecast or conclusion about what demand can be served at each airport. It is simply an essential ingredient to start what is a very complex analysis of future options.

Each airport's share of passengers in a particular market is strongly influenced by the availability of nonstop service. Today, San Francisco has the broadest array of nonstop service to 57 domestic airports, followed by San Jose with nonstop service to 26 domestic airports and Oakland with service to 21. Oakland and San Jose service is highly concentrated in the California corridor and Western US cities, whereas San Francisco International Airport is the major provider of international air service. The central question for the airport basecase forecast is what new services will be developed in the future at Oakland and San Jose airports, particularly in the Domestic markets, and therefore what share of regional domestic demand will remain at San Francisco International Airport. Oakland and San Jose airports currently serve 45% of the domestic local passengers together.

To get at this question we have looked at both the number of passengers that are within Oakland and San Jose airports' service areas today and in the future, as well as the way in which the incumbent airlines at each airport would most likely expand their service consistent with their route structure today. We know, for instance, each airline has a well articulated service strategy today, depending on their type of operation -- whether they employ a hub and spoke system (like United Airlines and other established "majors") or engage in point-to-point service (such as Southwest Airlines which attempts to define profitable city pairs and offer low cost service). In other words, no airline attempts to fly everywhere to serve every customer; rather each has a clear strategy that determines where it flies, what it charges and the type of customer they seek to serve.

We have combined this thinking with our airport catchment area analysis to come up with a basecase airport forecast. It is our belief that Oakland Airport could support new nonstop service in at least fourteen (14) new markets in 2010 (compared to 1998) and an additional eight (8) new markets in 2020. For San Jose, it is our belief that this airport could support service in eight (8) new markets in 2010 and an additional three (3) new markets in 2020. There would be additional nonstop service possibilities at San Francisco International Airport, about four new domestic nonstop markets and 13 new international nonstop services. Most of San Francisco International Airport's growth will, in fact, be in passenger travel to international destinations, with this airport serving over three quarters (76%) of all Bay Area international air travelers in 2020.

The resulting forecasts in Figures ES-1 and ES-2 shows that San Francisco Airport's expected share of regional air travel would decrease to 55% in 2020, while Oakland's share would increase to 22% and San Jose Airport's share would increase to 23%, including connecting passengers. Of Oakland's projected passenger volumes 11.7% of the passengers in 2010 would be associated with new airline service (traveling on nonstop flights not in place in 1998) and 16.8% in 2020. For San Jose airport, the comparable figures are 6.7% in 2010 and 9.2% in 2020. By 2020, Oakland and San Jose airports would be serving slightly over 50% of the Domestic local passengers, up from 45% in 1998. Figure ES-3 shows the distribution of air passengers by county to each airport.

Air Cargo Forecasts

The air cargo industry has grown rapidly as it responds to a number of domestic and international trends: growing overseas commerce, rapid delivery of small packages, just in time delivery of manufactured products, delivery of products that are seasonal or perishable in nature (e.g. agricultural products), and e-commerce. Air freight offers speed and reliability of service for manufacturers and shippers willing to pay higher costs to avoid delayed or uncertain delivery. Air cargo forecasts are most relevant when disaggregated into the types of air cargo transported from each airport. Generally air freight as a category consists of freight (domestic and international) and air mail carried under contract to the US Postal Service. Freight tonnages are much greater than mail tonnages as one would suspect.

From an airport capacity analysis we are concerned with the volume of cargo that will be generated in the future, the airport where it will be served, and how many aircraft operations are involved. Because a significant portion of air cargo is carried in the belly holds of passenger aircraft, not all the projected air cargo will translate into increased aircraft operations. The air cargo industry also has different time schedules than the air passenger flights, using runways in the late evenings when there is less demand from airline passenger flights.

Our analysis of air cargo is methodologically different than that for air passengers. Because of the difficulty in assembling air cargo by geographic origin and destination (as we did for the air travel markets), we have had to rely on a review of global trends and then our ability to translate these trends to airport growth rates. Each airport has developed a historic air cargo niche, and as we have explained above for airline strategies, we believe we can forecast how air cargo will develop within each airport's defined role. We do this by observing and assessing industry projections from a variety of sources. All of the forecasts for worldwide cargo growth are remarkably similar, averaging about 6.2% per year into the foreseeable future. Thus, in contrast to the air passenger forecasts above, the air cargo industry is still in a growth and development stage in most sectors (domestic mail being the anomaly).

San Francisco International Airport
International air freight, dominated by cargo to the Far East, is the leading cargo market at San Francisco Airport (46%) followed by domestic freight (32%) and mail. Due to the frequent widebody aircraft service offered at San Francisco Airport, the majority of the air freight on domestic and international flights (58% in 1998) is carried in the bellies of passenger aircraft.

The air cargo forecast for San Francisco Airport reflects the continuing significant growth in cargo to Asia which is largely responsible for the predicted tonnage increase in international air cargo of over 400% between 1998 and 2020. Domestic freight tonnage is projected to only increase by 55% due to the declining number of widebody aircraft in domestic service, the limited number of airlines operating freighters out of San Francisco Airport, and increasing competition from the trucking industry in the 2 and 3 day delivery market.

Oakland International Airport
Oakland's freight setting is dominated by two of the most successful air cargo "integrator" carriers, Fed Ex and UPS (an integrator airline picks up and delivers cargo from its customers). These airlines have established hub type operations at Oakland, due in part to its better freeway access to various areas of the region (although, as true most places in the Bay Area, freeway traffic is increasing leading to increased travel times and lower freeway reliability).

Oakland's air cargo projections are predicated on a more modest rate of growth for domestic air freight (as explained above for SFO) and a small push by the Oakland air carriers into the Asian cargo market. Because of the large existing domestic cargo volumes, even the moderate growth rate assumed for domestic cargo growth generates about 1.3 million additional tons (187% increase) compared to today's volume.

San Jose International Airport
San Jose serves as the focal point for cargo generated in the South Bay/Silicon Valley area, but is limited in its cargo role by the lack of widebody aircraft serving the airport as well as lack of backup land for air cargo facilities on the airport. As a result, a significant volume of cargo generated in the South Bay is transported to either Oakland or San Francisco Airports for shipment elsewhere. The forecasts assume a domestic growth rate similar to that for Oakland (but applied to a much smaller 1998 base).

Aircraft Operations

Commercial Passenger Operations
We next turn to the discussion of aircraft operations at each airport. The number of operations revolves around three important variables: the aircraft best suited to serve the air travel markets at each airport (number of seats and range), the frequency of service required (number of daily flights), and the load factors (how many seats are filled, on average).

Historically, load factors have been increasing and have been a major factor in airline profitability, particularly as airlines have learned to market unsold seats in new and creative ways. As the ability to fill seats reaches a threshold (it will never be possible to fill all seats on every flight), air passenger demand will more and more need to be served by added flights. At this point, the question arises as to what types of aircraft will be used and whether they are the larger aircraft (more seats and less frequent service) or smaller aircraft (fewer seats and more frequent service). This of course is a topic that preoccupies the airline planner and is of vital interest to the discussion of Bay Area airport capacity issues.

Trends are mixed in relation to these issues. Some airlines have been downsizing their aircraft to provide roomier and more comfortable seating in reaction to passenger feedback. Recent versions of workhorse short to mid range aircraft do not add seats, but extend range, such that these aircraft can now fly transcontinental and to Hawaii. While airport runway capacity issues may logically suggest the need to use larger aircraft, these aircraft are not necessarily suited to certain markets such as the California corridor due to the economics of their operation and longer turnaround times at the airport terminals. Also, airline aircraft purchasing patterns are changing, with airlines retaining existing models longer.

Our thinking regarding future aircraft types in use was to continue the operation of existing aircraft types with modest additions to airline fleets of larger versions of existing aircraft in the future. The net effect of these forecasts is to increase the average size of the domestic fleets serving the Bay Area, a reversal of a general but by no means universal trend among US airlines towards smaller aircraft since airline deregulation.

As a result, total Bay Area airline passenger flights of all types would increase from 639,000 in 1998 to 954,000 in 2020, about a 50% increase overall. The number of passenger flights would, however, grow at a slower rate than the growth in air passengers -- an average of 1.8% per year compared to 3.1% per year for air passengers -- reflecting the increase in seating capacity per aircraft. The combination of air passenger volumes, aircraft seating capacity, and load factors would combine to increase the average number of air passengers per airline operations as follows:

1998: 89
2010: 106
2020: 117

Individual airports would see increased operations over the forecast period. San Francisco Airport's total passenger operations would increase by 118,000 (30%), Oakland's by 104,000 (92%), and San Jose Airports by 92,000 (71%).

These forecasts can be further understood by looking at aircraft seating capacity in individual markets. In the California corridor, we project a significant increase in seats per flight between 1998 and 2010 (up from an average 134 to 154) and then a leveling off to 2020 as the seating capacity per flight only grows to 158 seats, on average. As a result, total California flights would increase about 35% from 1998 levels. In the small aircraft commuter market, the average size commuter aircraft will have about 40 seats in 2010 and 59 seats in 2020; overall operations would grow by about 39%. At San Francisco Airport, smaller commuter aircraft would comprise about 21% of airline flights but carry only about 7.5% of the air passengers.

It is also possible that by 2020 certain long haul international markets could be served by a new category of aircraft, termed New Large Aircraft( NLA), if such aircraft are produced. These aircraft, bigger than the largest passenger aircraft today, would most likely be put into long distance nonstop service to highly capacity constrained airports, such as London and Tokyo. If these aircraft are not manufactured, the number of estimated international operations at San Francisco would increase slightly as these aircraft would be replaced in our forecasts by the largest aircraft currently flying today, but which would have fewer seats (e.g. the 747 and its various versions).

Combining all the considerations above leads us the to following forecast results at each airport shown in Figure ES-3.

Passengers carried per operation and seats per operation go up at each airport between now and 2020
Passenger demand increases faster than the increase in the number of operations
San Francisco Airport continues to maintain the highest number of passengers carried per operation, due to the presence of larger aircraft in long distance services
As mentioned above, 21% of San Francisco Airport's operations will be by smaller commuter aircraft

Air Cargo Operations
At San Francisco International, where a majority of air cargo is carried in the bellies of passenger aircraft, we are forecasting a trend towards more cargo being carried in freighter aircraft along with increases in the average payload per operation. The total number of annual freighter operations would increase from 11,900 in 1998 to 42,700 in 2020. Unlike San Francisco International, very little future Oakland airport air cargo is expected to be carried in the bellies of passenger aircraft (since most of Oakland's passenger service will be "narrow body" type aircraft). Freighter operations would increase by about 130% to 61,300 annual operations in 2020, assuming higher load factors and an increase in widebody freighter aircraft as Oakland extends service into the Asian market. A unique characteristic of cargo activity at Oakland Airport is the use of small aircraft as freighters which feed traffic into the FedEx and UPS hub operations. These aircraft generally use the North Field and their operations are expected to number about 44,300 in 2020. At the San Jose airport, most of the air cargo is projected to be carried in freighters as well, producing about 9,400 operations in 2020. Some international cargo would be transported in the bellies of passenger aircraft as the airport gains widebody service to East Asia.

General Aviation and Military Operations
As explained at the beginning, the analysis of airport capacity isn't complete without considering the likely levels of general aviation and military activity at each airport. Of the two, the level of military operations is of limited impact, totaling a little over 4,000 a year for all three airports.

General aviation, on the other hand, can fluctuate and is difficult to forecast. In particular, the most relevant general aviation numbers are not the total operations at Oakland and San Jose (because each airport has specific runways for general aviation use), but the number of flights using the main air carrier runways in visual and instrument weather flying conditions.

Our general assumption is that general aviation activity at San Francisco and Oakland airports would not be that different in the future than occurs today. This is because general aviation, which has been undergoing a long and prolonged decline in recreational flying, still serves a business purpose for many corporations. Indeed, both business and individuals or groups of individuals chartering corporate aircraft are finding this mode of travel more convenient as airports and airline passenger flights become more crowded. It is also possible for corporate users to fly into general aviation airports in other metropolitan areas which are actually closer to their local ground destination in these areas.

We have arrived at our forecast at San Francisco International by assuming a constant level of operations throughout the forecast period to the level of activity accommodated in 1998. At the Oakland airport, we have assumed a level of general aviation operations, consistent with FAA forecasts, but have not yet divided these operations into North Field (Oakland's main general aviation facility) and the South Field air carrier runway (this division will be made as part of our airport runway capacity analysis).

For San Jose Airport, we have relied on their most recent master plan which assumes a decreasing number of general aviation operations consistent with the landside constraints on the number of aircraft that can be parked at the airport. The decrease in general aviation parking space is an outcome of the latest airport master plan which requires the conversion of some general aviation parking to other airport uses. As the number of aircraft based at the airport decreases, so will the number of operations. As with Oakland Airport we will need to split these operations in a subsequent step between those which use the shorter general aviation runways and those operations that takeoff and land on the main air carrier runway and affect the capacity available for air passenger and air cargo aircraft.

Comparison of Forecasts

Comparing the Bay Area air passengers forecasts for 2010 with other forecasts for the same year, the new regional forecasts of 82.3 MAP are about the same as the earlier "high" RASP forecast, but lower than the FAA's most recent forecast (called the Terminal Area Forecast) of 88.9 MAP and the combined forecasts from the individual airport master plans. Beyond 2010 there are only the recent forecasts prepared for San Francisco's runway reconfiguration analysis with which to compare our new forecasts. The San Francisco forecasts employ a trend line analysis approach. The new regional forecast of 111.1 MAP is slightly above the "low" forecast for the San Francisco analysis, consistent with our conclusion that the longer term growth patterns in Bay Area air travel will reflect a maturing air travel market.

Regarding the basecase airport forecasts, the most relevant comparisons are to the FAA and recent airport master plan forecasts for 2010. Our 2010 forecasts for Oakland and San Jose airports are similar to the recent FAA and Master Plan forecasts, while our San Francisco International Airport basecase forecast is lower than recent FAA or airport forecasts. This is the result of our comparatively lower regional passenger forecast and the gains in nonstop airline service at Oakland and San Jose airports projected in our basecase forecast.

Back to main RASP page