There are a number of limitations in current data about travel to and from the United States; multiple sources of data are required to develop a coherent picture of travel trends. As is generally the case when multiple sources of data are needed, there is reason for concern about comparability of data and the quality of the different datasets. There is no standard unit of measure or definition among the various datasets in the United States about what constitutes a trip or a visit; some datasets count the number of visitors or travelers, while others focus on the number of entries or border crossings at a particular location. This report generally uses the term “visit” rather than “trip” in an effort to enhance comparability amongst the various datasets. Also, it is not possible to readily determine from the current publicly available aggregate data how many travelers made multiple visits into and out of the United States, although this data could be derived in part for air travel.
International travel data can come from surveys, sampling of origin and destination information on tickets, or actual counts. International travel by U.S. residents has been a component of passenger travel surveys, such as the American Travel Survey conducted in 1995 by the Bureau of Transportation Statistics (BTS), now part of the Research and Innovative Technology Administration, which focused on long-distance travel and the 2001-2002 National Household Travel Survey (NHTS), conducted jointly by BTS and the Federal Highway Administration, which generated data on the percentage of long-distance trips taken outside of the United States. While no long-distance travel survey by the Department of Transportation (DOT) is currently planned, if a similar DOT survey were to take place in the future, then consideration should be given to asking additional questions on international travel. Answers to survey questions on specific countries visited, modes of transportation used, country of residence, and duration of international trips would enhance the transportation community’s knowledge of international travel ﬂows. This type of data would also inform other government agencies responsible for implementing passport and border entry measures. Because the sampling frame of the DOT survey only includes U.S. households, it does not provide information about the travel of residents of foreign countries within the United States. This is an important consideration for transportation planning, especially for states that attract a large number of foreign visitors but do not have a primary gateway for their entry. Important transportation planning questions include:
One DOT source of international travel data comes from sampling of air carrier data. Currently, 10 percent of all tickets issued by domestic carriers are sampled by the airlines, with the data transmitted to the Office of Airline Information at RITA/BTS. A 10-percent sample of the millions of international aviation trips on U.S. carriers results in a large sample size. Other sources of international travel data come from the Department of Commerce, International Trade Administration, Office of Travel and Tourism Industries (OTTI) as described on page 3. OTTI is the only published source for outbound air travel estimates of U.S. citizens. It also collects information on transportation modes, travel purpose, and destinations of foreign residents arriving in the United States by air.
Some North American travel data are produced through counts of passengers or border crossings. One source of actual count data comes from Customs and Border Protection (CBP) of the Department of Homeland Security, which provides RITA/BTS with counts on a dozen transportation related variables for inbound travel into the United States. These include some passenger variables: train passengers, passengers in personal vehicles, bus passengers, and pedestrians. Currently, the CBP data are at the port-of-entry (POE) level. Some POEs have more than one crossing, for example, the Buffalo–Niagara Falls POE has four highway bridge crossings and two rail crossings (one of which is on a passenger route). Accessing the POE model data at the individual crossing level would provide a wealth of information on the border crossings that served the most travelers.
Other international travel data that utilize count data are some security and cruise ship passenger data items. Security data on the number of prohibited articles confiscated for all ﬂights, both domestic and foreign, is an example. It would be helpful if international travel data indicated what proportion of the prohibited articles were intercepted from passengers boarding international ﬂights. The cruise ship data are for voyages marketed in North America. It would be useful to know the number of international cruise ship passengers that begin and/or end their trips in the United States. This would eliminate combining cruise ship passengers who do not begin or end their trips at U.S. ports with those passengers whose cruises have U.S. origins or destinations. This would also eliminate the double counting of some cruise ship passengers who ﬂy to Europe or Asia on American air carriers to board cruises that have been marketed in North America. These passengers are currently counted both in the aviation data and the cruise ship data.
In recent years, there has been considerable coordination between American, Canadian, and Mexican transportation and statistical agencies. Since 1991, the North American Transportation Statistics (NATS) Interchange has provided a mechanism for the participating countries to exchange information on transportation statistics, including transportation across borders. As a result of this trilateral coordination, some of the cross border travel data in this report come from Canada and Mexico—the two largest destinations for U.S. travelers and the largest origin of travelers to the United States. In the future, the NATS Interchange will continue to work toward reducing data gaps and to develop a greater understanding of the extent of comparability in transportation statistics in the three countries.
Developing a more complete picture of international travel to and from the United States is likely to be a continuing challenge and will require assembly of data from multiple sources to accurately portray travel and transportation trends. While such an approach has limitations in terms of data comparability, the need to update the international travel and transportation picture periodically is clear.