From: Mark Brown
To: All members of the TTRA State and Provincial Travel Researcher List Serve interested in the Canada travel market,
I wanted to make you aware of issues impacting Statistics Canada’s 2017 and 2018 monthly estimates of air traveler volumes to/from Canada, the USA, and overseas countries. Both issues pertain to the shifting from a declaration card-based system to an airport kiosk system for travelers entering/re-entering Canada. This information and its sharing with this research community originates from communications between NTTO and Statistics Canada.
ISSUE #1: apportioning airport traveler counts across the segments of Canada resident, USA resident, overseas resident for the period March to December 2017. Here is Statistics Canada’s official statement on this issue as provided on their website.
Data for Statistics Canada’s Frontier Counts program are produced using administrative data received from the Canada Border Services Agency (CBSA) on all international travellers who have been cleared for entry or re-entry into Canada. This includes residents of Canada, the United States and overseas entering Canada from abroad.
In 2017, the CBSA began introducing the electronic Primary Inspection Kiosk (PIK) system at airports in Canada. The PIK system replaces the E311 Declaration Cards that are completed by international travellers to Canada. As of the end of May 2018, the PIK system was deployed at the following airports: Macdonald–Cartier, Ottawa (March 2017); Vancouver International Airport (April 2017); Toronto International Airport T3 (June 2017); Edmonton International Airport (September 2017); Stanfield International Airport, Halifax (October 2017); Pierre Elliott Trudeau International Airport, Montréal (November 2017); Jean Lesage Airport, Québec (December 2017); and Billy Bishop Toronto City Airport (May 2018).
While awaiting receipt of PIK data, Statistics Canada prepared preliminary estimates for airports at which the PIK system has been deployed. These estimates were based on CBSA reports of total international travellers by airport, while the distribution between Canadian, American and travellers from individual overseas countries were modelled estimates based on historical data and trends, using methods similar to those used to do seasonal adjustment.
This release of Frontier Counts for the May 2018 reference month is the first release that incorporates PIK data from the airports where the system has been implemented. For the months of January 2018 to April 2018, preliminary modelled estimates of traveller counts at PIK airports have been replaced by PIK-based counts. The provincial and national totals to which these counts contribute have also been revised. The preliminary estimates of 2017 traveller counts for PIK airports will be revised at a later date.
While seasonally adjusted data have been revised since 2015, caution should be exercised when comparing with 2017 data that include international travellers to Canada by air for the months of March to December. This is because the revision of the preliminary modelled data for PIK airports in 2017 with actual PIK data has not yet been implemented. Nonetheless, this break in the series is temporary pending revision of the estimates using PIK data.
ISSUE #2: apportioning airport traveler counts across the segments of Canadians re-entering Canada directly or indirectly via transiting the USA. For perspective, in 2016, 1.6 million Canada residents re-entered Canada from abroad by air via USA airports. Here is Statistics Canada’s official statement on this issue.
Data users are also cautioned that the switch from E311 cards to PIK has impacted the historical comparability of some data series. Most notably, there has been an increased tendency of Canadian travellers returning from overseas trips via the United States to report that they are returning from the United States and not overseas via the United States. In the Frontier Counts, this has led to increases in the numbers of Canadian residents returning from the United States by air, and decreases in the numbers of Canadian residents returning from Countries other than the United States by air via the United States (as well as more aggregated series to which these data contribute).
The numbers of travellers to and from Canada by car and other modes of transportation are not affected by revisions in PIK data.
Since the Frontier Counts represent the administrative-data-based benchmark for Canadian tourism statistics, we cannot provide an accurate measure of the amount of overestimation of Canadians returning from the US by plane. However, examination of the degree of change in the January to May 2017-2018 numbers of Canadians returning by plane from overseas countries via the US gives an indication of the likely magnitude. We are looking into options to improve this measure in the future.
As a result of this change in respondent behaviour, users are advised to use caution when comparing 2018 PIK-based numbers of Canadians travelling to the United States or overseas with similar E311-based figures from previous years. Direct comparisons of the amount of increase or decrease between the two time periods are not advised.
In summary, this is my take on the implications of the current issues:
- Statistics Canada air traveler volume counts are becoming increasingly census counts, as sampling counts from E311 declaration cards are being replaced by 100% counts from the airport kiosks;
- Air traveler volumes from Canada to the USA and abroad for 2017 have been at least partially modelized since March 2017. These will be converted to actual census counts in spring 2019; Correct
- Air traveler volumes from Canada to the USA and abroad for 2018 were initially partially modelized for January, February, March, and April, but these data have already been revised with actual counts. Data for May 2018 forward will be based on actual counts. Correct.
- The extent to which the apportioning between direct and indirect flights into Canada has impacted the volume estimates is unknown and unknowable, at least at this time. Thus, Statistics Canada advises against comparing 2018 air traveler volume with data prior to 2018.