Records of the volume of United States travel into Canada are available according to motor, rail, and boat classifications, and estimates o the number of travelers by other means of transportation are made by a subtractive method. Table 15 shows the equivalent, in number of trips, of the data upon which the computations are based, and other tables following show additional detail. Table 15.-Number of Trips Across the Canadian Border, 1938-39 1 Precision to units is not inherent in the basic data, but the numbers are left unrounded in order that they may be more adapted to checking and identification by students of the account. Two or three significant figures probably comprise the limit of accuracy for most of the items. 2 Based on 3,081,557 cars at 3.17 persons per car. 3 Based on 2,943,159 cars at 3.12 persons per car. Based on 1,246,111 cars at 2.9 persons per car. Based on 1,253,694 cars at 2.91 persons per car. 6 Based on 1,318 cars at 2.8 persons per car. Based on 1,581 cars at 2.84 persons per car. • Adjusted. Based on 783,309 cars at 3.21 persons per car. 10 Based on 312.209 cars at 3.06 persons per car. Based on 419.351 cars at 3.14 persons per car. 12 Based on 244,461 cars at 3.05 persons per car. 13 The numbers of Canadian travelers by motor are not comparable for the years 1938 and 1939, nor are the numbers of Canadian travelers by "other" means comparable for those two years. The reason is that many of the "local" motorists, counted in 1938 as "other travelers," were in 1939 allowed for as occupants of cars on 24-hour permits. It is believed that during 1940 all "local" motorists will have been eliminated from the "other" account and added to the motorist account. The sum of the numbers of Canadian motorists and "other" travelers for 1939 is, however, comparable with the similar sum for 1938. 14 In estimating expenditures around 2,000,000 of these travelers were accorded special treatment as presumed commuters. This is a temporary arrangement. United States motorcars entering Canada for touring purposes are recorded by the Canadian Department of National Revenue according to type of permit obtained at time of entry. These permits are issued for periods of (a) 48 hours or less (prior to 1935, 24 hours or less), (b) 60 days or less, and (c) up to 6 months. The averages to be applied to these data were obtained in 1939 through distribution at the border, by United States immigration officers, of 24,000 questionnaire forms to residents returning from Canada by automobile. In years previous to 1938, from 1928, such questionnaires, although prepared by the Finance Division of the Bureau of Foreign and Domestic Commerce, had been actually distributed through the cooperation of the Dominion Bureau of Statistics, which instructed Canadian immigration officers to hand them to returning United States motorists. Since questionnaire returns from United States motorists visiting Canada report the class of permit obtained as well as expenditures in Canada, it is possible to estimate expenditures according to permit groups, which afford rough groupings by length of stay. Because of the small absolute number of returns in the 6-month permit group, it has been considered preferable to combine the 60-day and 6-month permit averages into a single average, in constructing which the constituent averages were weighted by the number of 60-day and of 6-month permit holders, respectively. Table 16 shows the per-car averages obtained from questionnaires returned by motorists entering Canada on the various types of permit, the number of United States automobiles entered in each class, and the estimated annual expenditures of United States motor travelers in Canada for the years 1932-39, together with certain collateral information compiled from questionnaire returns. Table 16.-Expenditures of United States Motorists in Canada, 1932-39 In several cases the averages used in calculating total expenditures represent slight adjustments of the averages computed from the questionnaire returns. Estimates of expenditures are temporary, pending improvements now in progress. 2 This average is as computed from questionnaire returns, and was applied to most of the 48-hour cars. Since the records of 48-hour permits issued at the New Brunswick boundary are said to include much local traffic (cf. Canada's Tourist Trade, 1935, Dominion Bureau of Statistics, Ottawa, 1936, p. 4, footnote 2), the computed average was applied to 40 percent of the cars entering New Brunswick on 48-hour permits in 1937 and subsequent years, and an arbitrary rate of $2 to the remainder. This is reflected in estimated total expenditures. Estimated total expenditures of 48-hour cars in 1936 and 1935 also reflect adjustments for this situation. The last decimal place is not significant. 1936 average; returns for 1937 were inadequate. Fizures shown, save as otherwise noted, give effect to revisions which result from recompilation of 1938 returns for the sake of comparability with standards set up for the 1939 compilation. Average from original compilation. 7 Adjusted. The component averages were weighted by the respective numbers of 60-day and 6-month permits Issued, adjusted. The computed average expenditure per 60-day car was $88.15 in 1939 and $93.92 in 1935. tures of Canadians in this country. (See table 14 and fig. 6.) The $349,000,000 represented 55 percent of the $639,000,000 combined payments and receipts for travel between the United States and the rest of the world in 1939. The corresponding proportion during 1938 was 53 percent. This $349,000,000 was also equivalent to the value of 42 percent of aggregate merchandise trade between the United States and Canada during the year. United States-Canadian travel expenditures during 1938 equaled half of the value of merchandise trade between the two countries during that year. Table 14.-United States-Canadian Travel Account, 1938-39 1 1 Data for previous years, although in various ways not comparable, will be found in The Balance of International Payments of the United States in 1938, p. 37, and in earlier issues. 2 Expenditures by many "local" motorists are unavoidably included in the account of "other travelers" for 1938, but 1939 expenditures by "other travelers" include expenditures by many local motorists through June only. This is a necessary result of shifts in the classification of volume-of-travel data at their source. It should be possible to make reasonably comparable estimates within the "motorist" and "other traveler" accounts for these years after the new series has been continued for a period in its present form. Since the change involves only a shift in the account to which a portion of the total volume of travel is assigned, the data on volume of travel upon which the entire United States-Canadian travel account is based are not, as a result, incomparable as a whole for the two years. 3 Rough exchange adjustments from Canadian currency figures, to give a total of $93,000,000 United States money spent by Canadians in the United States during 1939, are reflected in the United States dollar amounts shown in the Canadian motorist and rail accounts. The long frontier between the United States and Canada, the proximity to the border of centers of population, the cooperation of both countries in keeping border formalities at a minimum, and, in the case of motor travel, the excellence of highways in both countries are factors which have contributed to the importance of traffic across the northern land border of the United States. Motorists' expenditures for travel between the two countries overshadow those of travelers who cross this border by any other type of transportation. The decrease during 1939 of expenditures in Canada by United States residents, and of expenditures in the United States by Canadian residents, was in neither case a reflection of the trend of industrial activity, which in both countries was at a higher level throughout the year than during 1938. It can be surmised that the same factors which during 1939 considerably reduced travel expenditures between the United States and Europe and the Mediterranean area were connected, to a somewhat smaller degree, with the less marked decline in United States-Canadian travel expenditures. For the 8 months through August 1939 the number of cars recorded as entering Canada from the United States on 48-hour permits was only 2.8 percent less than for the corresponding months of 1938. The remaining months of 1939, during which Canada was at war, showed a decline of 8.2 percent in the number of cars entering Canada on 48-hour permits, however, as compared with the 4 final months of 1938. Cars entering Canada on 60-day or 6-month permits during the first 8 months of the year were 2.6 percent more than in 1938, but during the last 4 months of 1939, 5.2 percent less. The actual outbreak of war appears to have had less deterring effect upon travel to Canada than upon travel to Europe, and the same may be said regarding the fear of war-which, despite its realization, did not prevent a small increase for the entire year in the number of 60-day and 6-month cars entering Canada from the United States. The more significant influence, however, upon total expenditures of motorists-the most important type of traveler to Canada in effect upon expenditures—was a decline in their average expenditures, as mentioned below. Canadian travelers' expenditures in the United States during 1939 $93,000,000were down 8 percent from a revised estimate of $101,000,000 for the previous year. This percentage decline was much less than that shown in expenditures overseas of Canadian travelers as reported by the Dominion Bureau of Statistics,2 which fell off during 1939 by one-quarter of corresponding expenditures in 1938. Expenditures overseas by United States residents similarly decreased by onequarter, as shown in table IV; but travel payments accruing to the United States from visitors from overseas, on the other hand, increased by practically onequarter. The result of these various trends was an increase during 1939 in the proportion which travel moneys interchanged by residents of the United States and Canada represented of moneys paid and received by residents of these two nations for interchange of travel with all foreign countries. Considering travel between Canada and the United States according to method of transportation, it will be noted from table 14 that the automobile-travel account was the chief one which brought expenditures down for both Canada and the United States. Declines in average expenditures of United States motorists, both on long and on short stays, contributed largely to this drop in expenditures by motorists. There was recorded a trifling increase in the number of cars in the important group for which permits for stays in Canada of more than 48 hours were issued. This group regularly provides the largest part of automobile travelers' outlays. The number of cars not covered by such permits (a group numerically larger but of less importance in effect upon total expenditures) declined by under 5 percent from the corresponding figure for 1938. The volume of rail and of boat travel to Canada in 1939 rose slightly above that of 1938, with expenditures of rail travelers edging up to $56,000,000 from $55,000,000 and with expenditures of boat travelers remaining at $14,000,000 for 1939, the same as the revised estimate for 1938. The number of United States travelers to Canada by methods of transportation other than those discussed above was estimated to have increased slightly during 1939 as compared with 1938. Expenditures by this group were placed at $21,000,000 and $20,000,000, respectively, for these years. Estimated expenditures in the United States by Canadian motorists dropped from $49,000,000 in 1938 to $43,000,000 in 1939. Because of certain changes in the mechanics of recording travel, this figure of $43,000,000 includes some expenditures the counterpart of which for 1938 appears in the account of "other" travelers, so that the decline in expenditures by Canadian motorists is only partly depicted in the automobile account proper. Expenditures of boat travelers from Canada were put at $4,000,000, as against $3,000,000 in 1938. A decrease from $25,000,000, estimated spent by other types of Canadian travelers in the United States during 1938, to $21,000,000 in 1939 was due at least in part to improvement in the volume-of-travel data which effected during 1939 the inclusion in the automobile classification-rather than in the group of "other" travel-of some short-stay automobile travel, as just mentioned. A description of the manner of estimating the United States-Canadian travel account, as well as other detail, follows. The method used in estimating expenditures for travel between the United States and Canada consists, in general, of applying estimates of average expenditures by various types of travelers to the number of travelers in the groups to which the expenditure averages pertain. Statistics of volume of travel between the United States and Canada are thus the framework about which it has been necessary to build a system for making estimates. Since the residence of a given traveler determines the country to which his expenditures should be directly attributed, various series compiled by the Dominion Bureau of Statistics which afford this type of information have been found best adapted to the specialized requirements involved in making estimates of outlays by United States residents for travel in Canada, and vice versa. Questionnaires are employed in obtaining average expenditures, in the case of both United States and Canadian travelers, by automobile, train, and boat. During 1939 practically all 1938 averages have been recompiled in order to effect comparability with the averages used in obtaining estimates of expenditures during 1939. Canada's Tourist Trade, 1939. Canada, Department of Trade and Commerce, Dominion Bureau of Statistics, Ottawa. 1940. Records of the volume of United States travel into Canada are available according to motor, rail, and boat classifications, and estimates o` the number of travelers by other means of transportation are made by a subtractive method. Table 15 shows the equivalent, in number of trips, of the data upon which the computations are based, and other tables following show additional detail. Table 15.-Number of Trips 1 Across the Canadian Border, 1938-39 1 Precision to units is not inherent in the basic data, but the numbers are left unrounded in order that they may be more adapted to checking and identification by students of the account. Two or three significant figures probably comprise the limit of accuracy for most of the items. 2 Based on 3,081,557 cars at 3.17 persons per car. 3 Based on 2,943,159 cars at 3.12 persons per car. 6 Based on 1,318 cars at 2.8 persons per car. 9 Based on 783,309 cars at 3.21 persons per car. 13 The numbers of Canadian travelers by motor are not comparable for the years 1938 and 1939, nor are the numbers of Canadian travelers by "other" means comparable for those two years. The reason is that many of the "local" motorists, counted in 1938 as "other travelers," were in 1939 allowed for as occupants of cars on 24-hour permits. It is believed that during 1940 all "local" motorists will have been eliminated from the "other" account and added to the motorist account. The sum of the numbers of Canadian motorists and "other" travelers for 1939 is, however, comparable with the similar sum for 1938. 14 In estimating expenditures around 2,000,000 of these travelers were accorded special treatment as presumed commuters. This is a temporary arrangement. United States motorcars entering Canada for touring purposes are recorded by the Canadian Department of National Revenue according to type of permit obtained at time of entry. These permits are issued for periods of (a) 48 hours or less (prior to 1935, 24 hours or less), (b) 60 days or less, and (c) up to 6 months. The averages to be applied to these data were obtained in 1939 through distribution at the border, by United States immigration officers, of 24,000 questionnaire forms to residents returning from Canada by automobile. In years previous to 1938, from 1928, such questionnaires, although prepared by the Finance Division of the Bureau of Foreign and Domestic Commerce, had been actually distributed through the cooperation of the Dominion Bureau of Statistics, which instructed Canadian immigration officers to hand them to returning United States motorists. Since questionnaire returns from United States motorists visiting Canada report the class of permit obtained as well as expenditures in Canada, it is possible to estimate expenditures according to permit groups, which afford rough groupings by length of stay. Because of the small absolute number of returns in the 6-month permit group, it has been considered preferable to combine the 60-day and 6-month permit averages into a single average, in constructing which the constituent averages were weighted by the number of 60-day and of 6-month permit holders, respectively. Table 16 shows the per-car averages obtained from |