² See my “External Determinants of the Canadian Upswing, 1921-9,” Canadian Journal of Economics and Political Science, XVII, no. 1, 02, 1951, 50 ff.Google Scholar While the initial Canadian upswing was more rapid than the American in 1922, throughout the 1922-5 and also the 1921-5 period as a whole the Canadian upswing was less intense than those abroad.

³ Occasionally the phrase “relative cyclical conditions” will be used below to refer to this concept of divergent speeds of business expansion at home and abroad.

⁴ For the period 1919-25 and the 1926 capital accounts, see Knox, F. A., Dominion Monetary Policy, 1929-1934 (Ottawa: Royal Commission on Dominion-Provincial Relations, 1939), 89–93 Google Scholar; for 1927-9 and the 1926 current and monetary gold accounts, see Dominion Bureau of Statistics, The Canadian Balance of International Payments, Preliminary Statement, 1946 (Ottawa, 1947).Google Scholar

⁵ In five years if the long-term capital account balance is measured indirectly.

⁶ Canada generally had a positive balance of trade with Britain and a negative balance of trade with the United States, and Canada's sterling usually was sold in New York. Because of the proximity of the large-scale foreign exchange market at New York and the magnitude of British-American transactions, the sterling-Canadian exchange rate was usually brought into line with the sterling-American dollar rate. The Canadian-New York rate, in contrast, affords a reasonably good criterion of the external value of the Canadian dollar (partly because most of Canada's exchange dealings appear to have taken place in New York) and hence is used in all these correlations.

Because of the lack of official Canadian data on the earlier years under study and because of discrepancies discovered in some of the years for which Canadian exchange rate data have been compiled, all monthly exchange rate data on the relative values of the American and Canadian dollars used here are from the Federal Reserve Bulletin compilations of the noon buying rates of Canadian exchange in New York. The reciprocal of this rate is called the “price of the New York dollar in Canada.”

⁷ Not presented here for reasons of space. The monthly long-term capital account was estimated indirectly.

⁸ In no year was the correlation with the monthly long-term capital account stronger than with the current account balance. Moreover in 1926, the one year in which the external value of the Canadian dollar was inversely correlated with the annual current account balance, a direct correlation with the monthly current account balances occurred in nine of the months.

⁹ Because of this direct correlation between the current account balance and the exchange rate obviously monetary gold and short-term capital movements could not have stabilized the rate completely. Moreover, the seasonal capital flows induced in the relatively large stock of Canadian optional payment bonds by seasonal exchange rate fluctuations were not strong enough to stabilize the rate.

¹⁰ Thus in 1926 and 1929 when long-term capital imports were important in stimulating Canadian expansion the fluctuation in this account was responsible for all the divergences from the direct correlation between the current account balance and the external value of the Canadian dollar; in 1923-4 when Canadian expansion was still pardy dependent upon capital imports, most of the divergences from the general correlation are explained by this account; in 1925, 1927, and 1928 when long-term capital imports were negligible or even reversed, fluctuations of the monthly short-term capital account were as important as or more important than those in the long-term account in explaining divergences from the general correlation. Moreover, about half the divergences from the general correlation to which fluctuations of the short-term capital account contributed occurred in the three-month period, January to March, which is usually characterized by heavy monetary gold and short-term capital flows.

¹¹ See my “External Determinants,” 52 f.

¹² The trade in outstanding securities may also reflect short-term investments especially because of the large number of inter-listed stocks and Canadian optional-payment bonds, i.e., bonds possessing an external payment feature of some sort. (See Hackett, W. T. G., “Canada's Optional Payment Bonds,” Canadian Journal of Economics and Political Science, I, 1935, 161–70.CrossRefGoogle Scholar) Net purchases of outstanding securities increased in each of the years 1922 to 1925 so that they behaved in an equilibrating fashion. In so far as the rising value of the Canadian dollar made securities abroad cheaper and induced Canadians to buy them, a short-run equilibrating capital movement occurred here.

¹³ The short-term capital account (short-term capital plus monetary gold movements) is referred to here as the “balancing account” since its components appear to have fluctuated on the whole in a balancing fashion. Obviously long-run rather than short-run factors may have influenced particular changes in bank deposits abroad (say) and also some other items of the balance of payments moved in a balancing fashion even though they do not form part of this “balancing account.”

¹⁴ This increase may be attributed in part to rising American prices. The American price level rose by 5.5 per cent in 1924-5 while the Canadian exchange only appreciated by 1.2 per cent.

¹⁵ de Hevesy, Paul, World Wheat Planning and Economic Planning in General (London, 1940), 830.Google Scholar

¹⁶ For the domestic goods price index (1913 = 100) see the Federal Reserve Bulletins, 1920-6; export and import price indexes from Statistics, Dominion Bureau of, Prices and Price Indexes, 1913-1940 (Ottawa, 1942), 52.Google Scholar

¹⁷ See my “Internal Determinants of the Canadian Upswing, 1921-9,” Canadian Journal of Economics and Political Science, XVI, no. 2, 05, 1950, 184–98.Google Scholar

¹⁸ Moreover Canadian bankers were by temperament and training used to gold standard conditions.

¹⁹ Brown, W. A. Jr., The International Gold Standard Reinterpreted, 1914-1934 (New York, 1940), I, 396.Google Scholar See also Statist (London), June 5, 1926, 996. In Dominion Bureau of Statistics, Prices and Price Indexes, 1913-1926 (Ottawa, 1927), 136 Google Scholar, it is stated that “Canada passed from a de facto to a de jure gold basis by the lapsing of the restriction against gold exports” on July 1, 1926.

²⁰ Brown, , The International Gold Standard, 396, note 6.Google Scholar The Gazette (Montreal), 07 13, 1926, 18 Google Scholar, carries an extract from a Canadian Bank of Commerce Bulletin stating that “the Government has permitted the export of gold under license and Canada has practically been on a gold basis since July, 1923.”

²¹ Federal Reserve Bulletin, 07, 1926, 537.Google Scholar Thus, contrary to an opinion evidently held in some quarters, the first earmarking arrangements in Canada did not arise in the 1930's but early in the 1920's.

²² Globe (Toronto), 06 2, 1926, 6.Google Scholar

²³ Canada Year Book, 1936, 915.Google Scholar If the gold deposit in the “Central Gold Reserves” is included, the fall was from 13.8 to 12 per cent of total net liabilities.

The cash reserve ratio computed for single days at three-month intervals rose from 11.6 per cent at the end of March, 1922, to 13.4 per cent at the end of December, but then declined fairly continuously in the next five years reaching a trough of 6.7 per cent at the end of June, 1927. Thus a fall of 6.7 percentage points took place in four and one-half years. Contrast the behaviour of this ratio during the period 1900-1913 when it rose from a trough of 6.7 per cent in September, 1901, to a peak of 11.4 per cent in December, 1911, an over-all rise of only 4.7 percentage points in over ten years. See Viner, J., Canada's Balance of International Indebtedness, 1900-1913 (Cambridge, 1924), 166 f.Google Scholar Aside from this one sharp fall, however, the cash reserve ratio remained relatively stable throughout the 1921-33 period, the fluctuations in the secondary reserve ratio being much greater on the whole. Similarly absolute cash reserves remained relatively stable throughout the upswing while the “outside reserves” fluctuated widely, largely under the influence of the balance of payments position. See Table II.

The ratio of cash plus secondary reserves to total public liabilities fell only gradually throughout these four years, the secondary reserves on the whole having increased sharply along with the increasingly credit character of the current account while the paying back of Dominion notes reduced the internal cash reserves.

²⁴ Dominion notes in circulation per capita fell from an index of 624 in 1921 to 459 in 1925.

²⁵ From 1918 to 1922 Dominion notes borrowed under the Finance Act appear to have been used mainly for deposit in the “Central Gold Reserves” to enable the issue of a corresponding amount of chartered bank notes. There is a fairly close correlation during this period in the monthly banking data between the Dominion note deposits in the Central Gold Reserves and advances under the Finance Act.

²⁶ One reason for this drastic reduction of advances in 1922-3 may have been the desire of the banks to work off their advances because of the expectation that the Finance Act would be discontinued two years after the “conclusion of peace.” Another factor might have been the relatively high rate which the banks had to pay on Finance Act advances compared with what they could earn on call loans abroad.

²⁷ From 1923 to 1927 withdrawals would proceed until the deposits put in since 1923 were used up but would not go past that point.

²⁸ Since the rise of export prices in 1924 and 1925 resulted primarily from independent price rises on world markets, any possible tendency of other countries to curtail their imports from Canada could not be considered an adjustment process caused primarily by the increased value of the Canadian dollar.

²⁹ Total payments to individuals, Canada Year Book, 1945, 910.Google Scholar National income to 1925. ibid., 909; thereafter, Dominion Bureau of Statistics, National Accounts Income and Expenditure, 1926-1950 (Ottawa, 1952), 26.Google Scholar GNP to 1925 from Dominion Bureau of Statistics, Monthly Review of Business Statistics, 03, 1944, 15 Google Scholar; thereafter from National Accounts.

³⁰ Because of the large residuals in these years, a considerable part of which reflect unrecorded transactions in outstanding securities, these movements were probably even somewhat larger.

³¹ During World War I Canada's financial position with Great Britain was reversed as British purchases of Canadian war supplies offset the earlier war credits granted by the Treasury to the Canadian government. By the end of the fiscal year 1917 British advances to Canada exceeded Canadian advances to Britain by $11.3 million. Sessional Papers of Canada, Public Accounts, as cited by Curtis, C. A., “The Canadian Banks and War Finance” in Contributions to Canadian Economics, III (Toronto, 1931), 23.Google Scholar Because of the continued loans to Britain, however, in the calendar years 1918 and 1919 Canada was on balance a net creditor on long-term capital account ( Knox, , Dominion Monetary Policy, 93 Google Scholar) despite the relatively large Canadian borrowing in the newly tapped American market. In the seven fiscal years 1918 to 1924 Britain was a debtor to Canada on war account.

³² Although the “relative cyclical hypothesis” is not strictly applicable since the relative strength of the upswings in the two countries would not appear to be the primary determinant of this flow.

³³ Knox, F. A., Excursus “Canadian Capital Movements and the Canadian Balance of International Payments, 1900-1934” in Marshall, H., Southard, F. A. Jr., and Taylor, K. W., Canadian-American Industry (New Haven, 1936), 300.Google Scholar All non-government securities have been added together to form “industrial” securities here. The investment in industrials was largely in railway securities. Foreign investment in railway securities fell by $3 million in 1922 but rose by $2, $35, and $25 million in the next three years. Foreign investment in non-railway industrial securities became smaller, falling from $149 million in 1922 to $140 and $131 million in the next two years and rising slightly to $133 million in 1925.

³⁴ The average monthly price of the Canadian dollar in New York was below the gold export point during five months of 1928 and every month of 1929.

³⁵ Net transactions in outstanding securities behaved in an equilibrating manner in that net purchases fell sharply in 1928 and also in 1929. But since net purchases predominated in each of the four years most of these transactions cannot be attributed to exchange rate fluctuations or to the effect of the flow of international reserves on interest rates. Other factors would appear to have been major determinants of most of the transactions in outstanding securities.

³⁶ An official gold export of $40 million in 1928 ( Stokes, M. L., The Bank of Canada, Toronto, 1939, 24 Google Scholar) accounts for a primary money contraction of only half the size of the net credit in the balancing account.

³⁷ The secondary reserve ratio fell from a peak of 19 per cent at the end of March, 1926, to 7.6 per cent at the end of December, 1929, while the cash reserve ratio remained fairly steady (Table II). If the gold deposit in the “Central Gold Reserves” is added to the cash reserves the resulting cash reserve ratio falls from 11.2 per cent in 1926 to 9.2 in 1929.

³⁸ Cf. Curtis, C. A., “Credit Control in Canada,” Papers and Proceedings of the Canadian Political Science Association, II (Ottawa, 1930), 111.Google Scholar

³⁹ Statutes of Canada, 13-14 George V, c. 48 (R.S.C., 1927, c. 70). Finance Act advances appear to have been important as reserves for lending for the first time in 1927-8; formerly they appear to have been used mainly for the internal drain.

⁴⁰ Approved securities were Treasury bills, bonds, debentures, or stocks of the Dominion of Canada, the United Kingdom, any province of Canada, and of any British possession; public securities of the government of the United States; and Canadian municipal securities. Promissory notes and bills of exchange either had to be secured by documentary title to purposes. No advances were to be made on promissory notes for trading in stocks or for capital expenditures of any kind.

⁴¹ Cf. Curtis, C. A., “The Canadian Monetary Situation,” Journal of Political Economy, II, 06, 1932, 324.Google Scholar

⁴² The increase in import demand was directed mainly to the United States. From 1926 to 1929 total merchandise imports rose by 28.8 per cent; those from the United Kingdom rose by 18.3 per cent; those from other countries by 20.4 per cent; and those from the United States by 33.6 per cent. (Customs figures.)

⁴³ Since Dominion notes could be obtained so easily under the Finance Act there was a tendency for balances held abroad not to be called upon to provide foreign exchange as long as foreign money market rates were high. See Elliott, J. Courtland, Annalist, XXXII, 07 20, 1928, 125 Google Scholar and XXXIV, Oct. 18, 1929, 754.

⁴⁴ Report of the Royal Commission on Banking and Currency in Canada (Ottawa, 1933), 41.Google Scholar

⁴⁵ Ibid., 59.

⁴⁶ Cf. Curtis, C. A., “Canada and the Gold Standard,” Queen's Quarterly, XXXVIII, winter, 1931, 117.Google Scholar But this is true only in so far as any deliberate decision was made. There is no evidence that the Department of Finance attempted to exercise any control of credit. More than once the government stressed that it intended to assume none of the responsibilities of a central bank, although it performed important central bank functions. The government lived up to its promises!

⁴⁷ The index of export prices was 97.8, 94.2, and 92.2 in the three years 1927 to 1929; that of import prices was 97.7, 96.1, and 94.2 (1926 = 100). The terms of trade fell from 100.1 in 1927 to 98.0 and 97.9 in the next two years.

⁴⁸ The general wholesale price index fell from 100 in 1926 to 97.7, 96.4, and 95.6 in the next three years (Prices and Price Indexes).

⁴⁹ Hevesey, De, World Wheat Planning, 830.Google Scholar

⁵⁰ See my “External Determinants,” 52 f.

⁵¹ In his cursory treatment of “The Canadian Balance of Payments,” chap. XII of Cyclical Movements in the Balance of Payments (Cambridge, 1951)Google Scholar, Mr. T. C. Chang appears to reason thus: Expansion in the United States was more violent than in Canada in 1928 and the first half of 1929 and hence Canadians bought large quantities of American securities (pp. 209 ff.). In point of fact Canadian expansion was more violent in 1928 than the American. In the United States GNP and national income rose by 2.1 and 2.0 per cent (Kuznets data); in Canada they rose by 8.1 and 9.2 per cent. Moreover net purchases of foreign securities fell sharply in 1928 according to the revised D.B.S. estimates.

Incidentally Mr. Chang's statement that his 1926-38 balance of payments data came from Public Expenditure and Capital Formation is suspect. First, no such volume exists to my knowledge; he evidently was referring to Public Investment and Capital Formation since all his other references to the former “source” are appropriate for this publication. Secondly, his capital and monetary gold items cannot have come even from this volume since none appears in it.

Basing all correlations upon the 1926-38 period taken as a whole, Mr. Chang concludes that the typical cyclical pattern for Canada's current account balance is a surplus in prosperity and deficit during slump (pp. 203 f.)–a most extraordinary conclusion since Canada's current account balance grew progressively less active or more passive in the boom period 1926-9 and went in the other direction during the downswing 1930-3, the temporary worsening in 1930 being caused mainly by a rise in dividend and interest payments. Mr. Chang's peculiar conclusion is thus based mainly on the short cycle 1933-8 which saw no real boom. Even here however the fall in the current account balance in 1937 was caused chiefly by the sharp increase of merchandise imports with the first blush of boom conditions. The increasing surplus of the current account balance from 1933 to 1936 reminds one of the “period of the advance” from 1921 to 1925 when no real boom had as yet developed. Having derived a “pattern” which fits considerably less than half of the years he has dealt with, a peculiar period in itself, Mr. Chang takes pains to explain the 1927-8 “exceptions” to his generalizations by the youngness of Canadian manufacturing industry in those days indicated by our dependence upon American capital goods for our development boom. A brief glance at Canada's post-war international accounts shows that our manufacturing industry is still “young” since we are still dependent upon the United States for many of our capital goods. The 1933-8 cycle was the unusual one since it lacked any really strong investment boom.

Unfortunately, many of Mr. Chang's other conclusions are also suspect because they are similarly based upon correlations struck for the 1926-38 period as a whole. If one insists upon deriving a single generalization by correlation analysis to fit a lengthy period, surely it would be preferable to feed a single complete cycle (e.g. 1921-33) or integral multiples thereof into the calculator, especially if one is looking for cyclical behaviour.

⁵² New issues abroad of Canadian government securities rose from $62 million in 1928 to $113 million in 1929; that of industrial securities from $161 million to $200 million. Calculations based on data from Dominion Bureau of Statistics, The Canadian Balance of International Payments: A Study of Methods and Results (Ottawa, 1939), 193–6.Google Scholar Dominion, provincial, and municipal government securities have been combined as the “government” category for present purposes; the remainder as “industrial.”

⁵³ Similarly in 1922 when Canada's initial upswing was relatively rapid a large capital import occurred.

⁵⁴ While liquid businesses would pay off their loans quickly during the downswing, customers with slow loans carried at old values probably needed help for a considerable period.

⁵⁵ Angell, J. W., The Theory of International Prices (Cambridge, 1926), 527.CrossRefGoogle Scholar

⁵⁶ See Haberler, G., Prosperity and Depression (Geneva, 1941), 415–25Google Scholar, on the influence of varying degrees of mobility of capital on the transmission of cyclical impulses from one country to another.

⁵⁷ And the Commercial and Financial Chronicle.

⁵⁸ Preferred and common combined. The peak of common stock issues, $18.6 million, came in 1930 but $18.2 million were floated in 1929.

⁵⁹ Cf. Iversen, Carl, Aspects of the Theory of International Capital Movements (2nd ed., Copenhagen, 1936), 73.Google Scholar

⁶⁰ See Beach, W. E., British International Gold Movements and Banking Policy, 1881-1913 (Cambridge, 1935), 170 CrossRefGoogle Scholar, and White, H. D., The French International Accounts, 1880-1913 (Cambridge, 1933), 218 ff.CrossRefGoogle Scholar

⁶¹ Chang has found that British cyclical fluctuations were less intense than those abroad between the wars and that British capital exports were positively correlated with the British trade cycle. “The British Balance of Payments, 1924-1938,” Economic Journal, LVII, 228, 12, 1947, 482n. and 487.Google Scholar

⁶² White, , The French International Accounts, 220 f.Google Scholar Cf. also Hilgerdt, Folke, “Foreign Trade and the Short Business Cycle” in Economic Essays in Honour of Gustav Cassel (London, 1933), 279.Google Scholar

⁶³ See Bloomfield, A. I., “The Mechanism of Adjustment of the American Balance of Payments: 1919-1929,” Quarterly Journal of Economics, LVII, 05, 1943, 333–77CrossRefGoogle Scholar, which was drawn to my attention in 1950, for a somewhat similar hypothesis to explain the co-variation of the American trade and long-term capital balances. Bloomfield actually stresses domestic business fluctuations as “the common influence” but occasionally makes use of the hypothesis of American business fluctuations being “more intense than those in the rest of the world” (p. 376).

⁶⁴ White, , The French International Accounts, 218–21.Google Scholar

⁶⁵ Ibid., 220 f. The correlation between French activity and capital exports appears slightly stronger than in White's conclusions when the direction of movement is considered. If the eleven cases when no yearly change occurs in the crude measure of business activity are omitted, fifteen of the remaining twenty-two years reveal a positive correlation between the level of French activity and capital exports, i.e. over 68 per cent of these cases.

⁶⁶ Viner, J., Studies in the Theory of International Trade (London, 1937), 344.Google Scholar In correspondence Professor Viner has stated that his “scepticism was confined to the existence of a single and a priori ascertainable cyclical pattern applicable to different cycles and to different countries.”

⁶⁷ Cf. White, , The French International Accounts, 221.Google Scholar

⁶⁸ And the assumption that interest rates, stock prices, and profit rates are directly correlated with the speed of the business expansion. In so far as this is not true, e.g. if an inordinate stock market boom occurred in a country with a slow expansion, the long-term capital flows would become more complex.

⁶⁹ Cf. Mitchell, W. C., Business Cycles: The Problem and Its Setting (New York, 1928), 447 Google Scholar: “… prosperity, with its sanguine temper and its liberal profits, encourages investments abroad as well as at home, and the export of capital to other countries gives an impetus to their trade.”

⁷⁰ Moreover, the 1927 American recession and resulting decrease in capital supply may help account for the decrease in new Canadian issues there that year.