THE RELATIVE EFFECTS OF UNCERT’AINTY 0N IMPORT . VOLUME UNDER FLEXIBLE AND P’EGGED' EXCHANGE RATES; ~ THE CANADIAN EXPERIENCE 19517- 1964 r A Thesis for the Degree of Ph. D. 7 - MICHIGAN STATE UNIVERSITY PET ER JOHN GIN-MAN 1969 was» 0-169 MIChtgan State mvcrsity This is to certify that the thesis entitled The Relative Effects of Uncertainty on Import Volume Under Flexible and Pegged Exchange Rates: The Canadian Experience 1951—1964 presented by Peter J. Ginman has been accepted towards fulfillment of the requirements for Ph. D . ECONOMICS degree in ”WJQJ WW Major professor .a—I-I ABSTRACT THE RELATIVE EFFECTS OF UNCERTAINTY ON IMPORT VOLUME UNDER FLEXIBLE AND PEGGED EXCHANGE RATES: THE CANADIAN EXPERIENCE 1951—1964 By Peter John Ginman There has been a growing concern with the adequacy of the postwar international monetary system. The adoption of a flexible exchange rate system to replace the present pegged gold exchange standard has been suggested as one remedy to the problem. One of the arguments most frequently voiced against a flexible exchange rate system is that its very flexibility creates uncertainties which may lead to a contraction of trade volume. This thesis investigates the argument by testing the hypothesis that orderly flexible exchange rates will not appreciably dampen trade volume below those levels attainable under pegged rates. This hypothesis is tested on the Canadian flexible exchange rate experience of 1951 to 1964. The first phase of the study concentrates on the theoretical specification of a deterministic import demand model. This demand func- tion relates import volume to expected exchange rates, incremental forward/spot exchange rate differentials and income as explanatory variables. The expectation hypothesis for exchange rates takes the form of a distributed lag '1“— Peter John Ginman funCtion, where the parameter y is interpreted as the coefficient of expectations—-a proxy for flexible exchange rate engendered uncertainty. Theoretically, Y can vary between positive unity and zero. At these limits, respectively, exchange rate expec— tations either have little effect, as they would under pegged exchange rates, or, as the flexible exchange rate critics argue, have a substantial effect on importing behavior. If uncertainty plays an important role in determining import volume, v would have a value approaching zero. If Canada changed to a pegged rate on a priori grounds, one would expect the import function to shift upwards as im— porters adjusted their respective utility functions to the new trading circumstances where exchange rate expectations are always realized. The second phase of the analysis applies an iterative maximum likelihood estimating technique to a stochastic version of the import model derived in the preceding phase. The "best fit" value of V was found to be 0.6, indicating that a moderate level of uncertainty existed under the Canadian experience with flexible rates. As expected, the income variables and seasonal adjustment dummies explained most of the variation in import volume, while the expected exchange rate and incremental forward/spot exchange rate differential variables contributed little. The maximum Peter John Ginman likelihood parameters were then used to derive estimated import levels. To test whether the import function had shifted during the past May, 1962 return to pegged rates, these estimated imports were extrapolated for the eleven quarter period ending 196“ (IV). This produced further evidence upholding the hypothesis, namely, that imports exceeded expected values in only four out of the last seven quarters, allowing for an adjustment period, and that all but one of these fell within the confines of a stringent confidence band. These findings are further strengthened by tests indicating that the structural parameters of the model did not shift significantly when Canadian exchange rate policy changed. Thus, contrary to the assertions of the critics of flexible exchange rates, for the Canadian experience import volume was not appreciably affected by uncertainty arising from exchange rate variability. While the Canadian experience can provide many in- sights into the pegged versus flexible exchange rate con- troversy, caution must be exercised in an attempted general- izations of these findings. It must be kept in mind that the Canadian experience was unique for several reasons, the two most important being that Canada alone operated on flexible exchange rates in a world of pegged rates and she had close economic ties with the United States. Any con- clusions drawn ignoring these facts could only be con- sidered as purely conjectural. 'THE RELATIVE EFFECTS OF UNCERTAINTY ON IMPORT VOLUME UNDER FLEXIBLE AND PEGGED EXCHANGE RATES: THE CANADIAN EXPERIENCE 1951—196“ By Peter John Ginman A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Economics 1969 . - r. e . f ‘1’" r) ~-‘ ,JIr‘P" ©Copyright by PETER JOHN GINMAN 1969 ACKNOWLEDGMENTS Many people have provided considerable advise and have given me varying degrees of assistance with this dissertation. I would like to thank everyone who directly or indirectly assisted me. My major professor, M. Kreinin, was very patient and helpful in every respect. Professor J. Kmenta gave me considerable econometric ad- vice and most importantly, instilled the necessary confi- dence required to finish. I am also grateful to my col- leagues at Boston University for their suggestions and comments, especially Dr. Joseph Yance, who closely ad— vised me on the statistical techniques, as well as other portions of the thesis. The Boston University Computing Center was very understanding and extremely efficient with the programming and statistical computations. Funds for this analysis were provided by the Department of Economics at Boston University. I want to thank my wife, Noortje, for her unending support, sacrifices and patience as well as her many hours of labor typing the numerous drafts of this disser- tation. Finally, I accept the responsibility for any errors and omissions that may exist. 11 TABLE OF CONTENTS ACKNOWLEDGMENTS LIST OF TABLES. LIST OF FIGURES . . . . . . . . . . . . LIST OF APPENDICES Chapter I. II. III. IV. INTRODUCTION AND OBJECTIVES OF STUDY THE FLEXIBLE EXCHANGE RATE CONTROVERSY: THE ARGUMENTS AND THE LITERATURE. The Flexible Exchange Rate Mechanism Arguments for the Flexible Exchange Rate System . Arguments Against the Flexible Exchange Rate System THE CANADIAN EXPERIENCE WITH FLEXIBLE EXCHANGE RATES. Introduction . Pre- -September, 1950 Period: The 1950 to 1956 Period . . . The 1956- -May, 1962 Period . . Post-May, 1962 Period . . Conclusion THE MODEL . . . Introduction Generalized Import Demand and Supply Functions. The Role of Uncertainty in :Import Markets . . . ESTIMATION PROCEDURE, FINDINGS AND CONCLUSIONS . . Introduction . First Testing Procedure. iii Page ii vii viii 62 62 66 'W.“ .w 7:1“ .- w‘itfi ‘ *3 ne— . ' b—Z/rx Chapter Second Testing Procedure Summary and Conclusions. APPENDICES . . . . . . . BIBLIOGRAPHY iv ‘ KC“! ,‘ Table 3.1 A1 A2 A3 A4 A5 A6 Bl B2 B3 BA BB B6 B7 B8 LIST OF TABLES Foreign Exchange Practices of Canadian Corporations, 1950—1962. . . . Actual and Estimated Canadian Imports 1962 (II) to 1964 (IV) Inclusive Canadian Official Holdings of Gold and Foreign Exchange, 1950-62 Inclusive. Growth in Real Gross National Product, 1950-1962 . . . . . . . Canadian Unemployment, Quarterly 1956-1962 Canadian Wholesale and Retail Price Indices, Quarterly, 1956-1962 Canadian Balance of Payments, 1956-1962 Sales of New Canadian Security Issues to Non— —residents, 1956-1962 Iterative Maximum Likelihood Regression Results . Imports of Goods and Services are Adjusted for Seasonal Variations. . . . . Gross National Product. Canadian Domestic Price Index Foreign Price Index. Canadian Short Term Interest Rate Three Month Treasury Bill Rate--Quarterly, Mid—month Quotations Foreign Short Term Interest Rate, U. S. Three Month Treasury Bill Rate Quarterly, Average Daily Quotations Spot Exchange Rate, U. S. Dollars per Canadian Dollar . Page A0 83 91 92 92 93 9A 96 98 99 100 101 102 103 10“ 105 fig" .- ‘r ‘I— Tab 1e B9 B10 B11 B12 B13 B1H B15 B16 817 Forward Exchange Rate, U. S. Dollar per Canadian Dollar . . . . . . . . . Adjusted Spot Exchange Rate,. U. S. Dollars per Canadian Dollar . . . . . Forward/Spot Exchange Rate Differential . Incremental Forward/Spot Exchange Rate Differential . . . . . Import Volume, Imports of Goods and Services Unadjusted for Seasonal Variation . Gross National Product, Constant Dollars, Unadjusted for Seasonal Variation . Index of Relative Prices Adjusted for Exchange Rates . Expected Exchange Index Actual and Estimated Canadian Imports 1951 (I) to 1964 (IV) Inclusive vi Page 106 107 108 109 110 111 112 113 11“ Figure A. A. 1a 1b LIST OF FIGURES Domestic Market ith Commodity Import Demand Function ith Commodity. Canadian Import Market . . . Import Market . . . Residual Variance and Coefficients of Determination for Y Values 0.1 to 1.0 Actual and Estimated Imports 1950 (I) to 196A (IV), Inclusive . . . Page A6 A6 A8 61 77 82 LIST OF APPENDICES Appendix A. Canadian Background Statistics B. Raw Data. viii Page 90 97 CHAPTER I INTRODUCTION AND OBJECTIVES OF STUDY There has been a growing concern with the adequacy of the postwar international monetary system ever since the Western European currencies attained convertibility in 1958. Both critics and defenders of the present gold- exchange standard realize that adjustments to or complete revision of the system is in order if World trade is to continue its expansion. The increasing trends in inter- national cooperation on all fronts are but an overt mani- festation and realization that modern technology has shrunken the world and increased the degree of specializa- tion and interdependence among nations. Most criticisms and suggested revisions of the pre- sent system revolve around the historical ties between gold and national currencies. More specifically, the debate <3enters on whether to maintain or amend the present pegged eaxchange rate system or to adopt a flexible exchange rate. CDhis latter course of action would call for permanently nnoving away from pegged rates and allowing the value of a <30untry's currency to be determined solely by freely work- :ing international market forces. One of the arguments most frequently voiced against a flexible exchange rate system is that its very flexi- bility creates uncertainties beyond those normally associ- ated with market transactions. These uncertainties, it is argued, will dampen both export and import trade volumes as buyers turn to domestic and/or alternative suppliers who Operate on pegged exchange rate systems. But, as Milton Friedman argues, Instability of exchange rates is a symptom of instability in the underlying economic structure. Elimination of this symptom by administrative freezing of exchange rates cures none of the underlying difficulties and only makes adjust- ment to them more painful.1 The Canadian experiment with flexible exchange rates, as we shall see, provides supporting evidence for Friedman's arguments. This thesis tests the hypothesis that orderly flex- ible exchange rates will not appreciably dampen trade volume below those levels attainable under pegged rates. The Canadian flexible exchange rate experience of 1951 to 1962 is used as the proving ground; consequently, any generalization regarding the hypothesis should be re— stricted to this case. Perhaps the best approach to test the uncertainty liypothesis would be to examine the export side of the k 1Milton Friedman, "The Case for Flexible Exchange 'Rates," Essays in Positive Economics (Chicago: University of Chicago Press, 19537, p. 197. trade balance. This could be justified on grounds that merchandise exports are normally the most important earners of the foreign exchange that is necessary to finance im- port transactions. For a number of reasons, however, this argument does not have much strength for the Canadian case. Since international trade is relatively important to Canada and is heavily concentrated in the American and British markets, exchange rate uncertainty probably affects the Canadian trader to a larger extent than the foreign purchaser of Canadian goods. This view is further substantiated because foreign, especially the American, purchasers of Canadian goods have many alternative domestic and foreign sources of supply in contrast to the limited domestic alternatives for Canadian traders. While this allows the foreign traders to avoid exchange uncertainties, the Canadian importers who resort to using foreign alternative suppliers must confront the flexible exchange rate. Since the uncertainties arising .frcnn a flexible exchange rate system would seemingly affect 1119 Canadian importers more, the‘ :import side of the trade The objectives of this t0 construct a deterministic plausible relationships that this study concentrates on equation. thesis are threefold. First, model that explains the exist between import volume and. the principle determinants of import demand. These ifuiependent variables are, in turn, specified to reflect the inherent uncertainties under flexible exchange rates. Second, a stochastic model is used to estimate the probable extent of uncertainty during the Canadian experience with flexible rates. And third, the effect of this estimated uncertainty level on hypothetical trade volume under flex— ible rates is contrasted with the actual Canadian post May, 1962 experience under pegged rates. After a discussion of the literature surrounding the flexible versus pegged exchange rate controversy, presented in Chapter II, and the background information regarding the Canadian experience under both flexible and pegged ex— change rates of Chapter III, these objectives are examined in great detail in Chapters IV and V. A concise summary of the technique used and the conclusions derived from the study can be found at the end of Chapter V. CHAPTER II THE FLEXIBLE EXCHANGE RATE CONTROVERSY: THE ARGUMENTS AND THE LITERATURE The Flexible Exchange Rate Mechanism Under a flexible exchange rate system the total demand for, and of, a nation's currency is continuously equilibrated by the exchange market mechanism which ex- presses that rate at which units of the nation's currency will exchange for a unit of foreign currency. Assuming that the relevant supply and demand elasticities are sufficiently high, and that there is an absence of de- stabilizing speculation in the exchange market, there will always be some rate that will restore equilibrium in the current and capital accounts of the nation's balance of payments. Under ideal conditions, a deficit in the current ac<3<>unt would give rise to an increased supply of domestic Culrzrency. The market would react to this excess supply anti allocate it by depreciating the country's currency. ThEB increased foreign demand for the depreciating country's exports should allow the export sector to expand, this exllansion being financed out of rising profits resulting frfnn the increased domestic prices of these goods. Since 5 the depreciation increases the price of foreign goods expressed in domestic currency, the rise in export volume should be complemented by a concomitant decline in the country's imports. These adjustments should eventually lead to the restoration of equilibrium in the balance of payments. The length of time necessary to restore equilibrium is, of course, dependent on the existence of excess capac— ity at home, the magnitude of the initial changes in rela- tive prices, and domestic resource mobiltiy and adapta— bility. Because of these considerations there are many possible patterns that the equilibrium adjustment time path could take. For example, if the international adjust- ments to price changes are slow, the initial depreciation of the exchange rates might have to be quite large to accommodate a short-run return to balance of payments equilibrium. An overly .however, turn a deficit ciate the exchange rate forces by dampening the 018.1: ion . large initial depreciation could, into a surplus that would appre- and ultimately reverse the above net effect of the original depre- If short-term speculative capital inflows are stabil- izigrlg, however, the depreciation probably would never have 1x3 Ireach these proportions. Speculative inflows would come intxo play whenever speculators felt that the rate was only ten11:)orarily depreciated and would eventually rise in response to the movement towards a balance of payments surplus. These inflows are helpful to the deficit country since they temporarily finance the deficit and simul- taneously dampen the amplitude of exchange rate variations. In addition to this advantage of maintaining balance of payments equilibrium, many other arguments are put forth in behalf of this exchange rate system. Arguments for the Flexible Exchange Rate System This system insures continuous balance of payments equilibrium. The unfettered international market forces can quickly adjust the demand and supply of a nation's currency without being hampered by political considera— tions, domestic resource reallocations, and internal price and income adjustments. The exchange rate adjustments can quickly and easily realign internal market forces, bringing about balance of payments equilibrium without the compli- cations associated with the fixed gold exchange standard. Thea relative sensitivity of flexible rates to changing Imarlcet forces guarantees continuous adjustments and avoids that economic costs of prolonged disequilibrium as well as ther shock effects of sudden reactions to these problems in the' :form of monetary crises and austerity programs. The embriomic costs of these prolonged resource misallocations are high and would not occur under gradually adjusting flexible. rates . The growing number of flexible exchange rate advo— cates led by Friedman, Meade and Sohmen,l continuously stress the argument that free rates allow domestic authorities more independence to pursue domestic stabili- zation objectives than do fixed rates. The autonomous pursuit of domestic full employment by national govern— ments has been apparent ever since the gold standard "rules of the game" were violated during the inter-war period.2 Under the existing fixed exchange rate system, when govern- ments fail to respond to external payments disequilibrium with appropriate domestic policies, their only recourse, after draining reserves, must be to impose direct controls or to change the exchange rate's value. With the world's present commitment to currency convertibility and pres- sures for removal of all direct and indirect barriers to free trade, the flexible exchange rate seems to be the only plausible way to accomplish these ends. The proponents of the flexible rates argue that whifile domestic policy independence is gained under their synstzem, it is not completely independent, and that the immexritable ties that exist between a nation's domestic and ¥ lFriedman, op. cit., pp. 157-203; J. E. Meade, "The Case for Variable Exchange Rates," Three Banks Review, 1007321 (September, 1955); Egon Sohmen, Flexible Exchange {Rat€ES—-Theory and Controversy (Chicago: University of Chixzago Press, 1961). 2The classic discussion of the inter—war period is fOEuud in Ragnar Nurkse, International CurrencygExperience (Geneva: League of Nations, 19AA). external economic affairs cannot be completely ignored. Haphazard or extreme domestic policies will bring inter— national repercussions regardless of the exchange rate system. They also believe that while the rate is free to change, every effort should be made to minimize the range of fluctuations.3 According to them, flexible rates are not inherently unstable; instability arises from the basic economic forces in the economy and that if these are un- stable, no system, including a pegged rate system, will be able to work effectively. If conditions are not extremely out of line, normal domestic policies aimed at promoting growth with full employment and reasonable price stability, can always be pursued without over concern with their effect on the external balance. The limits of exchange rate variability will not be extreme, according to supporters of the flexible rate; consequently, entry into foreign trade will not be stymied. Furthermore, traders reduce uncertainty by hedging in the forward markets. Broad forward markets should develop spontaneously once experience with the flexible rate System indicates relative stability. An ancillary benefit of these markets is that they reinforce the factors that prompt stabilizing speculative inflows.“ 3As in the Canadian experience. See Chapter III, p. 28. “The literature regarding this will be explored further below. 10 Under the present pegged rate system, countries must constantly be aware of the external effects of domestic policies. If a country has to give higher priority to external conditions and hasn't sufficient policy tools to simultaneously manage the domestic economy, it may find a policy conflict that will lead to external relations dic- tating internal policies.5 Or, as the case would be in most countries, domestic policy considerations would take priority over external conditions, forcing the authorities to implement direct controls and/or after sufficient foreign exchange reserve drainage, to resort to large changes in the exchange rate's peg vis-a-vis the dollar. Milton Friedman believes that flexible exchange rates not only allow for a better harmonization of internal and external policies, but that the internal monetary policies can be more in line with domestic policy objectives.6 Sohmen also argues that flexible rates provide their big— gest service by reinforcing the effects of domestic mone- tary policies.7 For instance, if a country has an infla- tionary gap it normally reacts by raising interest rates. 5This was the case in the United Kingdom prior to their November, 1967 devaluation. Monetary policy became virtually useless at the domestic level because short-term interest rate policy was wholly based on balance of pay- ments considerations. Adherence to this policy cost the monetary authorities control over growth in the domestic money supply. 6Friedman, op. cit., p. 200. 7Sohmen, op. cit., pp. 83-90. 11 This reduces domestic spending, causes capital inflows which increase the demand for the nation's currency, thereby increasing the spot rate, causing increased im- ports and dampened exports. The deterioration of the trade balance takes pressure off domestic prices and/or tends to lower income. The net effect is that, via the flexible rate mechanism, external repercussions rein- force domestic policy. If a deflationary gap existed instead, the above forces reverse themselves enhancing growth in income and employment through an improved trade balance and again increase the effectiveness of monetary policy.8 A final advantage of freely fluctuating rates is that they do away with the need for gold and foreign ex- change reserves, except in the cases where governments use stabilization funds to influence the trend in rate fluctuations. There is no need to support rates if they are allowed to find their equilibrium levels. This is one of the reasons that has contributed to an international liquidity shortage. Under the flexible rate system, these 8See: Robert A. Mundell's "Flexible Exchange Rate and Employment Policy," Canadian Journal of Economics and Political Science, XXVII (November, 1961), pp. 509-517, and 1?Monetary Dynamics of International Adjustment Under Fixed and Flexible Exchange Rates," anrterlnyournal of Economics, LXXIV (May, 1961), pp. 227-257; J. M. Fleming, "Domestic Financial Policies Under Fixed and Under Float- ing Exchange Rates," IMF Staff Papers, IX (November, 1962), pp. 369-380. 12 large public balances could be eliminated and private speculators would satisfy private demands for liquidity.9 Agruments Against the Flexible Exchange Rate System While there are many arguments against flexible exchange rates, they can be condensed down to three princ- iple criticisms: (1) low supply and demand elasticities will prohibit and stifle the necessary equilibrating ex- change rate adjustments; (2) international trade and in- vestment volume will be reduced as the result of the uncertainties emanating from unstable flexible rates; and (3) as a subset of (2) there are strong arguments which indicate that unstable exchange rates will result from destabilizing speculation. A few other reservations re- garding this system will also be discussed at the end of this section. The low elasticities argument is based on the static precept of the Marshall—Lerner condition. This classic theoretical proposition . . . states, in effect, that depreciation will improve the balance of payments of a country and appreciation worsen it, if the sum of the elasticities of demand for a country's exports l and of its demand for imports is greater than one. 9Egon Sohmen, International Monetary Problems and the Foreign Exchanges (Princeton University international Finance Section, Special Papers in International Eco— nomics, No. A; April, 1963). 10Charles P. Kindleberger, International Economics (Ath ed.; Homewood, Illinois: Richard D. Irwin, Inc., 1968), p. 259. 13 For this to work, it is also necessary that supply elastic— ities are infinite and that trade is initially balanced. The early post World War II "elasticity pessimists" were convinced that elasticities were too low, and consequently 11 The that a depreciation would only worsen the deficit. implication underlying the view that devaluation of weak currencies would simply mean their further weakening is that the foreign exchange market is inherently unstable. These views have been countered by the theoretical 11This View can be found in the following literature on the subject: T. Balogh and P. P. Streeten, "The In- appropriateness of Simple 'Elasticity' Concepts in the Analysis of International Trade," Bulletin of the Oxford University Institute of Statistics, XIII (April, 1951), pp. 65-77 esp. p. 66 and the diagram on p. 76; A. J. Brown, "Trade Balance and Exchange Stability," Oxford Economic Papers, VI (April, 19A2), p. 105; P. T. Ellsworth, "Exchange Rates and Exchange Stability," Review of Eco- nomics and Statistics, XXXII (February, 1950), pp. 1-12; G. Haberler, "The Market of Foreign Exchange and Stability of the Balance of Payments," Kyklos, III (Fash. 3, 19A9), p. 19A; R. F. Harrod, "Convertibility Problems," Economia Internazionale, XII (February, 1955), p. 27; S. Laursen and L. A. Metzler, "Flexible Exchange Rates and the Theory of Employment," Review of Economics and Statistics, XXXII (November, 1950), p. 282, p. 295; J. E. Meade, The Balance of Payments and Mathematical Supplement (London: Oxford University Press, 19517: p. 323; P. A. Samuelson, "Dis- parity in Postwar Exchange Rates," in Foreign Economic Policy for the United States, ed. by S. E. Harris (Cam- bridge: Harvard University Press, 19A8), p. A01; G. Stuvel, The Exchange Stability Problem (New York: Augustus M. Kelley, 1951); H. A. von Stackleberg, translation of "Die Theorie des Wechselkurses bein vollstandiger Kon- currenz," Jahrbucher fur National Okonomie and Statistik, CLXI (19A9), pp. 1- 65, appears in International Economic Pa ers, No.1 (London and New York. Macmillan, 1951), l 7, 138-139. 1A 12 13 arguments of Sohmen and Morgan who have shown, re— spectively, that there always exist stable equilibria for foreign exchange supply and demand schedules on both sides of any possible unstable point, and that only in the im- probable case of instability in an underlying commodity market can there be an unstable foreign exchange market. The low elasticity critics also defend their view by using the behavioristic argument that businessmen, accus— tumed to reacting only to large sustained changes in the exchange rate, would not react very strongly to the short— term small changes that accompany a flexible exchange rate system. Furthermore, the relatively inelastic demands of many countries for imports of essential raw materials and foodstuffs also tend to support their low elasticity argu- ment. The recent successful adjustments of several pegged rates coupled with the increase in international competi- tion, have changed these attitudes and discussion of low elasticities have gradually diminished. The second argument against flexible exchange rates will be treated in a little more detail, since it is primarily towards this criticism that this thesis is l2Sohmen, Flexible Exchange Rates, Theory and Con— troversy, pp. 3-11. 13E. V. Morgan, "The Theory of Flexible Exchange Rates," American Economic Review (June, 1955), pp. 291— 292. 15 directed. The central thrust of this argument is that flexible exchange rate systems give rise to uncertainty that will dampen international trade and capital flows. At the core of this criticism lies the notion that un— certainty is due to the instability inherent in the system, and that even if it is not already present, its anticipa- tion is sufficient to bring about a lowering of inter— national transactions of all kinds. Since the remainder of this thesis will be devoted to an examination of the effects of uncertainty on import volume, it seems appro- priate to start the discussion with the arguments concern- ing the commodity trade restrictions that alledgedly result from flexible rates. Critics claim that trade volume will be diminished because of reluctance on the part of traders to contract in the face of exchange rate variability. They argue that there will be increased uncertainty regarding profit mar- gins because there may be changes in the exchange price that traders originally agreed upon. Traders can cut down on these risk factors by transacting in the forward ex- change market, but since this further adds to costs, to say nothing of the shortage of forward exchange when the market is thin, they will limit transactions only to those commodities for which there are relatively secure profit levels. Thus, not only is trade volume cut back, but also the type of transaction. 16 Long—term investment flows will also slow down under flexible rates, argue the critics, because either or both the lenders and borrowers may refuse to obligate them- selves to long-term contracts in the face of such risk. The creditor may demand that the repayments be made in his domestic currency as a hedge against exchange rate vari- ability and since this would shift the rish burden almost entirely on the borrower, he may not make the investment. Therefore, to avoid these long-run risks, the critics argue, we should maintain our present pegged rate system of exchange rates. The advocates of change to a flexible system believe that the maintenance of the present system of pegged rates does not really offer any more security or reduced risk than do the free rates. Long-term risk will be virtually the same under both systems, since there has been a history of, and a good prospect of future periodic adjustments of rates under the present system.lu Scammel argues that expected price levels in the trading countries, not the expected changes in exchange rates, are the ultimate cri- teria for determining the safety of long—term investments.15 Furthermore, Meade feels that fear of exchange controls 1”R. E. Caves, "Flexible Exchange Rates," American Economic Review, LIII (May, 1963), p. 122; W. M. Scammell, Internatioggl Monetary Policy (London: Macmillan, 1961), pp. 95-96; Sohmen, International Monetary Problems and the Foreign Exchange, pp. 6A—65. 15Scammel, op. cit., p. 96. 17 under pegged rates are more of a real threat to long-term investment than are variable exchange rates.l6 Cavesl7 "18 brought about by points out that the "solvency risks price and income changes under pegged rates may deter less risky long-term investments as much as the "conversion risks" under flexible rates. At the root of all the criticism lies the question of the stability of flexible rates. Defenders of the sys- tem feel that instability is due to fundamental problems and disturbances in the real forces of the economy and that instability of the exchange rate would occur, regardless of the system the country happens to be using. Under pegged rates, disturbances to foreign trade and investment may be less than with a flexible rate system in the absence of re— or devaluation, but the relative tranquility can be maintained over the long run only by introducing exchange and other controls. Haberler, for one, believes that small exchange rate variations are preferable and less disturb- ing to international trade and investment flows than are direct controls which may partially or totally exclude 16 p. 17. 17 18"Solvency risks" are unexpected increases or de- creases in quasi-rents that will affect the profitability of a particular investment and its rank on the list of profitable uses of funds. Meade, "The Case for Variable Exchange Rates," Caves, 0p. cit., p. 122. 18 certain markets.19 To the extent that participation in any market involves risktaking, then, of course, flexible exchange rates will give rise to uncertainty. But since under this system the rate is free to fluctuate in every respect, the question becomes, will this bring forth speculation that may ultimately become destabilizing in nature? Using the inter—war period as evidence,20 some critics argue that speculative flows will be destabilizing and consequently will enhance the amplitude and frequency of exchange rate variations. The reasoning behind these views is as follows: once real factors in the economy precipitate a depreciation of the rate, speculators, on the belief that the rate will continue to depreciate, move out of the currency, and by their own actions realize their anticipations. As soon as speculators realize that the rate has depreciated beyond the levels called for by the real forces in the economy, they will in anticipation of its appreciation begin purchasing it, ultimately driving it to an overvalued state, and so on. 21 Friedman concludes that Nurske's data is not evi- dence of destabilizing speculation. He argues that the 19G. Haberler, Currency Convertibilitv (No. 5A1 in the series National Economic Problems; Washington, D.C.: American Enterprise Association, 195A), p. 26. 20Nurkse, Op. cit., pp. 117-122. 21Friedman, op. cit., p. 176. speculative flows were responses to overvalued rates during that period and that consequently such movements were stabilizing, moving the exchange rates closer to their equilibrium levels. Speculative movements of these types are desirable and are stabilizing in their effect on rate movements. The very fact that the speculative moves can go in both directions is in and of itself an argument for flexible as opposed to pegged rates. There is a high probability that there will always be unanimity of opinion regarding the direction of change that the rate will ultimately take. This unity of expecta— tions can lead to changes in the peg that are exaggerated. Under flexible rates, on the other hand, there will be doubt as to what the true rate should be and opposing view- points will lead to stability. These arguments assume that speculators are well informed and that the country's finan- cial authorities will initiate appropriate stabilization policies when needed. Absence of either or both of these assumptions could lead to a breakdown of the system. 22 Meade and other proponents of flexible exchange rates feel that the governments should intervene in the exchange 22J. E. Meade, "The Future of International Pay- ments," in Factors Affecting the United States Balance of Payments, compilation of studies prepared for the Sub- committee on International Exchange and Payments, Joint Economic Committee, U. S. Congress, 87th Congress, 2d Session 1962, p. 250. 20 markets with an exchange equalization fund23 to overcome any change of inappropriate speculative movements. Friedman believes that the assumption regarding well informed speculators is inherent in the profit motive.2u He argues that since speculation is continuous, profits must exist, and since destabilizing speculation would lead to speculative losses--speculation on the whole must be of the stabilizing variety. There has been some criticism of this argument. Baumol's25 model shows that profitable speculation can be destabilizing by assuming that people buy (sell) pfpgp prices have already started to rise (fall), thereby increasing the speed of transactions as well as the frequency of fluctuations. Telser26 upholds Friedman's view by criticizing the assumptions of Baumol's model. Telser suggests that speculators react just before prices change their trend and consequently tend to reverse the flows and stabilize exchange rates. There have been studies to support both the critics and advocates of flexible rates. On the positive side, 23See Chapter III for a discussion of Canada's experience with this preposal, p. 28. 2uFriedman, op. cit., p. 175. 25w. J. Baumol, "Speculation, Profitability, and Stability," Review of Economics and Statistics (August, 1957), p. 7. 26L. G. Telser, "A Theory of Speculation Relating Profitability and Stability," Review of Economics and Statistics (August, 1959), pp. 295—301. 27 Tsiang examined the interwar period and found that the 1919-1926 data for three EurOpean countries did not ex— hibit any destabilizing speculation. In his study of the 1950-195A data covering the Peruvian flexible exchange rate experience, he found similar evidence.28 Both Rhom- 29 and Poole30 berg found that speculation tended to stab- ilize the Canadian rate. While not proving that speculation had a destabiliz— 31 and Eastman32 ing effect on flexible rates, Aliber study- ing the European inter-war experience and the Canadian post World War II experience, respectively, could find no evi- dence that it stabilized the rates either. To avoid the pitfalls that are common in empirical studies of this 27S. C. Tsiang, "Fluctuating Exchange Rates in Countries with Relatively Stable Economics. Some European Experiences After World War I," IMF Staff Papers, VII (October, 1959), pp. 2AA-273. 283. C. Tsiang, "An Experiment with a Flexible Ex- change Rate System: The Case of Peru, l950-l95A," IMF Staff Papers, V (February, 1957). 29R. Rhomberg, "Canada's Foreign Exchange Market: A Quarterly Model," IMF Staff Papers, VII (April, 1960). 3OW. Poole, "The Stability of the Canadian Flexible Exchange Rate," The Canadian Journal of Economics, XXXIII, No. 2 (May, 1967), pp. 205-217. 31R. Z. Aliber, "Speculation in the Foreign Ex- changes: The European Experience, 1919-1926," Yale Economic Essays, II (Sprin, 1962). 32H. C. Eastman, "Aspects of Speculation in the Canadian Market for Foreign Exchanges," Canadian Journal of Economics and Political Science, XXIV (August, 19587. 22 type, such as what measure of stability to use, the his- torical interpretation of data, etc., Kindleberger suggests using a more readily understandable index, such as the type of monetary and fiscal policies in Operation during the time period studied.33 If correct contra-cyclical policies are employed for the situation at hand, then speculative flows should be of the stabilizing kind. These relation- ships seemed to have held during the Canadian experience with flexible exchange rates. Another criticism that has received some attention is the fear that inflationary effects of depreciation could lead to further depreciation and therefore a generally un- 3A stable exchange rate. This argument has some credence in countries where imports are important in the country's cost— of-living index. Under these conditions, exchange rate depreciation would increase the price of these imports and thereby put pressure on the cost-of—living index, which in turn could force domestic prices and wages up and lead to a payments deficit which could cause a further rate depre— ciation. If speculators anticipated this inflation- depreciation cycle, they could further weaken the currency by moving funds to stronger currencies. The argument does 33C. E. Kindleberger, "Flexible Exchange Rates," Monetary Management, prepared for the Commission of Money and Credit (19635, pp. AlO-Al2. 31IScammell, Op. cit., pp. 97-98. 23 35 not carry much weight, according to Lutz since in most major trading countries domestic goods and services make up the majority of items in the cost-of—living index. Furthermore, there are generally lags between all of the phases of this cycle, so that the depreciation has time to offset the payments situation anyway. The final argument against flexible exchange rates is simple. They have never worked. The critics say that the many attempts during the inter-war period failed and the eleven-year Canadian experience also ended in failure. The rebuttal for the inter-war period experience is that conditions during the period did not allow ppy system to work, including pegged rate schemes. A second argument only points out that flexible rates are not the only thing a nation need adopt to maintain equilibrium. As the next chapter will explain at length, the failure of the Canadian experiment with flexible rates was not the fault of the system itself, but of the government's misunderstanding of its operation and persistence in pursuing policies that led to the system's demise. This brings us to a discus- sion of the Canadian experience from 1950 to 1962 and 35F. Lutz, "The Case for Flexible Exchange Rates," Banca Nazionale del Lavoro Quarterly Review, VII (December, 195A), p. 182. 2A the relevance of the preceding discussion to the Canadian case . 36 36The best recent summarial work of literature covering the exchange rate debates can be found in M. 0. Clement, R. K. Pfister, and K. J. Rothwell, Theoretical Issues in International Economics (Boston: Houghton Miflin Co., 1967), Chapter VI. Growing interest in this debate has led to several more pOpularized publications, one of the best of which is M. Friedman and R. V. Roosa, The Balance of Payments: Free Versus Fixed Exchange Rates, Fourth in the series of National Debate Seminars, American Enterprise Institute, Washington, D.C. (1967). CHAPTER III THE CANADIAN EXPERIENCE WITH FLEXIBLE EXCHANGE RATES Introduction Since the principle hypothesis of this study is tested by drawing on the recent Canadian experiment with flexible rates, it is important to have some familiarity with the external and internal economic forces that were in effect during the September, 1950 to May, 1962 period. This chapter undertakes this task and accordingly examines in turn: the pre-September, 1950 events that led to the adoption of flexible rates; the relatively successful experience of September, 1950 to 1956; the eventful 1956 to May, 1962 period that precipitated the demise of the flexible exchange rate experiment; and finally, the post— pegging period immediately following May, 1962.1 Pre-September, 1950 Period Canada shared the exchange experiences of most Western countries since World War I. She remained on the 1This chapter draws heavily on: H. H. Binhammer, "Canada's Foreign Exchange Problems: A Review," Kyklos, XVII (No. A, 196A), pp. 636-652; and L. B. Yeager, Inter- national Monetary Relations (New York: Harper & Row Pub- lishers, 1966), pp. A23-AAO; and other sources as noted. 25 26 gold standard until 1929, adopted a flexible rate until the start of World War II and Operated on a rate pegged to the U. 8. dollar at a ten per cent discount from 1939 to 19A6. During the four-year period from 19A6 to 19A9 the Canadian pegged rate officially changed twice before it was ultimately set free in September, 1950. In 19A6, speculative capital inflows, in anticipa- tion of revaluation and fear that the U. S. inflationary trends brought on by the elimination of U. S. wartime price controls would be transmitted to Canada, prompted the authorities to revalue the Canadian dollar to parity with its American counterpart. In the summer of 19A9, however, Canada had to devalue by ten per cent. This re- turn to the old discount vis a vis the U. S. dollar was prompted by many factors, such as: unchecked domestic inflation; pent-up demand for American goods; failure to maintain traditional surpluses with the United Kingdom as an offset to current account deficits; speculative capital outflows anticipating the ultimate devaluation; and a barrage of competitive European devaluations. Speculators responded almost immediately to the new rate, and the summer of 1950 witnessed an unprecedented inflow of foreign capital. Foreign exchange reserves rose by over $500 million, an increase of almost fifty per cent in the official holdings of gold and U. 8. dollars.2 2See Appendix A, Table A1. 27 While part of this capital constituted long-term invest- ments in the booming raw materials industries, the majority were short-term speculative funds.3 By the end of September, 1950, Canada's rate stabili- zation program had caused her official holdings of gold and U. S. dollars to grow almost to the $1.8 billion mark. Fearing that the impending public release of these statis- tics would spark further speculative inflows, the authori- ties concluded that a freely fluctuating, rather than a revalued pegged rate, would be the better policy choice for Canada. Canada's post World War II difficulty in finding the correct rate, coupled with her favorable experience with a flexible rate during the inter-war period of mone— tary instability, gave strength to this decision.” Accord- ingly, after first assuring the International Monetary Fund that the step was but a temporary experiment designed to allow the free market to decide the appropriate rate to be pegged in the future, Canada went on a freely fluctuating exchange rate on September 30, 1950.5 3Samuel 1. Katz, "The Canadian Dollar: A Fluctuating Currency," Review of Economics and Statistics, XXXV (August, 1953). pp. 236—237. “T. N. Brewis et a1., Canadian Economic Policy (Toronto: The Macmillan Company of Canada, Ltd, 1961), p. 261. 5See Peter M. Cornell, "Exchange Flexibility in Can- ada: Some Underlying Factors," Public Policy, A Year Book of the Graduate School of Public Acministration, Harvard University, 1958; and R. Wonnacott, The Canadian Dollar l9A8-l958 (Toronto: University of Toronto Press, 1958). 28 The 1950 to 1956 Period Throughout the 1950-1956 period, the flexible ex— change rate mechanism behaved in an almost textbook fashion reflecting the true equilibrium between total demand and supply of Canadian dollars. Numerous studies have shown that the relationship between the rate fluctuations and changes in domestic economic conditions were in line with a priori theoretical expectations.6 The only non-market hforce in effect was the Bank of Canada's Exchange Fund Account (EFA). This stabilization tool was used only to check destabilizing short-term trends and allowed any tenacious long—term movement to follow its course.7 Yeager found a strong positive correlation between the rate's fluctuations and purchasing power parity as measured by 6See Leland B. Yeager, "Some Facts About the Canadian Exchange Rate," Current Economic Comment, XX (November, 1958), pp. 39—5A; Samuel 1. Katz, "Le dollar canadien et le course de change fluctuant," Bulletin d'Information et de Documentation de la Banque Nationale de Belguique, 30th year, I, (May, 1955), pp. 7-8; R. Wonnacott, pp. 913., p. 123; E. P. Neufield, Bank of Canada Operations and Policy (Toronto: University of Toronto Press, 1958), p. 208; Rudolf R. Rhomberg, "Canada's Foreign Exchange Market: A Quarterly Model," IMF Staff Papers, VII (April, 1960), p. AA7; James C. Ingram, "The Canadian Exchange Rate, 1950—57," Southern Economic Journal, XXVI (January, 1960), p. 207; R. E. Artus, "Canada Pegs Its Dollar," Tpg Banker, CXII (June, 1962), p. 363. 7John Pippenger, "The Canadian Experience with Flex- ible Exchange Rates," American Economic Review, LVII (May, 1967). pp. SAS-SSS. 29 relative price changes, as well as interest rate differen- tials between the U. S. and Canada.8 Short—term speculative capital movements appear to have been of the stabilizing variety. As soon as it be— came apparent that the flexible rate was not going to vary significantly, short—term capital movements settled into a pattern with inflows matching rate declines and outflows following rate increases. This behavior is also implied by Rhomberg's econometric model of Canada which showed that a one per cent depreciation of the Canadian dollar attracted about the same amount of capital inflow as did a one per cent difference in Canadian over U. 8. interest rates.9 One of the most significant attributes of the change to flexible rates was the end of pre—l950 "hot" money move— ments that had plagued Canada. One of the best arguments supporting flexible ex- change rates is that they allow considerable freedom in the choice of domestic economic policies and minimize their repercussions on the economy's foreign sector. Flexible rates do not completely circumvent the interdependency of domestic and foreign policy, but the free-playing inter— national market forces cause the exchange rate to adjust 8Yeager, Op. cit., pp. 39, 5A. 9Rudolf R. Rhomberg, "A Model of the Canadian Economy Under Fixed and Fluctuating Exchange Rates," Journal of Political Economy, LXXII (February, 196A), p. 12. 30 the external accounts in a complementary fashion.lO Domestic and foreign economic policies are, however, in- extricably linked and since the flexible rate adjusts to changes in real economic forces, any prolonged misuse of domestic policy could cause a perverse reaction of these forces and upset this complementarity. With Canadian exports and imports of goods and ser- vices averaging about twenty and twenty-three per cent of 11 the relationship between domestic and GNP, respectively, foreign economic policies is very important. During the 1950—56 period, the relative freedom to pursue independ- ent12 domestic policies offered by the free exchange rate allowed Canada to effectively halt the inflationary ten- dencies caused by heavy capital investment in resource development. The tight monetary policy had the proper deflationary domestic effects and simultaneously en- couraged capital inflows, which in turn pushed the ex- change rate to a premium, causing a dampening of exports and an expansion of imports. Thus, simultaneously this 10For a good discussion of the close interdependence between monetary policy and the freely fluctuating exchange rate, see Robert A. Mundell, "Problems of Monetary and Ex- change Rate Management in Canada," The National Banking Review, 11 (September, 196A), pp. 77-86. 11See Appendix B, Tables B2 and B3. l2Gerald K. Helleiner points out that independent Canadian monetary policy was due to the existence of a flexible exchange rate in "Connections Between United States and Canadian Capital Markets, 1952-1960," Yale Economic Essays, II, No. 2 (Fall, 1962), pp. 398-A00. 31 policy attracted capital which expanded the economy's capacity and encouraged imports, thereby relieving the excessive pressures of the overtaxed economy.13 The relatively smooth Operation of the Canadian experiment with flexible exchange rates during the 1950- 56 period began to progressively falter in the second half of the decade as misinterpretations of the system led to the pursuit of inconsistent policies. The 1956-May, 1962 Period The year 1956 marked the beginning of a new era in the Canadian flexible exchange rate period. During that year, most of the principle economic indicatorslu began to register evidence that the Canadian economy was slowing down. Economic growth as measured by annual growth in real GNP, averaged 6.7 per cent for the 1950—56 period, began to decline seriously after peaking in 1956 at 8.6 per cent, and averaged only 2.8 per cent for the 1956-62 period. This decline can mostly be attributed to the de- cline in Korean War generated demand, the general down- turn in the United States' economy, the change in world supply and demand conditions for primary product, and 13Rudolph G. Penner in his article, "Inflow of Long-term Capital and the Canadian Business Cycle 1950- 1960," Canadian Journal of Economics and Political Science, XXVIII, No. A (November, 1962), pp. 526-5A2, re- futes the commonly held thesis that long-term capital in- flows had a deflationary effect on the economy. 1“All the indicators referred to in this section can be found in Appendix A, Tables Al through A6. 32 competition in the production of manufactured goods from the Japanese and Europeans. Unemployment, the cohort of declining economic growth, began to rise to conspicuous levels in the 1956—62 period. After averaging only 3.2 per cent for the 1950-56 period, it steadily rose, peaking at 11.3 per cent of the work force in February, 1961. During the 1956-62 period it averaged almost twice the rate of the early period, or 6.3 per cent.15 These high rates were partially due to a surplus of unskilled labor that had been generated by the earlier fast growth of primary industries, Hungarian re- fugees, and a shortage Of vocational training schools. Both wholesale and retail prices were fairly stable throughout the 1956-62 period. One may have expected some deflation to have taken place given the economic trends, but Canadian wage and price structures, similar to those of the United States, tend to be subject to downward rigid- ities. The slight increases in these price indices are mainly attributable to unbalanced sectoral demands pres- sing on inelastic sources of supply that ironically existed in the midst of overall excess capacity. Wholesale and re- tail price indices (1953 = 100) averaged 101 to 99, re— spectively, for the 1950-56 period and rose to 105 and 110 in the 1956-62 era. The highest recordings on both 15Only 1956-62 information are provided in Tables A3 and AA of Appendix A. Check table sources for further documentation of earlier period averages. 33 indices occurred in 1961, the year preceding the exchange crisis. The balance of payments also reflected these trends. From 1950 to 1956, the current and capital accounts were in a healthy condition. The flexible exchange rate comple- mented the domestic expansionary policies and allowed sufficient capital inflow to cover current account deficits. These earlier deficits averaged only C$312 million in con- trast to C$l,222 million over the 1956-62 period. The deficit on merchandise account steadily declined over the 1956-62 period. Imports increased in the first few years, reflecting their complementarity with direct foreign in- vestments as well as the premium on the Canadian dollar, while in the later years of the period they declined, relatively, as income levels fell and the nature of the capital inflows changed. Exports, on the other hand, remained fairly stable throughout the period. The balance on other current transactions did not fare so well. Tourism receipts were consistently well below payments for travel abroad by Canadians. Interest and divident payments were high throughout the period, reflecting the growth in Canadian indebtedness abroad. These payments represented over half of the non-merchandise deficit in five out of the seven years of the period. There were continual deficits in all the other current 3A account items as well, with the majority of these remain- ing outpayments representing immigrant remittances to families abroad. From 1956 to 1962 the overall current account deficits averaged over a billion dollars per year. Short and long-term capital movements covered the current account deficits in four out of the seven years, from 1956 to 1962. In 1957, 1959 and 1960 the authorities were forced to draw down official reserves to maintain overall balance in the external accounts. As we shall see, the insistence on a tight monetary policy in the face of declining economic growth, aside from worsening that trend, had important ramifications in the capital markets. The policy produced a disparity between domestic and American interest rates which caused capital inflows. Of the two most important categories of long—term inflows, namely direct investment and foreign investment in Canadian securities, only the latter was stimulated by these events. The interest rate differential had a pronounced effect on American participation in Canadian security issues at the government, provincial, community and cor— porate levels (see Table 6). This single component of the capital account indicates more than any other balance of payments entry the close interdependence between domestic and foreign economic policies. These inflows were most responsible for the premium on the Canadian dollar as the 35 floating exchange rate correctly equilibrated the excess demand for Canadian dollars with their supply. These current and capital account disequilibria accompanied the general slowdown in Canadian economic activities. Mr. Coyne, then governor of the Bank of Canada, felt that the simultaneous presence of rising unemployment and inflation was the result of excessive imports and large foreign, mostly American,l6 capital inflows. His solution to these problems consisted of a combination of tight monetary policy and controls on im- ports and foreign borrowing by Canadians. He reasoned that tight money would reduce capital expansion and thereby reduce inflation while at the same time dampening import demand. He hoped that by promoting increased pro- tection for the secondary industries and introducing legislation forcing the "Canadianization" of American businesses in Canada that American capital inflows could be stepped. But given the inefficient oligopolistic structure of Canadian secondary industry and the small domestic market, protection would only have aggravated 16For example, foreign investment amounted to al- most $2A billion in 1959, with 75 per cent, or $18 billion originating in the U. S. By 1959, the foreign ownership of major industry had grown to 51 per cent of manufactur- ing, 63 per cent of petroleum and natural gas, and 59 per cent of mining, according to H. I. MacDonald, "Prob- lems for Canada in the World Economy," in Economics: Canada, ed. by M. H. Watkins and D. F. Forster (Toronto: McGraw-Hill Company of Canada Limited, 1963), p. 17A. War-r or“— *5. a; 0e_.jq'_?~-“-_ «a: .2» 36 domestic problems, closed the door on much needed inter- national competition, and passed the costs on to the al— ready burdened Canadian community. Fortunately, these measures were never carried too far before the 1962 cri- sis.l7 That Coyne did not understand the relationship of domestic to foreign economic policies, especially under a flexible rate system is quite evident given his persistent application of a tight money policy. By keeping the money supply restricted through high interest rates he forced Canadians to borrow abroad and thereby perpetuated the capital inflows he desperately wanted to terminate. The free rate, doing its job well, kept the Canadian dollar at a premium and thereby maintained the balance of payments deficits. Tight monetary policy continued to aggravate the situation until Coyne was forced out of office in 1961. Fortunately, the full impact of these policies were prob- ably never felt by the distressed economy. Harry Johnson, in his study of monetary policy as a short—run stabiliza— tion instrument undertaken for the Canadian Royal Commis— sion on Banking and Finance, points out that the lags l7See Harry G. Johnson, The Canadian Quandary: Eco- nomic Problems and Policies (Toronto: McGraw-Hill Company of Canada Limited, 1963), pp. 11—21; and Jacob Viner, "American Capital and the Canadian Economy," in M. H. Watkins and D. F. Forster, Economics: Canada, op. cit., pp. 188—19A. 37 between policy need and actual implementation, and imple- mentation and effect ("inside" and "outside" lags) are very long. He concludes, ". . . that the effect of mone- tary policy on the Canadian economy is imprecise, slow, and variable; there is a relationship present, but it is extremely hazy."l8 If these lags were shorter, in all probability the crisis of 1962 would have occurred much sooner. Until Coyne's ouster in 1961, both he and Finance Minister Fleming failed to admit any relationship between monetary policy, capital inflows, and the overvalued cur- rency. After continual criticism by academic economists, backed by a steadily worsening situation, Fleming finally reversed his position. To stimulate business activity, discourage capital inflow and lower unemployment, he lowered interest rates and rejected overt nationalistic economic policies. Unfortunately, the return to easy money was accom— panied by official intervention in the exchange market. Rather than allow the free rate to depreciate itself in response to the lowered interest rates, the E. F. A. was employed to artificially depress the exchange rate. While some connection was finally seen between the monetary policy, unemployment and capital flows, the crucial link between these and the free rate was misconstrued. The l8Johnson, op. cit., p. 187. 38 reduction in capital inflow would have reduced the premium on the Canadian dollar, lowered the exchange rate, caused an increase in exports, dampened imports, and thereby reduced the current account deficits that had been so rampant since 1956. Canadian reserves fluctuated widely during the eleven month period from June, 1961 to May, 1962 (see Table A1). The E. F. A. found itself first manipulating the rate down— ward, gaining reserves, and then trying to maintain an overvalued rate, causing a steady loss of reserves, until the Spring crisis ended with the Official adoption of a fixed, devalued rate on May 2, 1962 at a pegged price of 92.5 U. S. cents.19 The relative freedom in the formula- tion of domestic policy afforded by a free rate became obvious during this period as . . . Canada had seen the ironic adoption of contractionary monetary and fiscal policies to cope with an ermegency stemming from official action to depreciate the Canadian dollar, in turn undertaken in hopes of expanding employ- ment and production.2 The presistent problems of current account deficits, intense capital flows, speculative and otherwise, the change to a manipulated exchange rate, and finally, the pegging of the rate, prompted many people to blame the 19During the l2—month intervention, the rate depre- ciated by 9.5 per cent, bringing the total decline in the value of the Canadian dollar to l2.A per cent. 2OYeager, International Monetary Relations, 100. cit., p. “39. 39 freely fluctuating rate for Canada's economic problems in the 1956—62 period. These allegations could not have been more incorrent. The free rate did its job well-—it equated the total demands and supplies of Canadian dollars on the exchange market. It was the misinterpretation of the advantages of flexible rates and the persistent follow- ing of incorrect policies that gave rise to the trouble. Throughout the entire flexible exchange rate period Canadian importers had the opportunity to use the forward exchange market as a hedge against uncertainty.21 An October, 1962 survey of importing corporation undertaken as part of a Royal Commission study sheds some interesting light on this activity.22 Table 3.1 provides a summary of the survey's findings. In general, the forward markets were not too extensively used by either classification of firm, but there is a noticeable difference in the foreign exchange operations of small and large firms. As the table indicates, a high proportion of the small firms dealt ex- clusively in the spot market, and while the percentage of large firms that chose to cover their positions is greater, over half of the survey respondents in this classification still dealt exclusively on the spot market. In sum, the 21The theoretical implications of this mechanism and its incorporation into the model are discussed at length in the next chapter, pp. 53-56. 22J. H. Young and J. F. Helliwell, "The Effects of Monetary Policy on Corporations," Royal Commission on Bank— ing and Finance (196A), Chapter XI, pp. A19-A26. AO survey's evidence indicates that all categories of Canadian corporate importers made regular spot purchases during the stable period of the flexible exchange rate experience, and increased their activities during the 1960-62 period of relative instability, and virtually ceased forward opera- tions after May, 1962. TABLE 3.1.—-Foreign exchange practices of Canadian corpora- tions, 1950—1962. Percentage distribution of survey replies Occasional Always use Always use ASFIEE Of the spot orfgiigpgnt the forward Total market transactions market Under $10 million 76% 6% 18% 100% $10 million and over 51% 10% 39% 100% Source: J. H. Young and J. F. Helliwell, "The effects of Monetary Policy on Corporations," Royal Commis- sion on Banking and Finance (196A), p. A20. Post-May, 1962 Period After a brief stabilization period immediately follow- ing the official actions of May 2, 1962, things took a turn for the worse. The indecisive election in June, 1962, coupled with the contradictory remarks by the Conservative minority, renewed the loss of confidence and capital out- flows began to mount. Reserves fell catastrophically be- tween May 30, 1962 and June 22, 1962 as the E. F. A. A1 introduced an operation designed to support the newly pegged rate. These actions plus a commitment to deliver forward exchange of $239 million brought the reserve level down to $861 million. On June 2A, 1962 the Prime Minister announced the exchange emergency and proclaimed that the May 2 rate would be maintained at all costs. To back up his statement, the government mobilized $1,050 million in cash and standby credits from inter— national sources and instituted special measures including an increase in the bank rate to 6 per cent; temporary sur- charges from 5 to 15 per cent on non-essential imports; and a decrease in the duty exemptions allowed Canadian tourists. By September, 1962 confidence had been restored, monetary policy eased and capital inflows resumed on a regular basis. As the foreign exchange reserves accumu- lated, the government reduced its commitments to the foreign nations and international organizations that had come to its rescue. Conclusion Oversight and a misunderstanding of the important relationships between domestic and foreign economic poli- cies caused Canada to experience an exchange crisis and the loss of a unique and successful experiment with a freely fluctuating exchange rate. Policy makers incor- rectly diagnosed the causes of declining economic growth and followed restrictive monetary and fiscal policies. A2 The failure to recognize the dependency between monetary policy, capital inflows, balance of payments deficits, and an artifically overvalued currency, thrust blame on a perfectly working floating exchange rate which led to active interventions in the foreign exchange markets as well as a series of nationalistic policies to thwart cap- ital inflows from the United States. These actions forced a return to a fixed rate in May, 1962. Canada's lessons were learned too late. Its actions had forced it to once again become a member of the fixed gold exchange standard community wherein its domes- tic policies could no longer be sought independently of exchange rates and reserve level considerations. The next chapter formulates a deterministic import demand model that incorporates the possible motivational factors ex- plaining import level during this Canadian experience with flexible exchange rates. CHAPTER IV THE MODEL Introduction This chapter seeks to explain the plausible rela- tionships that exist between uncertainty, caused by expec- tations about future exchange rate levels, and current import levels. Using a Marshallian partial equilibrium framework, theoretical import demand and supply functions are derived from domestic excess demand and supply func- tions to show the differential impacts of pegged and flexible exchange rate systems, after which both approaches are incorporated into a model for the purpose of investi- gating the hypothesis. The focus of these discussions is on identifying the specifying the determinants of the im- port demand function. The next chapter will express this function in an empirically testable form as well as present the parameter estimates for use in deriving conclusions regarding the hypothesis. Generalized Import Demand and Supply Functions For expository purposes, the following discussion assumes that all functions are linear in two variables with all other explanatory variables held constant. This 43 AA approach simplifies the algebra and allows for parallel graphical presentations. Buyers and sellers of imported commodities are assumed to be rational economic agents with an awareness Of alternatives and knowledge of all the relevant market datam sufficient to maximize their respective utility functions. Any non-market forces, such as tariffs, quotas, exchange stabilization and/or ex- change control schemes, etc., that add to or affect the costs of purchasing goods are assumed to be given and will not be dealth with in any explicit way. Before the effects of uncertainty arising from the flexible exchange rate mechanism can be discussed, the relevant import functions have to be derived. The follow- ing discussion is in terms of the ith commodity so that l aggregation problems can be avoided at this stage. The domestic demand for the ith commodity can be expressed as: A.1 le = fl (poi, ps, Y), i=1,...,n where Qd is the quantity of the ith commodity de- l manded per unit of time; PC is the domestic price of the ith commodity; 1 PS is the price of domestic substitutes, and Y is a measure of income appropriate to the ith commodity. 1These and other estimation problems are discussed in the following chapter. ”— A5 For any given state of technology (i.e., production func— tion) the domestic supply function of the ith commodity is expressed as: A.2 Q = f (p , w ), si 2 ci 1 where Qi and p0 are defined above and w i i is the price of inputs. A linear representation of equations A.1 and A.2 is shown in Figure A.1a. At prices below the market clearing price pC , say pC , the domestic demand for the ith com— 1 i modity exceeds itszdomestic supply by a quantity A-B. This is the quantity of imports that will be demanded at pmil equal to C-D in Figure A.1b. The import demand func- tion portrayed in Figure A.1b is then the difference be- tween equations A.l and A.2, or “03 Qd - Q = Md where the import demand function is generalized as follows: “on Mdi = f3 (r, poi, pfi’ Y), where pi and Y are defined as before, and r is the exchange rate expressed in terms of the number of units of foreign currency per unit of domestic currency, and pf is a foreign price index appropriate for the i ith commodity. ._..“\""-l .soaooEEoo noa .soaeoesoo soapocom panama anomaH ADV npfi poxnmz ofiummaoo Amy .H.: mhswfim .B.D\HWU Ed A6 A7 A parallel process could be followed to derive ex- cess supply functions in each country producing the ith commodity. The sum of these excess supply functions would represent the import supply function. From the Canadian point of View, however, this function would appear highly or even infinitely elastic, since changes in her import demand for the ith commodity would not effect its world price.2 This import supply function is expressed as follows: A.5 P = fA (pC ,p3r’F)’ si 1 fi i where ps and F are the import supply price (c.i.f.) to i 1 Canada and freignt and any other duty on imports of the ith commodity, respectively, and the other variables are as defined above. By combining certain independent variables into an index of relative prices, r/, as follows: A.6 r/ = pf/rpc, the import market equilibrium can be portrayed graphically as in Figure A.2 below. With the independent variable de- fined as above in A.6, it is easy to see the importance that the type of exchange rate system would have on the import market. Under a pegged exchange rate system, the 2This assumption is discussed and adopted in the following chapter, p. A8 level of imports would be largely determined by relative prices, but under flexible or freely fluctuating rates the relation would be compounded by changes in the rate as well as relative prices. The increased likelihood of change in the independent variable would tend to increase the degree of uncertainty involved in import transactions. Since this could lead to a dampening of import activities, it is important to understand the role of uncertainty and of measures to counter its effects. That is the task of the next section. Figure A.2. Canadian Import Market. A9 The Role of Uncertainty in Import Markets Uncertainties exist in all markets that lack perfect knowledge. When commodity markets become international in scope, the level of uncertainty increases further due to the introduction of the vagarious foreign exchange market. At this stage in the formulation of the model, it is important to examine the ramifications that uncertainty has on the demand for imports under flexible exchange rates. This discussion is confined to the effects of un— certainty on current real transactions with only passing reference given speculation, stabilizing or otherwise, in the short—term capital and foreign exchange markets. Al- though importers are not overtly involved in speculation, in the normal sense Of that term, their very actions have the same effects on the foreign exchange markets as those of professional speculators. Their decisions to hold off or purchase foreign exchange depend on their expectations regarding the future level of the exchange rate, as well as the short-term interest rates at home and in the countries with which they transact business. This behavior will enhance existing uncertainties and consequently the transactions of professional speculators as well. Because the degree of uncertainty about the future exchange rates grows more than in proportion to their time horizons, both importers and professional speculators have difficulty in predicting the future levels of exchange 5O rates. Assuming that the importer must pay for an import in foreign exchange at some given data in the future, he has to estimate a probability distribution of possible outcomes to guide his actions, such as: A.7 P (r*) = f (r*) dr*, that yields an expected value of the exchange rate on the given future date, where P is the probability and r* and dr* are the expected exchange rate and incremental changes in it, respectively. After netting out the costs involved in the trans- actions, he gets an estimate of the profits yielded. If he is spending an amount X on the purchase of the foreign exchange and the unit cost of the transaction is equal to x, his profits would amount to: A.8 w = X (r* - r - x), where r is the present observed rate. Assuming that the importer has the following utility function, that relates all possible gains and losses to a utility index, A.9 U = U (n) = U (n (r) ), where U (0) = 0 and U" < 0, indicating diminishing margi- nal utility for each additional dollar earned, or not earned. _ 51 Given the preceding relationships, the expected value of the importers satisfaction can be expressed as: 00 I U (n) . P (n). O A.10 E (U) It is possible that this expected utility could turn out negative even though the expected value of profit is posi- tive. There is a greater likelihood of this occurring the greater the degree of uncertainty about the future rate, (i.e., uncertainty could outweigh the expected value of profits and he may rationally not import). There is a greater chance of this occurring the faster the marginal utility of gain diminishes, and the greater the dispersion of the expected exchange rate.3 Thus, the rational im- porter, while not overtly speculating, can alter the total demand and supply of currency and in the face of seemingly profitable ventures may not import due to uncertainty caused by flexible rates. An alternative explanation of importing behavior in the face of uncertainty can be made in terms of the Hicksian elasticity of expectations concept.“ Hicks defines "the 3This discussion is an adaptation, in part, of J. Vanek, International Trade: Theory and Economic Policy (Homewood, 111.: Richard D. Irwin, Inc., 1962), Chapter 10, pp. 158—185. “J. R. Hicks, Value and Capital (2d ed.; London: Oxford, l9A6), pp. 203-206. 52 elasticity of a particular person's expectations of the price of a commodity x as the ratio of the proportional rise in expected future prices of x to the proportional rise in its current price."5 The elasticity of price expectations can vary between the limits of zero and one. An elasticity of zero indicates that current price changes have no effect on expected future prices, while an elas— ticity of one implies that previously expected constant long—run equilibrium prices will remain constant at cur— rent price levels. By allowing this spread in the range of elasticities, Hicks recognizes that past prices, deemed ”normal" may not be the sole explanation of price expecta— tions. Since in the real world conditions are seldom, if ever, "normal," our concept of the elasticity of expecta- tions will express a change in current price as a deviation from what was the past expected normal price instead of a deviation from what prices were up to now. If r*t and r/t are the expected normal price and the actual price of the 1th commodity in time period t, respectively, the Hicksian definition of the elasticity of expectations, v, implies log r*t — log r* -l A.11 v = . ,* , O§vt'lr/l) + (1 - y>t E2vr/ + + t—2 0.. <1 - y>t‘1yr/l The logarithmic likelihood function for M1’ M2, . Mn is _ n 2 41. 5.8 L — — log (2 w o ) - 202 2 [Mt -d - "045 t l (l-Y) _ t 2 Maximizing L with respect to 0, BO’ (1 - Y) and 90 is equivalent to minimizing 75 (l - Y) n (l - Y) S 2 [M -d -B W — t=l t o t (60 - a) (l - Y)t]2 with respect to the same parameters. Since we know that Osy3 (2293.6585) SEE (98.5373) 10See Appendix B, Table Bl for estimated parameters of other values of Y. llSlightly higher values of R2 were found for the logarithmic formulation for all values of Y. These results are rejected, however, because they are misleading in that the transformation of the regressands into their natural logarithms probably accounts for most of the differences in R2 values and also because the "t" values consistently indicate lower levels of significance for all variables, for all values of Y. .o.fi 0p H.o mwzfim> r Mom coapmcfishmpmo mo mpcmfioammmoo pom mommapm> annuammm H.m whowfim .Hm manna .m xficcoaa< "mohzom >0.H 0.0 0.0 s.0 0.0 m.0 0.0 mfl0 «.0 H.0 0.0 F 00000. as _ _ _ _ r _ _ .00s0 u.oahm 1.0mpm m 1.0m0m 1.0:5m Owa. 1 v.0mwm 1.0mwm 1.o>~m r.om>m 1.0mnm 77 ommm. 1 I.oomm 1.0Hmm -.0m00 -.omwm 0 -.0000 omwm. I -.omwm -.oowm 1.05mm 1.0mmm U «c in 78 The partial correlation coefficients and relevant elastici— ties are as follows: r/ ..... -o.26A5 ; -0.2756 r'd .... 0.1965 y ...... 0.8772 ; 0.9u58 Ql ..... —0.078i Q2 ..... 0.3956 Q3 ..... —0.5828 (l—Y)t.. 0.2573 Unfortunately, we cannot use the Durban-Watson test for autocorrelation since it is not applicable to regression models wherein the lagged value of the dependent variable is also one of the regressors.l2 The empirical findings of the first testing procedure tend to support the basic hypothesis of this thesis. A coefficient of expectations of 0.6 implies that uncer- tainty about future exchange rates affects present import- ing activities. It should be emphasized, however, that this value of Y is only a partial indicator of the relative importance of uncertainty to Canadian importers and has to be evaluated in conjunction with the second testing pro— cedure below. Further indications of the relative unim— portance of the index of relative prices, r/, is indicated' l2J. Durban and G. S. Watson, "Testing for Serial Correlation in Least—Squares Regression," Pts. I and II, Biometrica, 1950 and 1951. 79 by only a 90 per cent significance level and a partial correlation coefficient of -0.26A5. While the aApriori expectation of an inverse relationship between r/ and M was realized, no such designation can be made about r'd. Only if expectations regarding exchange rate variability affects the utility functions of all importers in the same direction and to the same degree is it possible to make similar predictions regarding the sign of the incremental forward/spot exchange rate differential variable, r'd. The findings that r'd is statistically insignificant sup- ports the arguments presented in Chapter III which showed that for the Canadian experience hedging activities via the forward exchange mechanism played a relatively minor role.13 This is further indicated by the relatively low partial correlation coefficient of 0.1965. Y, the real GNP variable, predictably explains most of the variation in import volume. Its high significance level and partial correlation coefficient of 0.8772 sup- port this view. The statistical significance levels of the seasonal variables and their partial R2's, given the geographical location of Canada, also corroborate all a priori expectations. The elasticities of the r/ and Y variables are relatively meaningless since the elasticity of r/ is really the total elasticity of an index of relative 13See Chapter III, pp. 38-A0. 80 prices adjusted for changes in the exchange rates and the income elasticity is understandably large since it includes the important investment components and their attendant impact of import volumes. This first testing procedure has given us some in- sights into the relative impact of uncertainty on Canadian import volume. The incorporation of our estimate of Y, the coefficient of expectations into the model will allow for a closer examination of the role of uncertainty result- ing from flexible exchange rates. Second TestingiProcedure The second testing procedure involves using the maxi- mum likelihood coefficient of expectations, 0.6, to solve equation 4.21 for various values of r*.lu Once this series is generated for the fifty-six quarter period 1951 (I) to 1964 (IV), it is used with other basic data15 over the same period in the structural import demand function, equation 5.1. Estimated import levels, M, are then derived by using the maximum likelihood parameters of equation 5.9 in equation 5.1a. l”Since Canada adopted flexible exchange rates on September 30, 1950, the end of the third quarter, it is reasonable to assume that until the fourth quarter of 1950 expectations were always realized, i.e., Y = 1.0. Based on this assumption, r*§_1 of equation 4.2 is equal to 1.2791 generating an r for 1950 (IV) of 1.2460 and thereby allowing the r* series to be constructed. 15See Appendix B, Tables B2 to B17. 81 5.1a fit = -2086.6238 — 494.4801 rg + 2680.7834 r'dt + 0.2253 Yt - 21.0459 Q1 + 110.6218 02 t t - 198.6400 Q + e 3 3t t where 6t = Mt'Mt Figure 5.2 plots both the actual and estimated im- ports over the fifty-six quarter period, and clearly shows the disparity between actual and estimated imports in the post—May, 1962 period. The extrapolated import values are, by assumption, those values that would have occurred had Canada remained on a flexible exchange rate system, ceteris paribus, through 1964 (IV). If, as argued in Chapter IV16 uncertainty is a significant determinant of importing activity, one would expect, after an initial adjustment period, that import volume would increase over those levels associated with the assumed maintenance of the flexible exchange rate system. Notice that there is not any pronounced shift of the import demand function after May, 1962. Out of the total eleven quarter period after the return to a pegged rate, only five quarters registered increases in import volume. If the standard error of the estimate is used to establish confidence limits, the 95 per cent level would 16Chapter IV, pp. 60-61. 82 . Awhmfiaom cmapmcmo so mcoaflaazv .0>H00H00H AsHV 000H 00 AHV 0mmfi masonsH pmsmeusmm 0:0 as: new mead NomH Homfl owmm mmmfi wmma nmmm mmmn \ —-..—f. —.—~—_..—.—.—__._____._.hsr_—».—L—_ ..- iu< mmmm 0mmm ammfl - ---- y 261 ‘a New 21(2 | it> ;_ .— 9 ea paBSad on uanqeg g 83 TABLE 5.l.--Actual and Estimated Canadian Imports 1962 (II) to 1964 (IV) inclusive. Millions of Canadian Dollars Actual Estimated Residual Imports Imports Error 1962 II 2217 2236 - 19 111 2077 2438 -361 1v 2074 1986 88 1963 I 1864 1982 —118 II 2183 2292 -109 III 2155 2514 -359 IV 2283 2114 169 1964 I 2153 2139 14 II 2492 2466 26 III 2353 ‘ 2619 —266 IV 2507 2262 245 Source: Appendix B, Table B17. be approximately M t 208 or M : Can.$l98 million.17 This confidence band covers all but one of the cases where actual imports exceed the hypothesized flexible exchange rate import levels. If a four quarter adjustment period is allowed to enable importers to completely adjust to the 18 new pegged rate system over half of the remaining seven 17This is a far more stringent test of the shift than the standard error of the forecast, which would have a widening confidence band for each successive extrapolation period. See L. R. Klein, A Textbook of Econometrics (Evanston, Ill: Row, Peterson & Co., 1953), Chapter IV, VI. 18This interim transition period is not only to allow adjustments of the importer's utility function, but 84 quarters show the shift but again in only 1964 (IV) does the import level exceed the stringent confidence band used here as a guideline. The tightness of the observed shifts of import levels around the estimated import levels indi- cates that they are nothing exceptional and for the most part deviate by no more than what one would expect on a purely random basis. Thus, this testing procedure seems to indicate that, at least for the Canadian case, uncer- tainty brought on by flexible exchange rates had little effect on import trade volume. An alternative test must be made before any final conclusions regarding the hypothesis can be drawn. This analysis is based on interpreting a shift in the import demand function that results primarily from changes in Canadian exchange rate policy. To check on the validity of this ceteris paribus assumption, shift dummies are introduced for each independent variable and tested for statistical significance. A multiple regression was run on the following model for the full fifty-six quarter period: = x v Mt d + Blle + 82 r t + 83 Z1 + Bur dt + 8522 + B6Yt + 8723 + 8801 + 8902 + 81003 also to allow him to draw down inventory hoards of essen- tial imports no longer necessary under pegged rates. 85 where D = (9 1951( I) to 1962(I ), inclusive 1 i 1962(11) to 1964(IV), inclusive, and, Z1 2 D1r*t 22 = Dlr'dt 23 = DlYt 19 The results of this analysis are as follows: 5.1b Mt = —893.82 + (0)chl — 474.00r*t + 4278.49 Zl - (264.80) (2890.87) + 1584.76r'dt — 4264.94 Z2 + 0.20 Yt (2515.40) (3159.21) (0.02) - 0.04 Z3 + 1.84 Q1 + 146.92 Q2 - (0.05) (46.00) (47.24) - 96.91 03 . R2 = 0.8524 (47.50) SEE = 117.60 0w = 1.38 The shift dummy coefficients 81’ B3, 85 and B7 are all statistically insignificant indicating that the ceteris paribus assumption of the preceding analysis was valid despite the over twelve year period that the study covers. Since the structure of the model used here did not appear 19Johnston, op. cit., pp. 221-228. 86 to shift, more weight has to be put on the conclusions discussed above. It is to these conclusions and some necessary qualifications that are required that we now turn. Summary and Conclusions This study has investigated the effects that uncer- tainty, engendered by a system of flexible exchange rates, has on the volume of aggregate Canadian imports. The general conclusion is that Canadian import volume was not appreciably affected by the fluctuating rates. The inquiry took place in two phases. The first phase used an itera- tive maximum likelihood process to find the "best fit" value of Y, the coefficient of expectations. Given the model formulated in this thesis, this value of Y is 0.6, a level acceptable on a priori grounds. Since this value is less than but approaching unity, where exchange rate expectations are always realized, as under pegged rates, it indicates that a moderate level of uncertainty did exist for the Canadian experience under flexible rates. The second phase of the analysis involved using the estimated parameters of the "best fit" equation, found in the first step in the structural import demand function. The coefficients were used to derive estimated import levels under the uncertainty conditions imposed by the 0.6 coefficient of expectations. Extrapolations of these 87 levels into the post-May, 1962 period provided further evidence upholding the hypothesis, namely, that while imports increased in four out of the last seven quarters, allowing for an adjustment period, all but one of these fell within the confines of a stringent confidence band. These findings are further strengthened by alternative tests indicating that the structural parameters of the model did not shift significantly. Thus, contrary to the assertions of the critics of flexible exchange rates, for the Canadian experience import volume was not appreciably affected by uncertainty arising from exchange rate vari- ability. While the Canadian experience can provide many in- sights into the pegged versus flexible exchange rate con- troversy, caution must be exercised in any attempted generalizations of these findings. It must be kept in mind that the Canadian experience was unique for several reasons, the two most important being that Canada alone Operated on flexible exchange rates in a world of pegged rates and she had economic ties with the United States. Any conclusions drawn ignoring these facts could only be considered as purely conjectural. The only plausible generalization that could be drawn from the Canadian experience is that it worked and provided an orderly mechanism for conducting all forms of international transactions, as long as it was allowed to 88 operate in accord with the dictates of free market forces, as in the 1951-1956 period. This knowledge alone is encouraging! APPENDICES APPENDIX A CANADIAN BACKGROUND STATISTICS 91 TABLE Al.—-Canadian official holdings of gold and foreign exchange, 1950—62 inclusive. (millions of US dollars) Quarter Year Jan-Mar Apr—Jun Jul-Sep Oct-Dec 1950 A Gold 501 506 533 569 B Foreign Exchange 671 699 1,006 1,227 1951 A 604 635 673 786 B 1,108 1,041 941 949 1952 A 856 877 878 879 B 936 939 972 976 1953 A 896 924 951 977 B 960 869 823 833 1954 A 1,007 1,028 1.046 1,066 B 837 812 860 865 1955 A 1,083 1,102 1,125 1,138 B 808 803 820 766 1956 A 1,124 1,102 1,106 1,109 B 762 792 799 828 1957 A 1,103 1,106 1,104 1,103 B 831 831 842 788 1958 A 1,092 1,083 1,080 1,077 B 781 823 844 860 1959 A 1,077 1,073 1,051 962 B 824 857 884 915 1960 A 951 925 901 998 B 916 868 930 959 1961 A 885 899 920 940 B 1,036 1,066 1,032 1,150 1962 A 959 848 682 702 B 842 792 1,623 1,892 Source: IMF, International Financial Statistics, Vol. IV, No. 6 (June, 1951), pp. 14-15, for 1950; UN Monthly Bulletin of Statistics, various issues for 1951- 1962. 92 TABLE A2.——Growth in real gross national product, 1950-1962. (millions of constant (1949) Canadian dollars) Gross Annual National Percentage Product Change 1950 17,471 10.7 Total Percentage Increases 1951 18,547 10.6 1950-62 62.2 1952 20,027 7.8 1950-56 36.3 1953 20,794 3.8 1956-62 18.1 1954 20,186 -3.0 1955 21,920 8.6 Arithmetic Mean of Annual 1956 23,811 8.6 Percentage Increases 1957 24,117 1.3 1950—62 4.9 1958 24,397 1.2 1950-56 6.7 1959 25,242 3.5 1956—62 2.8 1960 25,805 2.2 1961 26,468 2.6 1962 28,111 6.2 Source: Dominion Bureau of Statistics, Canada Year Book 1963-64 (Ottawa, Queen's Printer and Controller of Stationary, 1964), p. 1016. TABLE A3.-—Canadian unemployment, quarterly, 1956-1962. Year Quarter Annual Jan—Mar Apr—Jun Jul—Sep Oct-Dec Average 1956 5.9 3.4 2.0 2.7 3.4 1957 6.1 4.0 3.2 5.3 4.6 1958 10.1 7.1 4.8 6.3 7.1 1959 9.3 5.7 3.7 5.2 6.0 1960 9.4 6.7 5.1 6.8 7.0 1961 11.1 7.4 4.9 5.6 7.2 1962 8.8 5.7 4.2 5.3 5.9 Source: Compiled from monthly data from various issues of United Nations Monthly Bulletin of Statistics (New York: Office of the United Nations). 93 TABLE A4.——Canadian wholesale and retail price indices, quarterly, 1956-1962. (1953 = 100) Year Annual 7 —M1- Apr—Jun Jul—Sep Cot—Dec Average 1956 A 101 102 103 103 102 B 101 101 103 104 102 1957 A 10' 103 103 102 103 B 104 105 106 107 106 1958 A 10? 103 103 104 103 B 107 108 1‘8 109 108 1959 A 105 105 105 104 104 B 109 119 109 111 110 1960 A 304 105 104 104 4 B 110 ‘10 111 112 111 1 61 A 105 105 106 107 106 B :12 :1 112 12 112 1962 A 107 108 109 110 108 B -_i 112 113 114 112 “;rp:_lei from m nthly dati from Naticns Mo onthly Bulletin of :Latistical C)ffice of tUe United price index, B — retail price _"‘§¢a.' ... '47.— 94 000 0m mm :m 00 om 00 0000000000 00 I 0 I 00 I 0 I 0m III m I .000 .00000>0< mpmcmo 0o pomscpw>ou mo COHpQHLomQZQ Hmufiamu 62w mcmoq me I :m I OH I mm I m o m mmwpfihzomm cmwmhom Ca pcmEpm®>CH 00m 000 000 000 00m 000 0000000>00 000000 002 000 I 00 I 00 I 00 I 00 000 I 000000000 >0 000000 000 000 000 000 000 mmm 000000000I000 00 000000 00 0cmspmm>cH 000009 mpcmEm>oS 0000qmo E009 mcoq 0200000 0000000 :00 I 000 I m00.0I 000.0I 000.0I 000.0- 050.0I 0000000 0000000 0000000 000.0I 000.0I 000.0I 000.0I mm: 000 I 000 I 000000000000 ucmhhjo meuc CO moCMHmfl 000.0 000.m m00.0 .0 0mm.“ 0; 000.0 00000 000 0 000.0 000.0 _m 000 000 000 000000000 0000000 00000 000 mom 000 0mm 000 000 000 000 0 00000000 000 0000000 000 000 mm0 000 000 000 mam 000000>00 000 00000000 000 000 000 000 000 000 000 000000000000 00>000 mpcszwL m00.m 000.0 m00.0 000.0 000.0 000.0 00000 0m0 000 000 000 000 000 000000000 0000000 00000 000 000 000 000 000 000 00: 00000000 000 0000000 000 000 m00 000 000 000 000000>00 000 00000000 mom 000 000 00m mom 0mm 000000000000 00>000 . mpafimomx 0:00pommcmse 0:00030 00:00 000 m00 000 I 000 I 000 0-0 I .00 I 00000 00000000002 00 0000000 m00.0 000.0 000 m 000 0 000.0 000 m 000 0 0000000 00m.0 000.0 00m.m 000.0 000.0 000.0 am0.0 0000000 000.55 mmficcmzohmz 0000000 0200000 0000 0000 0000 0000 0000 0000 0000 .AmLmHHoo cwfiomcmo mo 000000000 0000I0m00 .00000000 00 0000000 00000000II.m0 00000 95 .mmo0pm 0000000 000: 00 000:00 :0 000000000230 0: 00:05000 050000> w:00500:0 0:000000:m0u 00:00 0:0 0000>000 000:0052 00:50000 m::000>0a 0:0 000000:H 0:0 :0 000000 00: 0000000 0:0 0500:0 w:00300:0 00:05000 0:00:0H000005 mmcoflumuficmw0o Hm:OHum:0mu:H :0 00:05000000 00:000:505 0:0 Umo0nm 000sp0o:0ax0 20000005 0:0 :00000:0000a00 00000000 m:0::ao:0 w:oapomw:00p p:05:00>ow ”00:30 .mp:00w050 0:0 000:00000::0 00czaocfi 00:05000 0:00030 00:00 000m .0000>00m 00:00000 0:0 00:000000000 .mmm:0w:: 00 0000000000 0H0000: 000 :00:3 0:00000 Im:00u 00:00 0:0 0000>000 000:0039 0000:30000 000000 00: 0000000 0:0 0500:0 00 m00umcm0p w:00300:0 0500:0 0300:00000005 ”00:00: 0000005500Ico: m 00 000:0000500 Hm:ofiu:ufipm:0 0:0 00:00000 00:0:00 :0 00:05:00020000 20000005 0:0 0000500000 w:0:000:005 000 mp:05:00>om :w00000 a: 09:05000 00:00:0 :00:3 0:00:000:000 p:05:00>om 000::0 .mp:00M0550 0:0 000:00 I000::0 .00oax0 000 0090000>0 :000030000 000w ”00:00:0 0000000: 0:00030 00:00 HH02 0000000 002 000 I 000 I am 00 000 I 000 00 I 00000002 00000000 00 000000 002 0 II II II II II 00:000000< 0000:0:00 00:00pm:000:H Hmfio0am 00:00 mwm Hm I II mm I II II m0 I :0000000 0:50 00000:oz .Hmcofiumc00ucH 002 :0 0w:m:o 0mm I 000 I am 00 000 I 000 mm I 00000000 00 000000 0m:m:0xm :w000om 0:0 0000 00 mm:0000: 00000000 Hmm mmm mom mnm mmH mm OH I mucwEm>oZ HMuHQmo €008 phonm H0908 00m mmm Hma mmm mm mm 00 mu:050>oz 0000000 00:00 ma I mm I mmH ma mofi :m I om I m0mcw000om Mo ww50vao: 00000: :mflomcmo :0 0w:m:o 00:050>oz amuHQMQ 5009 p0o:m mmm 000 000 m00.0 000.0 00m.0 000.0 0000E0>02 0000000 0000 0000 00000 000 I NMH mm mm 000 m0 mmH .0.0.: I .m:oauomw:m0e 0000000 5005 m:oq 96 TABLE A6.-—Sales of new Canadian security issues to non- residents, 1956-1962 (millions of Canadian dollars) 1956 1957 1958 1959 1960 1961 1962 Bonds and Debentures Government of Canada 9 16 76 56 3O 37 155 Provincial govern- ments 22“ 136 168 334 103 66 1M8 Municipal governments 112 123 1&8 158 133 M7 74 Corporations 252 462 2M2 112 155 343 331 Total 597 737 63k 660 U21 M93 708 Common and Preferred Stocks 7O 61 U3 M7 26 “5 20 Total' 667 798 677 707 uu7 538 728 Source: Dominion Bureau of Statistics, Sales and Purchases of Securities Between Canada and Other Countries, Monthly, Cat. No. 67-002. APPENDIX B RAW DATA 9&3 .cso uoc Hafiz coammmpmop mooou mocowopoom unmeaspmpmo o.a Ass.:mmmv Azm.mzv Asz.mqv Am©.msv Amo.ov AH:.MHNNV Amfl.ommv swam.mzmm :Haom. mzam.mmmz+ mmmw.fiomn msmm.:fia+ wamm.mH| momm.o+ wom~.oo~m+ smmm.mnmn omw~.mwmml m.o ‘ Aam.mmmmv Am:.msv Aom.m:v Amm.:zv Amo.ov “mo.zammv Aom.mmmv mmma.mw~m mommw.. mumm.wafiz+ mms>.oomn mmmo.maa+ moma.ma| mmmm.o+ msa:.mz~m+ Omma.mazl mmom.owmmn m.o Aeo.ocmmv Ao:.m:v Amfl.m:v Aws.mev Amo.ov Amo.aommv Amm.memv :HNH.Hmwm ammom. momfi.~mmm+ Hmow.mmal :mow.HHH+ momm.oH| msmm.o+ :wmo.:anm+ mmmm.mmzn mwm~.mmamn w.o Amm.mmmmv Azm.m:v Amm.mqv mma.:sv Amo.ov Aom.mmamv Am:.wmmv mmmm.monm Homwm. mama.:asm+ oo:©.m0HI mamo.oafl+ cmzo.am| mmmm.o+ :mmn.om©m+ Hom:.:mzn mmmo.mmoml 0.0 “mo.mommv Aww.msv Awm.mzv Aom.::v Amo.ov Aza.mmamv Ama.mmmv owao.mmum wwmmw. waoz.zmmm+ momm.wmai mma~.moa+ o:ma.mmn Hmmm.o+ msom.mmom+ oomw.zmmn owoo.nmmfin m.o Amm.azmmv AHm.osv Am:.msv Amm.:sv Amo.ov Amw.mfimmv Aom.ommv - mamm.mm~m mmaom. nm:w.:mmmu mmmm.mmat sumo.moan m:ma.mm| :0mm.o+ memo.mmmm+ mazm.zmmu :mmo.mwmfin 3.0 Amm.mozmv Am©.~:v Azm.m:v Aom.mav Amo.ov Azo.m:mmv Amm.mozv oamm.m>mm mwoow. m:mm.>zam+ meam.amau mmaw.moa+ >o:m.:mu Hmam.o+ mmma.onzm+ mmm>.mmol mem.awoal m.o ‘ Am~.mommv Amm.mzv Amm.mz, Amm.wzv Amo.ov Amm.mmmmv nom.mmov mm»a.>~mm mwoow. mmmn.m:om+ wwmm.mmau Fm::.moau 22m2.0ml mmom.o+ mmmo.~amm+ oomo.oo>| sawm.oomau m.o can no: Hafiz scammmpwmp mopom monomopaqw pcmcHEpmme H.o mm mm 6A»-Hvsm+ mommu momm+ Hosmu pwmm+ po.LNm+ »\sam- n a: » .mpHSmms cofimmmpwmp noozaamxfla Esefixme o>aumpmqul.Hm mqmda 99 TABLE B2°--Imports of goods and services are adjusted for seasonal variations. (Millions of current Canadian dollars) Quarters I II III IV 1950 1,137 1,332 1951 1,259 1,563 1,484 1,307 1952 1,202 1,368 1,353 1,477 1953 1,335 1,597 1,496 1,415 1954 1,264 1,478 1,379 1,453 1955 1,397 1,605 1,665 1,776 1956 1,709 2,071 1,942 1,993 1957 1,841 2,131 1,968 1,873 1958 1,666 1,945 1,852 1,960 1959 1,761 2.177 2,091 2,102 1960 1,902 2,167 2,031 2,060 1961 1,900 2,157 2,165 2,265 1962 2,047 2,425 2,278 2,283 1963 2,071 2,436 2,442 2,593 1964 2,463 2,866 2,689 2,850 Source: Dominion Bureau of Statistics (DBS) National Accounts: Income and Expenditures by Quarters Table 2, line 11, various issues.. 100 TABLE B3.--Gross national product. Current dollars, unadjusted for seasonal variations. Quarters Year I 11 III IV 1950 5,350 4,721 1951 4,452 5,100 6,328 5,290 1952 5,129 5,739 5,171 5,956 1953 5,461 6,036 7,389 6,134 1954 5,526 6,082 6,939 6,324 1955 5,806 6,604 7,813 6,909 1956 6,564 7,327 8,826 7,826 1957 7,175 7,856 8,894 7,984 1958 7,185 8,104 9,170 8,435 1959 7,688 8,538 9,661 8,897 1960 8,152 8,713 9,963 9,100 1961 8,144 9,010 10,118 9,572 1962 8,986 9,863 11,313 10,399 1963 9,575 10,415 12,079 11,171 1964 10,444 11,511 12,903 12,145 Source: DBS, National Accounts: Income and Expenditures, by quarters, Table 1, lines 9 and 13. 101 TABLE B4.--Canadian domestic price index. deflator GNP. Implict price by quarters, Table 19, line 12. (1957 = 100) Quarters Year 11 III IV 1950 78.6 80.3 1951 82.2 84.9 88.0 90.2 1952 90.6 90.8 90.1 90.5 1953 90.8 90.3 91.2 91.2 1954 92.2 92.9 93.5 93.9 1955 93.0 9.30 93.3 94.7 1956 95.4 96.6 97.4 98.8 1957 99.3 99.8 100.4 100.5 1958 100.9 102.0 101.8 102.8 1959 103.6 104.1 104.8 105.6 1960 105.4 105.9 106.0 106.5 1961 106.7 106.6 106.6 106.9 1962 107.6 108.1 108.7 109.2 1963 109.6 110 1 110.4 111.1 1964 111.6 112.3 113.3 113.8 Source: DBS, National Accounts: Income and Expenditures, 102 TABLE B5.—-Foreign price index. Implicit price deflator, Imports of goods and services. (1957 = 100) Quarters Year I II III IV 1950 91.2 92.9 1951 97.9 102.7 102.4 100.1 1952 95.6 92.8 91.6 92.2 1953 92.4 93.1 93.9 93.7 1954 92.9 93.8 94.0 93.3 1955 93.3 93.5 94.1 96.4 1956 97.2 97.3 97.6 _ V 97.5 1957 98.9 99.8 100.4 101.0 1958 102.1 101.1 101.3 101.5 1959 101.1 101.1 100.9 100.6 1960 101.0 101.8 102.3 102.5 1961 103.6 103.3 105.9 107.2 1962 108.2 109.4 109.7 110.1 1963 111.1 111.6 113.3 113.6 1964 114.4 115.0 114.3 113.7 Source: DBS, National Accounts: Income and Expenditures, by quarters Table 19, line 11. 103 TABLE B6.—-Canadian short term interest rate three month treasury bill rate--quarterly, mid—month quotations. Year I II III IV 1950 0.56 0.62 1951 0.70 0.75 0.81 0.91 1052 0.91 1.02 1.11 1.25 1953 1.44 1.58 1.82 1.90 1954 1.74 1.58 1.30 1.14 1955 1.03 1.33 1.65 2.45 1956 2.58 2.71 2.90 3.51 1957 3.72 3.77 3.88 3.67 1958 2.79 1.61 1.54 3.07 1959 3.73 4.90 5.58 4.98 1960 4.45 3.04 2.53 3.23 1961 3.15 3.06 2.50 2.59 1962 3.10 3.64 5.22 4.05 1963 3.71 3.38 3.56 3.64 1964 3.82 3.66 3.73 3.76 Source: International Monetary Fund (IMF) International Financial Statistics, various issues. 104 TABLE B7.-—Foreign short term interest rate, U.S. three month treasury bill rate quarterly, average daily quotations. Quarters Year I II III IV 1950 1.23 1.35 1951 .40 1.53 1.62 1.65 1952 .64 1.67 1.83 1.92 1953 .04 2.20 2.02 1.48 1954 .08 0.81 0.87 1.03 1955 .26 1.51 1.86 2.35 1956 .38 2.60. 2.60 3.06 1957 .17 3.16 3.38 3.34 1958 .84 1.02 1.71 2.79 1959 .80 3.02 3.55 4.30 1960 .94 3.09 2.39 2.36 1961 .38 2.32 2.32 2.48 1962 .74 2.72‘ 2.86 2.80 1.963 .91 2.94 3.28 3.50 1964 .54 3.48 3.50 3.68 Source: International Financial Statistics, various issues. Ila—— 105 TABLE B8.--Spot exchange rate, U.S. dollars per Canadian dollar. (Average Noon rate) Quarters Year I II III IV 1950 .9070 .9533 1951 .9529 .9396 .9459 .9614 1952 .9994 1.0190 1.0120 1.0228 1953 1.0308 1.0177 1.0294 1.0320 1954 1.0246 1.0177 1.0294 1.0320 1955 1.0246 1.0148 1.0144 1.0019 1956 1.0012 1.0089 1.0200 1.0350 1957 1.0434 1.0459 1.0499 1.0328 1958 1.0193 1.0348 1.0341 1.0332 1959 1.0304 1.0398 1.0477 1.0529 1960 1.0506 1.0257 1.0274 1.0212 1961 1.0102 1.0058 .9689 .9651 1962 .9545 .9317 .9279 .9292 1963 .9281 .9282 .9252 .9275 1964 .9258 .9252 .9269 .9306 Source: Bank of Canada. Statistical Summary, Supplement various issues. Above is inverse of reported Lates. . -“.‘ awn-«4"» . '5. - ,' ”Hm . - ‘H Wm...» ‘95,... F J. TABLE B9.-—Forward exchange rate, U.S. dollar per Canadian dollar. (Average Noon Rate) 106 Quarters Year I II III IV 1950 .9577 .9577 1951 .9565 .9429 .9494 .9638 1952 .9984 .0171 .0354 .0276 1953 .0200 .0076 .0081 .0190 1954 .0262 .0128 .0271 .0298 1955 .0243 .0150 .0148 .0017 1956 .0008 .0075 .0171 .0302 1957 .0386 .0427 .0460 .0288 1958 .0167 .0334 .0337 .0334 1959 .0287 .0357 .0422 .0496 1960 .0493 .0260 .0275 .0205 1961 .0097 .0049 .9687 .9656 1962 .9542 .9306 .9234 .9269 1963 .9264 .9278 .9250 .9275 1964 .9257 .9255 .9269 .9302 Source: Bank of Canada. Statistical Summary, SUPPleEEflE: various issues; above is inverse of reported rates. 107 TABLE B10.--Adjusted spot exchange rate, U.S. dollars per Canadian dollar. Quarters Year I II III IV 1950 .9603 1951 .9596 .9468 .9535 .9684 1952 .0066 1.0255 .0192 .0296 1953 .0369 1.0239 .0315 .0278 1954 .0179 1.0100 .0251 .0309 1955 ..0270 1.0166 .0165 .0009 1956 .9992 1.0078 .0170 .0306 1957 .0378 1.0397 .0449 .0295 1958 .0099 1.0288 .0359 .0304 1959 .0211 1.0212 .0276 .0461 1960 .0455 1.0261 .0260 .0126 1961 .0071 1.0063 .0016 .0017 1962 .0032 1.0078 .0180 .0097 1963 .0059 1.0039 .0025 .0013 1964 .0026 1.0021 .0022 .0003 Source: Tables B6, B7, BB; derived from 1 + if r = r ————r— a s 1 + 1c '-_— 108 TABLE B11.—-Forward/spot exchange rate differential. Quarters Year 1 11 111 1v 1950 .9973 1951 .9967 .9959 .9957 .9952 1952 .9919 .9918 1 0159 .9980 1953 .9837 .9841 .9773 -9914 1954 1.0082 1 0028 1.0020 .9989 1955 .9974 .9984 .9983 1.0008 1956 1.0016 .9997 1.0001 .9996 1957 1.0007 1.0029 1.0011 .9993 1958 1.0067 1.0045 .9979 1.0029 1959 1.0074 1.0142 1.0142 1.0033 1960 1.0036 .9999 1.0016 1 0078 1961 1.0071 1 0063 1 0016 1.0017 1962 1.0032 1 0078 1.0180 1.0097 1963 1.005 1 0039 1.0025 1.0013 1964 1.0026 1 0021 1.0022 1.0003 Source: Tables B9 and B10 derived from rd = r1 109 TABLE B12.--Incremental forward/spot exchange rate differ- ential. Quarters Year I II III IV 1950 .9994 1951 .9994 .9992 .9998 .9995 1952 .9967 .9999 .0243 .9824 1953 .9857 .0004 .9931 .0144 1954 .0169 .9946 .9992 .9969 1955 .9985 .0010 .9999 .0025 1956 .0008 .9981 .0004 .9995 1957 .0011 .0022 .9982 .9982 1958 .0074 .9978 .9934 .0050 1959 .0045 .0068 .0000 .9892 1960 .0003 .9963 .0016 .0063 1961 .9993 .9992 .9953 .0001 1962 .0015 .0046 .0101 .9918 1963 .9962 .9980 .9986 .9988 1964 .0013 .9995 .0001 .9981 rdt Source: Table 811 derived from r'd = rd TABLE Bl3.——Import volume, imports of goods and services, unadjusted for seasonal variation. (Millions of constant Canadian do1lars) 110 Quarters Year I II III IV 1950 1,025 1,292 1,247 1,434 1951 1,286 1,522 ,1449 1,306 1952 1,257 1,474 1,477 1,602 1953 1,445 1,715 1,593 1,510 1954 1,361 1,577 1,467 1,557 1955 1,497 1,717 1,769 1,842 1956 1,758 2,128 1,990 2,044 1957 1,861 2,135 1,960 1,854 1958 1,632 1,924 1,828 1,931 1959 1,742 2,153 2,072 2,090 1960 1,883 2,129 1,985 2.010 1961 1,834 2,088 2,044 2,113 1962 1,892 2,217 2,077 2,074 1963 1,864 2,183 2,155 2,283 1964 2,153 2,492 2,353 2,507 Source: Tables B2 and B5; derived from Mv 111 TABLE Bl4.—-Gross national product, Constant dollars, unadjusted for seasonal variation. (Millions of Canadian Dollars) Quarters Year I II III IV 1950 4,953 5,444 6,807 5,879 1951 5,416 6,007 7,190 5,864 1952 5,661 6,320 5,739 6,581 1953 6,014 6,684 8,102 6,726 1954 5,993 6,547 7,421 6,735 1955 6,243 7,101 8,374 7,296 1956 6,881 7,585 9,062 7,921 1957 7,226 7,872 8,859 7,944 1958 7,121 7,945 9,008 8,205 1959 7,421 8,202 9,219 8,425 1960 7,734 8,228 9,399 8,545 1961 7,633 8,452 9,492 8,954 1962 8,351 9,124 10,408 9,523 1963 8,682 9,460 10,941 10,055 1064 9,358 10,250 11,388 10,672 Source: Tables B3 and B4; derived from Y = $3: . C TABLE BlS.—-Index of relative prices adjusted for exchange 112 rates. Quarters Year I II III IV 1950 1951 1.2499 1.2874 1.2302 1.1543 1952 1.0558 1.0030 1.0046 .9961 1953 .9872 1.0131 1.0002 .9956 1954 .9834 .9921 .9766 .9628 1955 .9791 .9907 .9943 1.0160 1956 1.0176 .9984 .9824 .0535 1957 .9545 .9561 .9525 .9731 1958 .9927 .9578 .9623 .9556 1959 .9471 .9340 .9190 .9048 1960 .9121 .9372 .9394 .9425 1961 .9611 .9635 1.0253 1.0391 1962 1.0535 1.0864 1.0872 1.0847 1963 1.0924 1.0920 1.0979 1.1018 1964 1.1074 1.1068 1.0886 1.0736 Source: Tables B4, B5 and B8; derived from r/ = EE- . rPc TABLE B16.-—Expected exchange index. 113 Quarters Year 11 111 IV 1950 2460 1951 2718 2468 1913 1952 0458 .0211 0061 1953 .0057 1 0024 .9983 1959 .9910 .9824 9706 .955 .9847 9904 0058 1956 .0042 9911 .9685 1957 .9577 .9546 0657 1958 .9675 9644 .9591 1959 .9412 9278 9140 1960 .9276 9346 9393 1961 .9490 9988 0230 1962 1 .0848 0688 0784 1963 .0899 0942 0987 1964 1.1057 0054 0823 Souece: Generated by solving equation 4.21 for 0.6. See Chapter V, fgotnote 14. .ma.m coapmzum 80pm U0>Hhm© mmmeHpmm ocm mam manme "mopsom mam mmmm momm mmm- mamm mmmm mm mmam mmam aa mmam mmam amma mma aaam mmmm mmm- aamm mmam moan mmmm mmam maau mmma amma mmma mm mmma amom ammu mmam mmom ma 1 mmmm mamm mm . aama mmma mmma m moam maam ma mmom aaom ma u mmam mmom aa omma amma amma mm 1 mmom oamm mm . mmom mmma mm omom mmam ma mmma mmma mmma mm aaom omom mm aaom mmom ma moam mmam aa I mmma mama mmma am 1 mmma amma moa- mmma mmma mm 1 oaom amma om . mama mmma mmma ma 1 mmma amma ma mama mmma ama aaom mmam mma mmma amma mmma maa mmma aaom ma mama mmma mam mama mmam mma mmma mmma mmma 1. mm mama mama mm 1 mmma mmma moan mmma mama a mmaa mmaa mmma H mm . mmma mmma maa- mmma mmaa mm . mmma mmma maa- mmaa amma amma aaa- amma oama aaa- mmma mmma a aama mama ma mmma maaa mmma om mmma mmma mmm mama mmaa mma- aama amaa am I amma mmma mmma ma . mmma mmma am mmma maaa mm mmaa mmma aaa mmaa mmma amma w SH 2 w E Z w Na 2 w mm 2 >a aaa aa a hmpthG ampmaaoo cmammcmo mo mcoaaaazv .msamaaoca A>avamma om aavamma mpsomsa cmammcmm mmmeapmm 66m amapoauu.mam mamas BIBLIOGRAPHY BIBLIOGRAPHY Articles and Periodicals Aliber, R. Z. "Speculation in the Foreign Exchanges: The *1 EurOpean Experience, 1919-1926." Yale Economic Es— ' says, II (Spring, 1962), 171-245. Artus, R. E. "Canada Pegs Its Dollar." The Banker, CXII, (June, 1962), 362-369. i Balogh, T., and Streeten, P. P. "The InapprOpriateness of Simple 'Elasticity' Concepts in the Analysis of International Trade." Bulletin of the Oxford Uni— L» versity Institute of Statistics, XIII (April, 1951), 65-77. Baumol, W. J. "Speculation, Profitability, and Stability." Review of Economics and Statistics (XXXIX (August, Binhammer, H. H. "Canada's Foreign Exchange Problems: A Review." Kyklos, XVII (No. 4, 1964), 636—652. Brown, A. J. "Trade Balance and Exchange Stability." Oxford Economic Papers, VI (April, 1942), 101-108. Caves, R. E. "Flexible Exchange Rates." American Economic Review, LIII (May, 1963), 120—129. Durban, J, and Watson, G. S. "Testing for Serial Correla- tion in Least-Squares Regression." Pts. I and II. Biometrica (1950-51), 409—428. Eastman, H. C. "Aspects of Speculation in the Canadian Market for Foreign Exchanges." Canadian Journal of Economics and Political Science, XXIV (August, 1958), 355’3720 Ellsworth, P. T. "Exchange Rates and Exchange Stability." Review of Economics and Statistics, XXXII (February, 1950), 1—12. Fleming, J. M. "Domestic Financial Policies Under Fixed and Under Floating Exchange Rates." IMF Staff Papers, IX (November, 1962), 369-380. 116 117 Haberler, G. "The Market of Foreign Exchange and Stability of the Balance of Payments." Kyklos, III (Fash. 3, lgug)’ 193—218. Harrod, R. F. "Convertibility Problems." Economia Inter- nazionale, XIII (February, 1955), 20-34. Helleiner, Gerald K. "Connections Between United States' and Canadian Capital Markets, 1952—1960." Yale Economic Essays, II (Fall, 1962), 398—400. Ingram, James C. "The Canadian Exchange Rate, 1950-57." Southern Economic Journal, XXVI (January, 1960), 207-218. Katz, Samuel I. "Le dollar canadien et le course de change fluctuant." Bulletin d'Information et de Documenta— tion de la Banque Nationale de Belguique. 30th year, I (May, 1955), 7-8. . "The Canadian Dollar: A Fluctuating Currency." Review of Economics and Statistics, XXXV (August, 1953), 236—237. Klein, L. R. "The Estimation of Distributed Lags." Econometrica, XXVI (October, 1958), 553-565. Laursen, 8., and Metzler, L. A. "Flexible Exchange Rates and the Theory of Employment." Review of Economics and Statistics, XXXII (November, 1950), 281-299. Lutz, F. "The Case for Flexible Exchange Rates." Banca Nazionale del Lavoro Quarterly Review, VII (December, 1954). 175—I85. Meade, J. G. "The Case for Variable Exchange Rates." Three Banks Review, XXVII (September, 1955), 3-27. Morgan, E. V. "The Theory of Flexible Exchange Rates." American Economic Review, XLV (June, 1955), 279-295. Mundell, Robert A. "Flexible Exchange Rate and Employment Policy." Canadian Journal of Economics and Political Science, XXVII (November, 19615, 509-517. . "Monetary Dynamics of International Adjustment Under Fixed and Flexible Exchange Rates." Quarterly Journal of Economics, LXXIV (May, 1961), 227-257. "Problems of Monetary and Exchange Rate Management in Canada." The National Banking,Review, XI (Sep- tember, 1964), 77-86. 118 Nerlove, M. "The Market Demand for Durable Goods: A Comment." Econometrica, XXVIII (January, 1960), 132-142. Orcutt, G. "Measurement of Price Elasticities in Inter- national Trade." Review of Economics and Statistics, XXXII (May, 1950), 117-132. Penner, Rudolph G. "Inflow of Long—term Capital and the Canadian Business Cycle, 1950-1960." Canadian Journal of Economics and Political Science, XXVIII (November, 1962), 527-542. "fll‘fil' I‘. I Pippenger, John. "The Canadian Experience with Flexible Exchange Rates." American Economic Review, LVII (May, 1967), 545-555. S Poole, W. "The Stability of the Canadian Flexible Ex- change Rate." The Canadian Journal of Economics, XXXIII, No. 2 (May, 1967), 205-217. "‘ ...-”VII 1' .o \ . Rhomberg, R. R. "A Model of the Canadian Economy Under Fixed and Fluctuating Exchange Rates." Journal of Political Economy, LXXII (February, 1964), 1-29. . "Canada's Foreign Exchange Market: A Quarterly Model." IMF Staff Papers, VII (April, 1960), 439- L856 0 Stone, R., and Rowe, D. A. "The Market Demand for Durable Goods." Econometrica, XXV (July, 1957), 423-443. Telser, L. G. "A Theory of Speculation Relating Profit— ability and Stability." Review of Economics and Statistics, XLI (August, 1959), 295-301. Tsiang, S. C. "An Experiment with a Flexible Exchange Rate System: The Case of Peru, 1950-54." IMF Staff Papers, V (February, 1957), 449-476. "Fluctuating Exchange Rates in Countries with Relatively Stable Economics: Some European Exper- ience After World War I." IMF Staff Papers, VII (October, 1959), 244-273. Yeager, Leland. "Some Facts About the Canadian Exchange Rate." Current Economic Comment, XX (November, 1958). 39-54. 119 Books and Reports Brewis, T. N., et a1. Canadian Economic Policy. Toronto: The Macmillan Company of Canada Ltd., 1961. Christ, C. F. Econometric Models and Methods. New York: John Wiley and Sons, Inc., 1966. Clement, M. O.; Pfister, R. K.; and Rothwell, K. J. Theoretical Issues in International Economics. Boston: Houghton Mifflin Co., 1967. Cornell, Peter M. "Exchange Flexibility in Canada: Some Underlying Factors." Public Policy. A Year Book of the Graduate School of Public Administration, Harvard University, 1958. Dhrymes, P. J. "Efficient Estimation of Distributed Lags with Autocorrelated Error Terms." Discussion Paper No. 23, Department of Economics, University of Pennsylvania, 1968. Friedman, Milton. Essays in Positive Economics. Chicago: University of Chicago Press, 1953. Friedman, Milton, and Roosa, R. V. The Balance of Pay— ments: Free Versus Fixed Exchange Rates. Washington, D.C.: American Enterprise Institute, 1967. Haberler, G. Currency Convertibility. No. 541 in the series "National Economic Problems." Washington, D.C.: American Enterprise Association, 1954. Hicks, John R. Value and Capital. 2d. ed. London: Oxford University Press, 1946. Johnson, Harry G. The Canadian Quandary: Economic Prob— lems and Policies. Toronto: McGraw-Hill Co. of Canada Ltd., 1963. Johnston, J. Econometric Methods. New York: McGraw- Hill, 1963. Kemp, M. The Demand for Canadian Imports 1926-1955. Toronto: University of Toronto Press, 1962. Kindleberger, C. E. "Flexible Exchange Rates." Monetary Management. Prepared for the Commission on Money and Credit, 1963. Kindleberger, Charles P. International Economics. 4th ed. Homewood, Illinois: Richard D. Irwin, Inc., 1968. Klein, L. R. A Textbook of Econometrics. Evanston, Illinois: Row, Peterson & Co., 1953. Meade, J. E. The Balance of Payments and Mathematical Supplement. London: Oxford University Press, 1951. Nerlove, Marc. Distributed Lags and Demand Analysis for Agricultural and Other Commodities. USDA Agricultural m Handbook No. 141, Washington, D.C., 1958. Neufield, E. P. Bank of Canada Operations and Policy. Toronto: University of Toronto Press, 1958. Nurkse, Ragnar. International Currency Experience. Geneva: League of Nations, 1944. Officer, L. An Economic Model of Canada Under the Fluctuat— — ing Exchange Rate. Harvard Economic Series. Cam— bridge: Harvard University Press, 1968. Samuelson, P. A. "Disparity in Postwar Exchange Rates." Foreign Economic Policy for the United States. Edited by S. E. Harris. Cambridge, Mass.: Harvard University Press, 1947. Scammell, W. M. International Monetary Policy. 2d ed. London: Macmillan, 1961. Sohmen, Egon. Flexible Exchange Rates - Theory and Con- troversy. Chicago: University of Chicago Press, 1961. International Monetary Problems and the Foreign Exchanges. Princeton, N.J.: Princeton University International Finance Section, Special Papers in International Economics, No. 4, 1963. Stackelberg, H. von. "Die Theorie des Wechselkurses bei vollstandiger Konkurrens." Jahrbucher fur National— Okonomie und Statistik, CLXI—(1949), 1—65. Trans— lated in "international Economic Papers," No. 1, pp. 104—159. London and New York: Macmillan, 1951. Stuvel, G. The Exchange Stability Problem. New York: Augustus M. Kelley, 1951. Vanek, J. International Trade: Theory_and Economic Policy. Homewood, 111.: Richard D. Irwin, Inc., 19 2. 121 Watkins, M. H., and Forster, D. F., eds. Economics: Canada. Toronto: McGraw-Hill Company of Canada Ltd., 1963. Wonnacott, R. The Canadian Dollar 1948~1958. Toronto: University of Toronto Press, 1958. Yeager, L. B. International Monetary Relations. New York: Harper & Row Publishers, 1966. Young, J. H., and Helliwell, J. F. "The Effects of Mone— r7 tary Policy on Corporations." Royal Commission on ' ’Banking and Finance. Ottawa: 1964. Zellner, A., and Geisel, M. S. "Analysis of Distributed Lag Models with Applications to Consumption Function Estimation." Manuscript, Department of Economics, University of Chicago (January, 1968). 1’ Public Documents Bank of Canada. Statistical Summary, Supplement. Ottawa: Various issues. Dominion Bureau of Statistics. National Accounts. Income and Expenditures by Quarters. Ottawa: various issues. . Sales and Purchases of Securities Between Canada and Other Countries. Ottawa: Cat. No. 67—002. . The Canadian Balance of International Payments. Ottawa: Cat. Nos. 67-201 and 67-001. International Monetary Fund. International Financial Statistics. Washington, D.C.: various issues, Meade, J. E. "The Future of International Payments." Factors Affecting the United States Balance of Pay— ments. U. S. Congress, 87th Cong., 2d Sess., 1962. United Nations. Monthly Bulletin of Statistics. New York: Statistical Office of the United Nations, various issues 1950—1964. 4.! .9 vi .4: :15" - :F ..l .9... . .N ”"171711’77fl1177711717177“‘s 030