The Effects of Energy Price Changes on Commodity Prices, Interprovincial Trade, and Employment 9781487573713

Citation preview

THE EFFECTS OF ENERGY PRICE CHANGES ON COMMODITY PRICES, INTERPROVINCIAL TRADE, AND EMPLOYMENT

J. R. Melvin

The Effects of Energy Price Changes on Commodity Prices, lnterprovincial Trade, and Employment

PUBLISHED FOR THE ONTARIO ECONOMIC COUNCIL BY UNIVERSITY OF TORONTO PRESS TORONTO AND BUFFALO

©

Ontario Economic Council 1976 Printed in Canada Reprinted in 2018

Library of Congress Cataloging in Publication Data

Melvin, James R The effects of energy price changes on commodity prices, interprovincial trade, and employment. (Ontario Economic Council research studies; 3) Bibliography: p. 1. Petroleum industry and trade - Ontario. 2. Petroleum products - Prices Canada. 3. Interstate commerce - Canada. I. Ontario Economic Council. II. Title. III. Series: Ontario Economic Council. Ontario Economic Council research studies; 3. 76-27870 HD9574.C23052 338.2'7'2809713 ISBN 978-0-8020-3337-6 (paper)

This study reflects the views of the author and not necessarily those of the Ontario Economic Council.

Contents

1 Introduction 3 2

Interprovincial trade restrictions S

3 A general equilibrium analysis of energy costs 17 4 Provincial policy options 78

s

Conclusions 98 BIBLIOGRAPHY

105

THE EFFECTS OF ENERGY PRICE CHANGES ON COMMODITY PRICES, INTERPROVINCIAL TRADE, AND EMPLOYMENT

1 Introduction

OUTLINE OF THE STUDY

Canadian economists have always been concerned with the issue of free international trade and a number of studies have examined the effect of the removal of tariffs. As a recent example, the Economic Council of Canada has been involved in an extensive study of the possibilities of free trade between Canada and the United States. Much less attention, however, has been paid to restrictions on interprovincial trade, at least in part because of the fact that since provinces do not have the power to impose tariffs and since restraints on interprovincial trade by provincial governments are forbidden by the BNA Act, this has never been seen as a major problem. However, recent events and a better understanding of the effects of taxes on trade suggest that further research in this area will be necessary if Ontario is to avoid being placed in a disadvantageous position relative to other provinces in Canada. The present study will examine the effects of interprovincial trade restrictions with specific attention being paid to energy sources, particularly petroleum and natural gas. Chapter 2 will consider the broad question of interprovincial trade restrictions and will be specifically concerned with the income redistribution effects of a policy in which more than one domestic price for petroleum and natural gas exists, and an analogy will be drawn with tariffs used to restrict international trade. Specifically, we shall argue that since any tariff system can be duplicated by a system of taxes, any attempt by a provincial government to maintain different prices in the rest of the country can be viewed as an attempt to impose a tariff. We shall then examine the terms-of-trade effect of such a

4 The effects of energy price changes two-price system in order to investigate the income redistribution effects that such a system could have on the Canadian economy. We shall consider first the case where petroleum is assumed to be only a consumption good and then the more realistic possibility of petroleum products being both a consumption product and an intermediate good for use as an input in other industrial processes. Chapter 3 will continue the investigation of the effects of energy cost changes by considering a simple empirical model. Specifically, through the use of the Ontario and Canadian input-output tables, we shall attempt to investigate the over-all price effects of an energy cost increase. These price effects will be composed not only of the direct cost to each industry but also the indirect cost through interindustry flows. In the light of this analysis chapter 4 will discuss some of the policy options available to the Ontario government. Specifically, we shall argue that two kinds of policies are open. First, there are policies designed simply to counteract the effects of higher energy costs; second, we can conceive of policies of a retaliatory nature. Although in general such retaliatory policies would not be beneficial to the Canadian economy as a whole, it would seem useful to make it clear that provinces other than the oil-producing provinces may have a monopoly in some areas of interprovincial trade. The final chapter will present conclusions and make suggestions for possible future research.

2

I nterprovincial trade restrictions

INTRODUCTION

In recent years there have been a number of areas in which conflicts have arisen between provinces on questions of interprovincial trade. For example, there was a dispute between Manitoba and British Columbia over the marketing of fresh eggs. There have also been conflicts between eastern and western Canada with regard to the marketing of feed grains. Of much more significance, however, have been recent discussions regarding the possibility of a two-price system between eastern and western Canada for petroleum and natural gas. Both Alberta and Saskatchewan have expressed an interest in such a plan, and although a recent agreement has been reached between the provinces and the federal government regarding the marketing of petroleum, recent indications are that new conflict may arise regarding the marketing of natural gas. The recent international petroleum crisis has certainly improved the bargaining position of the oil-producing provinces, and the fact that this crisis seems likely to continue, or at least that prices are not likely to be reduced significantly in the next few years, underlies the need for an economic study of the consequences of such action. The form that these proposals has taken has varied among provinces, and in some cases precise policies have not been indicated. In general, however, they are intended to raise the price to other provinces while keeping the cost to the provincial users virtually unchanged. This, then, is a kind of provincial export tariff which is intended to improve the welfare of the petroleum-producing provinces by allowing them to take advantage of their monopoly trade position.

6 The effects of energy price changes They are attempting to use an export tariff to improve their terms of trade. In other words, they are attempting to impose an optimum tariff. There are two obvious effects of such a policy. First, the tax results in a transfer of income from non-petroleum-producing provinces to Alberta and Saskatchewan. Such a situation has an obvious analogy with the case in which, in the international sector, one country imposes a tariff against another. In the remainder of this chapter we use standard international trade tools to investigate the effects of such a two-price system . After reviewing the literature on taxes as a tariff, we consider the terms-of-trade effect of such a tariff imposition. The second effect of the two-price system would be the increase in production costs which would automatically take place in those provinces in which fuel prices are higher. Although it is clear that some cost increases will occur, their exact extent is not obvious. Concern has been expressed in several areas about how such higher fuel costs would be disadvantageous to the Ontario industrial sector. In a recent C.D. Howe Research Institute publication, for instance, Judith Maxwell suggests that 'this abrupt price increase will unquestionably create financial hardship for low- and middleincome families and for companies that are large oil consumers' (Maxwell, 1973, 7). Although Maxwell is referring to price differences which occur because of the Ottawa Valley line, her argument, if true , should apply to any situation in which different prices for petroleum products exist within the country . Later she observes that 'Canada's ability to guarantee manufacturers a stable supply of fuel at prices below the current international level may be an attraction to locate plants in Canada which will export to energy-short markets' (ibid., 13). One purpose of chapter 3 will be to determine exactly how much of a cost advantage such a two-price system for petroleum would be. Specifically, we shall attempt to estimate the commodity price effect of specific increases in energy costs for Ontario. These cost increases will then be used to calculate some preliminary employment changes under various assumptions about the trade sector. For comparison, similar calculations will be done for Canada and the United States. Most of the discussion of possible petroleum and natural gas price changes has been concerned with the possibility of a two-price system within Canada, the implicit assumption being that uniform price increases throughout Canada will be neutral interregionally. The empirical results do not support this conclusion, however, and it is found that, on account of differences in interindustry structures, the commodity price effects of a uniform increase in petroleum prices will differ across Canada . Similar differences are found between Canada and the United States.

Interprovincial trade restrictions 7 y

FIGURE I TAXES AS A TARIFF

The proposition that a system of taxes could be used to duplicate a tariff structure was first proved by Mundell (l 960). Intuitively, this proposition seems quite reasonable, for it is clear that a tariff system changes prices both for consumers and producers in the domestic economy vis-a-vis the rest of the world. It seems quite reasonable to suppose that a system of domestic taxes could bring about the same kind of distortions between domestic and foreign prices as is brought about by tariffs. This proposition for the two-good case is illustrated in Figure l. Here TT' represents the production possibility curve,A the free trade production point, and C the free trade consumption point. To simplify our discussion, we will assume that the terms of trade are given, in other words, that we are considering the small-country case. With free trade the world terms of trade and the domestic price ratio will both be equal to the slope of the line AC. Suppose we now impose a tax on commodity Y that is collected from domestic producers; we are therefore imposing a production tax on commodity Y . We further assume that imports of Y are not taxed, so that consumers are unaffected by the tax and can continue to buy commodity Y at the world terms of trade AC. Because the world terms of trade are given, the entire effect of the tax will be reflected in price received by producers. Thus production will shift to point A', implying a reduction in the production of Y and an increase in the production of X. The world terms of trade are now A'C', parallel to AC, and consumers, since the prices they face are not affected, will consume at C', the point where the highest indifference curve is tangent to the line A' C'.

8 The effects of energy price changes Suppose we now impose a consumption tax on commodity X exactly equal to the production tax that we applied on commodity Y . This tax will have the effect of shifting consumption away from X towards Y so that the new consumption point will be a point such as C" on the terms-of-trade line A' C'. Thus, in equilibrium, producers will be producing at A', consumers consuming at C", with the trade vector equal to A'C". But this situation is exactly equivalent to one in which we had imposed a tariff on commodity X equal to the tax. It is clear, therefore, that a tariff of t per cent on commodity X can be exactly duplicated by a production tax of t per cent on commodity Y and a consumption tax of t per cent on commodity X. The importance of this proposition for interprovincial trade is obvious. For while provinces do not have the authority to impose tariffs, they certainly do have the authority to impose taxes . Of course, the model that we have considered so far is not of much relevance for trade between provinces, for it is not reasonable to suppose that any particular province is so small relative to the rest of Canada that it faces fixed terms of trade. Indeed, it is clear that if any province did face fixed terms of trade vis-a-vis the other provinces in the country, it would not be in its interest to impose the kind of taxes indicated in Figure 1, because welfare has been reduced in going from C to C". Such taxes can be beneficial from the point of view of trade only if the combination of the two taxes succeeds in improving the terms of trade for the imposing province. Such a situation is shown in Figure 2, where again A and C represent the free trade production and consumption points respectively, and where A' and C" represent the production and consumption points after the imposition of the two taxes . It is clear that in this situation, C" represents a higher level of utility than does C and therefore the province has been made better off by the two taxes. 1 Note that the price line through C", assumed to be tangent to an indifference curve at that point, has the same slope as the price line which is tangent to the transformation curve at A' . We have so far considered two situations, one in which the tax-imposing country had no monopoly power in trade and another in which the tax-imposing province or country has some monopoly power in trade. The third possibility of

From an examination of Figure 2 it might appear that the production tax alone would have resulted in an even higher level of welfare since it is tempting to consider the point B, assumed to be a tangency between A'B and an indifference curve, as an alternative to C". Point Bis not possible, however, for if the consumption tax were removed, the terms of trade would change until a new equilibrium would be reached, which would be on a lower indifference curve than C" .

Interprovincial trade restrictions 9

FIGURE 2

y

FIGURE 3

some interest for our present discussion is the case in which the tax-imposing province has complete monopoly power. In this situation we would expect the domestic, that is, the provincial, price line to remain unchanged and the entire increase in cost to be passed along to the other provinces. Such a situation is shown in Figure 3, where again A and C represent the free trade production and consumption points. Now, with the imposition of the production tax on Y and the consumption tax on C, the production point, instead of moving to A' as in

10 The effects of energy price changes Figure 2, remains at A and the new terms of trade give rise to a consumption point C". It is clear that for the same level of taxes, C" represents not only a higher level of welfare than C, but also a higher utility level than the C" of Figure 2. This is not surprising, of course, but simply illustrates the fact that the larger is the monopoly power in trade, the larger the gains from the imposition of tariffs or, as in this case, taxes. We must keep in mind that the situation represented in Figure 3 is a very special case, for we are in fact assuming that the export good of this province, commodity Y, has a very low price elasticity of demand in the rest of the country. In other words, we are assuming that the other provinces are willing to buy almost the same quantity of commodity Y regardless of price. If this were taken to be true for all price ratios, then the implication would be that this province should charge an infinite tariff. This situation is just the opposite of the case where there is no monopoly power in trade and where the optimal tariff is zero. Of course, it is unreasonable to expect that the elasticity of demand for product Y is zero at all prices; nevertheless, it may be reasonable to suppose that for relatively small price changes the elasticity of demand is close to zero. Figure 3 is relevant, then, only for those cases in which the tax is small enough that it does not much affect total Canadian consumption. But what the actual quantitative limits on the size of the tax are is difficult to predict. It is conceivable, for example, that a 100 per cent tax might not much affect the amount of petroleum products that the rest of the provinces would use. We have now analysed the welfare and terms-of-trade effect of taxes from the point of view of the province imposing the tax, but at least as important for our purposes is the effect on the other provinces. From this point of view the cases of interest are those represented in Figures 2 and 3. In the situation as shown in Figure 1, we would need to assume that the oil-purchasing provinces were perfect competitors relative to the rest of the country, and this is clearly not a case of interest for Canada. The situation for the oil-purchasing provinces which corresponds to Figure 2 is represented in Figure 4. Here T0 T0 ' is the transformation curve, and we are assuming that the oil-importing provinces do have some capacity for the production of petroleum products. With free trade, production in these provinces is at Ao and consumption at C0 . The imposition of the tax shown in Figure 2 now gives rise to a change in the terms of trade so that the new production point is Ao' and the new consumption point Co'. The terms of trade line Ao' Co' in Figure 4 has the same slope as the line A'C" in Figure 2. It is clear that the tax imposed by the oil-exporting provinces has made the oil-importing provinces worse off, for in Figure 4 C0 ' clearly represents a lower level of utility thanC0 .

Interprovincial trade restrictions I I y

~------------~Ti:,--x FIGURE 4 y

FIGURE 5

One way to represent the situation corresponding to Figure 3 for the oil-importing provinces is to suppose that those provinces have no capacity to produce commodity Y assumed to be petroleum or natural gas. This means that the production possibility set for the oil-importing countries is the single point on the X-axis in Figure 5. The free trade consumption point would be 0 , and when the tax is imposed by the oil-exporting provinces the terms of trade would shift to with consumption at Note that with this change in the terms of trade the oil-importing province is now receiving less oil for a larger quantity of X than it did with free trade, and that this corresponds precisely with the situation represented in Figure 3.

To'

C

To'Co'

Co'.

12 The effects of energy price changes PETROLEUM AS AN INTERMEDIATE PRODUCT

In our discussion to this point we have been implicitly assuming that both commodities in our model have been consumer goods only. In other words, we have been implicitly assuming that petroleum products are used for consumption purposes only, used, for example, as fuel in the family automobile. While the pure consumption effects of changes in petroleum prices are certainly of importance, of at least as much significance are the effects that changes in petroleum prices have on production costs. We must therefore take account of the fact that petroleum is used not only as a final consumer product but also as an input into the production process of other commodities. This means that we must consider an intermediate good model of the kind discussed by, among others, McKinnon (1966), Vanek (1963), Melvin (1969), and Warne (1971). The kind of model we have been implicitly assuming so far is the traditional two-good, two-factor, model in which the production functions can be represented as

X = f/Kx,L)

(1)

Y = ~(Ky, Ly).

(2)

For a model in which we explicitly take account of intermediate goods, we would write the production functions as

(3)

(4) This is precisely the intermediate-good model investigated by Warne, and we can use his diagrammatic derivation of the production possibility curve to analyse the effect of a tax. Figure 6 is a diagrammatic representation of the model given by equations (3) and (4), along with the assumption that the total amounts of Kand L available are fixed in supply. In Figure 6 the axis Xy represents the amount of X used in the production of Y, and the axis Yx represents the amount of Y used in the production of X. fy *, fx * represent the production functions for Y and X under the assumption that the amount of capital and labour is fixed and equal to the total available, and that the only variable input is the intermediate product, X for Y and Y for X. This gives rise to a transformation curve STT'S'. It is clear from this diagram that if equilibrium can be represented as the tangency of a price line, with the transformation curve somewhere in the region

Interprovincial trade restrictions 13 y

FIGURE 6

TT', then the analysis is exactly the same as it was in the preceding figures. This situation would occur if both provinces were net producers of both commodities in final equilibrium. It is quite possible, however, that in equilibrium the oil-importing provinces may have a negative net output of petroleum products. This would occur if their total production of petroleum products was not sufficient to meet the demand for these products in the production of other commodities. With free trade, production would take place at a point such as A 0 and consumption would be at C0 . Now, if the oil-producing provinces impose a tax, production in the oil-importing provinces would take place at a point such as Ao' and the corresponding consumption point would be C0 '. Comparing Figures 5 and 6, we see that the existence of intermediate goods is likely to give rise to a larger reduction in welfare when a tax is imposed by the oil-producing provinces than when intermediate goods are not present. This, of course, is just the corollary to a conclusion previously derived in the literature, namely, that the existence of an intermediate good allows for the possibility of there being larger gains from trade than in a situation in which no intermediate good exists (see Melvin, 1969). We have shown that because petroleum can be viewed as an intermediate product, the imposition of a tax by the oil-producing provinces is likely to result in a larger reduction of welfare for the oil-importing provinces than would be the case for a good which is not an intermediate product. The corollary to this

14 The effects of energy price changes y

FIGURE 7

propos1t10n is also true, of course. Because petroleum is an intermediate product, the gain for the oil-exporting provinces associated with the imposition of a tax is likely to be considerably greater than it would be in the case where the product is for final composition alone. This can be seen from Figure 7, where, as before, C and C" represent respectively the consumption point with free trade and the consumption point after the imposition of the tariff. It is clear that the gains from trade are larger than if production had been confined to the sector TT' of the production possibility set. Of course, the fact that with intermediate products the gains from trade for the country imposing a tariff are larger, and the losses to the other country are greater, is not surprising, for it is simply a reflection of the fact that with an intermediate-product model the effective production possibility set is larger and this gives rise to an expanded possibility for gains from trade. This fact is important, however, because it indicates not only that it is beneficial to produce a product which is used extensively as an intermediate product in other economic regions but also that because of the intermediate-good nature of the commodities, there are significant advantages to the imposition of a tariff. We are assuming throughout, of course, that this 'tariff can be imposed through the use of taxes. In this section we have argued that because taxes can be used to duplicate the tariff structure, taxes can have a significant effect on the terms of trade and the

lnterprovincial trade restrictions 15 welfare of the two economic regions involved. It is important to understand, of course, the precise form that such taxes would have to take in order to improve welfare. All that is required is some kind of production tax that would raise the price of the product to all consumers. For petroleum, a price above that in a perfectly competitive system could be achieved through the imposition of a production tax. Of course, in the case of petroleum, a production tax of this kind may not be required, because the petroleum-producing provinces have considerable control over the price of the product. The government of Alberta claims to have the authority to set the price that other provinces will be required to pay for petroleum products. And even if this claim is not true, Alberta can easily influence the price of petroleum through changes in royalties. Since the provinces are the effective marketing agents for petroleum, the oil-producing provinces have an advantage over provinces producing commodities sold in more competitive markets. Theoretically, it is possible to imagine a case in which a province such as Ontario imposes a production tax on some of the products in which it has a comparative advantage. In fact, however, the BNA Act restricts provincial taxation to taxation at the retail level. There is a significant difference between taxes collected at the retail level, or destinationprinciple taxation, and taxes collected at the production level, or origin-principle taxation. With destination-principle taxation all consumers of the product within the tax jurisdiction are taxed regardless of where the commodity originated. With origin-principle taxation, the tax is imposed at the production level, and anyone who buys the product pays the tax regardless of the economic region in which he lives. In Canada, then, the provinces are restricted to taxation on the destination principle and are not allowed to impose origin-type taxes. However, provinces are allowed to impose corporation income taxes. Under the assumption that this tax is, in fact, reflected in commodity prices, it is really an origin-type tax. This topic is considered in chapter 4. In the case of petroleum, therefore, the imposition of a tax which will improve the terms of trade for the producing province is a simple matter. Any measure which will raise the price to all purchasers of the product will do. Of course, such a tax does not duplicate a tariff, for it produces a distortion only at the production level. The tax system completely equivalent to a tariff structure would require either a consumption tax on the other products in the province or a consumption subsidy to provincial consumers of petroleum. While such a tax system would be needed to maximize terms-of-trade benefits, the production tax by itself will significantly increase welfare. It must also be remembered that the maximum gain requires the imposition of the optimal tax structure. Even in international trade, significant difficulties are encountered when one attempts to find the optimum tariff, and the problem

I 6 The effects of energy price changes will be even more difficult for provinces since provincial trade statistics generally do not exist. While we have used petroleum in our examples, the same basic conclusion will be true for any product over which provinces have monopoly power and thus some control of the domestic terms of trade. And of course the same argument is true regardless of who has control over price. Thus, while we have argued that the BNA Act specifically prohibits provinces from imposing origin-type taxation, essentially the same results can be obtained if private companies decide to charge higher prices for their products. In the petroleum example, from a terms-of-trade viewpoint it makes no difference whether the provincial government or the oil companies raise the price. Of course, one would not expect individual companies to attempt to adjust the terms of trade in order to improve the welfare of the provinces in which they reside, for in general we would expect companies to attempt to maximize their own income rather than the welfare of the communities in which they are located. The point is that there will be a terms-of-trade effect any time the price of petroleum or of any other major commodity is changed. Government decree, company price-policy changes, corporation tax changes, or even property tax changes can all change the terms of trade and result in an intranational redistribution of income. Of course who benefits from this terms-of-trade improvement will depend crucially on how the change is generated.

3

A general equilibrium analysis of energy costs

THE MODEL

The theoretical model employed to this point has been in the neoclassical tradition and has assumed that substitution is possible between inputs in the production of both goods. While this model is an excellent vehicle for theoretical consideration it does not lend itself well to empirical investigation, particularly if a large number of sectors are to be considered. Such an investigation would, for one thing, require knowledge of the form of the production functions for all sectors of the economy. Such information is not available, and we could certainly not expect to make such estimations in this study. Thus, in order to consider the general equilibrium effects of petroleum price increases we must shift to a fixed-coefficient, or input-output, model. But this new model must first be compared to the neoclassical full employment model used up to now. It is crucial, after all, that we understand to what extent the theoretical conclusions that we arrived at earlier will carry over to the fixed-coefficient case. Although input-output tables have been extensively used in empirical investigation of general equilibrium problems, there is a surprising lack of theoretical discussion of them. 1 We begin with the simplest model containing all the basic elements we want to employ. In particular, we assume two goods, X 1 and X 2 , and two primary There is an extensive literature on the generalized Leontref model, but unfortunately it is not of much relevance here. For an early and comprehensive discussion see Koopmans (1951).

18 The effects of energy price changes

FIGURE 8

factors of production, Kand L; and we assume that X 1 and X 2 use each other as an intermediate product in the production process. Both production functions are of the fixed-coefficient variety and are assumed to exhibit constant returns to scale. The two production functions can be written as X 1 = min

{a1 X21 ,/31 L 1 ,-y1 K1 } ,

X 2 = min {aiX12 ,/3il 2 ,-y2 K2

} .

(5)

(6)

We further assume that the total amounts of capital and labour available for production are fixed and, at least for the moment, that tl1e total amount of capital and labour are used, so that full employment exists. Thus, we have

(7)

(8) The production possibility set can now be drawn by maximizing either X 1 or X 2 subject to varying amounts of the other output and subject to (7) and (8). Geometrically the production possibility set can be constructed by extension of either the technique of McKinnon (I 966) or Warne (1971). In Figure 8 the outputs

Analysis of energy costs 19 of Xi and X2 are measured in the first quadrant, axis X2 i represents the amount of X 2 used in the production of Xi, while axis X 12 represents the amount of X 1 used in the production of X 2 . From the production functions (5) and (6), we have a relationship between the quantities of each commodity used in the production of the other, and because of the assumption of fixed coefficient and constant returns to scale, these can be represented by straight lines in the first and third quadrants. Thus OA represents the ratio in which X 2 is used in the production of Xi, and OC represents the ratio of the use of X 1 in the production of X 2 • The next step is to define, from equations (5) and (6), the maximum amounts of X 1 and X 2 that can be produced under the assumption that there is no constraint on the input of the intermediate product. In other words, treating K and L as the only two constraints we calculate maximum production of the two final outputs. Suppose these two final outputs are X 1' and Xi', But to produce Xi' of final output we would need X 2 / of the intermediate input. Similarly, to produce X 2' we would need X 1/ of the intermediate input. Thus the net output points associated with Xi' and X 2' are A and C respectively. It is clear that neither A nor C could be produced in a closed economy for both would require imports of the intermediate good used. We now have two points on the gross output curve;Xi' and X 2'. A third point can be found by noting that there is only one full-employment point in our model. Algebraically, from equations (5), (6), (7), and (8), and recalling that the weighted average of the capital/labour ratio in the two industries must add up to K/L, we can solve for this point. Alternatively, the point can _be found geometrically from the familiar Edgeworth-Bowley Box diagram and is simply the point where the two capital/labour ratios for Xi and X 2 intersect. Either method would give us a point such as £' in Figure 8. The curve Xi' E'X 21 is the gross output curve, for we have yet to take account of the fact that some of the output of each industry will be used up in the production process of the other. To find a net output curve, that is, the locus of output points available for consumption, we must take account of this intermediate use of the two outputs. First, consider E' implying gross outputs of Xi" and X/. From OA we see that the production of Xi" implies that X 2 /" be used as an intermediate input, and the production of X/ implies, from line OC, that X 12" be used as an intermediate input in that industry. To find the net outputs we now subtract the intermediate use of the two outputs from the gross output. Subtracting X 12" from Xi" we get point Mand subtracting X 2 i" from X/ we get point N. These two points give point£, the full -employment net-output point. This procedure can be simplified by observing that point M can be found from point E' by drawing the line E'M with slope the negative of OC, and that N can be found by drawing from £' a line E'N with slope the negative of OA . Of

20 The effects of energy price changes

FIGURE 9

course, this will apply for any point on the gross output locus. For example, the point on the axis of the net curve for X 1, that is, the point which implies maximum net output of X 1 and zero net output of X 2 , can be found by drawing, from the origin, a line OG with slope the negative of OA, and then by drawing from G a line GT with slope the negative of OC. This gives us T, and T' can be found in an exactly analogous way . The net transformation curve is thus the curve AEC. For the closed economy only TET' is possible, for points between A and T and between T' and C would require imports of the intermediate products in order for production to take place. We have redrawn the production possibility curve on Figure 9 and eliminated some of the unnecessary lines. This production possibility curve can be used in much the same way as the neoclassical production possibility curve, which implies full employment throughout its entire length. The major difference is that in the fixed coefficient model there is a single full-employment point E, and for all price lines which give rise to production at that point, namely, price lines with slope between that of AE and EC, there will be a fixed output configuration and a constant distribution of factors between industries. Our basic conclusions about the terms-of-trade effects of a tariff will be unchanged, however, and the conclusions that we reached in chapter 2 about trade in intermediate products will also hold for this model. To take just one example, suppose the terms of trade are represented by the line P in Figure 9. If trade in intermediate products is not possible, then production will be at T and consumption at C. If, on the other

Analysis of energy costs 21

FIGURE 10

hand, trade in intermediate goods is possible, then the production point will be A, the consumption point will be C 1 , and it is clear that welfare has been improved, for C 1 represents a higher level of community welfare than C. Before proceeding to the algebraic formulation of our fixed-coefficient model, there are a few observations which can be made from the geometry. First, we can easily derive the conditions for the production model to be feasible, the conditions which guarantee that at least some net output of both goods is possible. The condition is simply that the lines OA and OC make an angle of less than 180°. If this is not so, then none of the points on AEC could be in the first quadrant. The condition that OA and OC make an angle of less than 180° can alternatively be expressed by saying that X 1/X21 > X 1JX2. Putting this in the more usual production coefficient terms we have that I /a 21 > a 12 or, alternatively, that I - a 12 q21 > 0. This, of course, is just a Hawkins-Simon condition. Another point we can observe in the geometry is that it is not necessary for point£ to be in the first quadrant, because the situation as shown in Figure 10 is possible. This presents no fundamental problem for the model in which trade is allowed, but it does present a situation where full employment is impossible unless international trade is allowed. Consumption, of course, must always be in the first quadrant, and since full employment requires production in the second quadrant, only with a price line such as P', implying production at E and consumption at C, is full employment possible.

22 The effects of energy price changes We will now turn to the algebraic formulation of the model and initially assume that only three goods are produced: X 1 , X 2 , and X 3 • In equations (9), (10), and (11) we have written, for the three industries, the condition

(9) (10) (11) that total revenue equals total cost. Here p 1, p2 , and p 3 are, respectively, the prices of commodities 1, 2, and 3, and w and rare the wage rate and return to capital. Of course these two relationships will hold for any production functions, but in general we would not have the fixed relationships between the levels of inputs and the levels of output which exist in the fixed coefficient model. Because of our assumption of constant returns to scale we can divide equations (9) by X 1 p 1 , (l O) by X2 p2 , and (11) by X3 p 3 to obtain (12) (13) (14) Here oii is equal to XijP/XjPj, ui is equal to Liw/xipi, and vi is equal to K/l~Pt Equations (12), (13), and (14) give the allocation of cost for each dollar's worth of output. Now suppose that for some reason exogenous to the system the price of one of the commodities rises. Assume, for example, that the government imposes a tax on all purchases of commodity 3. From the point of view of the producers of commodity 1 the immediate impact of this price change will be that they will pay more for inputs of commodity 3. In particular, the 0 31 in equation (12) must now be replaced by 0 31 (1 + t). Since total revenue must still equal total cost and since we are assuming that this change in the price of a unit of X 3 does not affect either the wage or the return to capital, it must be reflected in a higher price for commodity 1. This price change, due directly to the increase in the cost of commodity 3, is not the only price change which will affect the first industry. Just as the price that producers of commodity 1 must charge will rise, so there will be an increase in the prices of the other products when the tax on X 3 is reflected in their costs.

Analysis of energy costs 23 In other words, there will be an indirect price effect as well as a direct price effect. We would expect the prices of all other commodities, in our case the prices of X 1 and X 2 , to increase so that the price of all inputs will rise. Thus, for example, in equation (12) we would replace a 11 with a 11 (l + p1) and a 21 with a 21 (l + p2) . The p is defined to be the total change in the commodity price. Note that just as there are indirect price changes for commodities 1 and 2, so there will be an indirect price change for commodity 3, the industry on whose output the tax was imposed. In other words, the price of commodity 3 will increase, first because of the tax and second because of the indirect price increases of all products used to produce commodity 3. The total price changes for industry 1 are shown in equation (15).

Performing the indicated multiplication, and subtracting (12) from (15) we obtain (16) In an analogous way we can write the expression for the change in price in industries 2 and 3, except that the left-hand side of the equation for industry 3 will be bil + t), the term p3t reflecting the fact that, because we have assumed an ad valorem tax, when prices rise tax collections will rise and there will thus be a further indirect price increase. Our system of three equations can now be written in matrix notation as

(17) Here p is the vector of price changes, A the coefficient matrix, a 3 the column vector (a 31 , a 32 , a 33), and t the tax change. Solving (17) for p we obtain (18) This equation is completely general and applies to any number of sectors. From 18, in order to calculate the price effects of the tax we need to know the matrix A, the tax, and the vector a3. Of course a 3 is just the relevant row of the coefficient matrix and is known if A is known. DIRECT AND INDIRECT EFFECTS FOR ONTARIO

Using the model of the previous section and the 1965 Ontario input-output table we have calculated the direct and total price effects of an increase in the costs of

24 The effects of energy price changes TABLE l Price effects of hydrocarbon fuel price increases for Ontario 2

Industries l 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37

Agriculture, Forestry and Fishing Mining Meat and poultry Daily products Grain mills Biscuits and bakeries Sugar and confectioneries Other food Soft drinks Distilleries, breweries , and wineries Tobacco and tobacco products Rubber products Leather and leather products Cotton yarn and cloth Synthetic textiles Knitting mills Clothing Other textile mills Sawmills Furniture and fixtures Other wood Pulp and paper mills Paper products Printing and publishing Iron and steel mills Other primary metals Fabricated and structural metals Metal stamping, pressing, and coating Other metal fabricating Miscellaneous machinery Motor vehicles and aircraft Other transportation equipment Electrical appliances Electrical industrial equipment Communication equipment Other electrical products Clay, lime, and cement

Direct effect 4.98 1.23 0.17 0.57 0.30 0.67 0.21 0.23 0 .82 0.23 0.08 0 .30 0.31 0.17 0.61 0.20 0.06 0.25 1.22 0.15 0.27 0.58 0.82 0.09 0.15 1.51 0.00 0.15 0.26 0.49 0.13 0.30 0.46 0.15 0 .35 0.27 1.25

3 Ratio Total/ Direct 2/1

Total effect

4

Rank change

(48) (32) (10) (40) (28) (16) (12) (46) (41) (5) (49)

7.102 2.468 5.283 5.189 4.221 2.697 1.749 3.495 2.112 1.321 4 .001 2.553 2.396 1.727 3.503 1.777 1.404 l.871 2.676 l.505 1.921 1.910 2.695 0.904 1.137 3.438 0.689

(3) (19) (5) (6) (7) (15) (32) (10) (23) (42) (8) (20) (22) (35) (11) (33) (39) (29) (16) (37) (30) (31) (17) (47) (45) (12) (49)

1.42 2.00 31.08 9.10 14.07 4.02 8.32 15 .19 2.57 5.74 50.01 8.51 7.72 10.15 5.74 8.88 23.40 7.48 2.19 10.03 7.11 3.29 3.28 10.04 7.58 2.27

0 -10 + 33 + 11 + 18 - 2 + 4 + 23 -12 - 8 + 39 + 6 + 2 + 4 + 3 + 4 + 9 3 - 6 3 - 2 -15 - 5 - l - 4 - 7 0

(42) (30) (19) (44) (27) (20) (43) (22) (29) (7)

1.208 1.514 1.551 1.342 1.418 1.846 1.419 l.956 2.096 2.675

(44) (38) (36) (43) (40) (34) (41) (27) (25) (18)

8.05 5.82 3.16 10.32 4.72 4.01 9.46 5.58 7.76 2.14

- 2 - 8 - 17 + l - 13 - 14 + 2 - 5 + 4 - 11

(3) (9) (38) (17) (25) (13) (36) (33) (11) (34) (47) (26) (24) (39) (14) (37)

Analysis of energy costs 25 TABLE 1 continued 2

Industries 38 Other non-metallic mineral products 39 Petroleum refineries and coal products 40 Plastics and synthetic resins 41 Paint and varnish 42 Pharmaceuticals and medicines 43 Other chemical 44 Miscellaneous manufacturing 45 Construction, maintenance, and repair 46 Transportation, storage, and trade 4 7 Utilities 48 Communications and other services 49 Unallocated sector

Direct effect

3 Ratio Total/ Direct 2/1

Total effect

4

Rank change

1.43

(6)

3.090 (13)

2.16

- 7

7.87 0.37 0.59 0.33 5.00 0.23

(I)

(21) (15) (23) (2) (35)

10.870 (I) 6.355 (4) 3.869 (9) 2 .027 (28) 7.946 (2) 2.071 (26)

1.38 17.17 6 .55 6 .14 1.58 9 .00

0 + 17 + 6 - 5 0 + 9

1.24 2.10 0.55 0.26 0.11

(8) (4) (18) (31) (45)

(21) (14) (46) (48) (24)

2.05 1.42 2.00 3.30 19.16

- 13 - 10 - 28 - 17 + 21

2.548 2.999 1.104 0 .860 2.108

hydrocarbon fuels . These estimates are presented in Table 1. Before proceeding with the discussion of these results it will be useful to review the main assumptions and limitations of our model. We have repeatedly referred to the fact that we are assuming a fixedcoefficient model. By definition this means that no substitution is allowed among inputs. To the extent that this is not the case our estimates will be biased on the high side, for if substitution is possible between hydrocarbon fuels and other kinds of fuels, such as electricity, then we would expect that when the price of hydrocarbons rises there will be a shift away from these fuels, resulting in a lower increase in the price of the product being produced. Of course, the degree to which such substitution is possible will vary from industry to industry, and in many industries we would no_t expect electricity to be a very good substitute for oil and natural gas. A second limitation of the data affecting our calculations for Ontario is that the input-output table does not isolate petroleum and natural gas, but includes them along with coal. We are forced to assume that the price of these fuels increases by the same amount, realizing that in fact this seems very unlikely . To the degree that coal is a substitute for petroleum and natural gas we would naturally expect any increase in the price of the latter to result in price increases for the former, and in the limiting case where substitution is perfect the price

26 The effects of energy price changes rises will be identical. As a check on the difference that this aggregation makes to our results we have made the same calculations for the Canadian input-output table, where petroleum and natural gas, and coal are included as different industries. The results of these calculations are reported in the next section. In all of our calculations we are assuming that the cost increases are entirely passed along in price increases and that there is no change in the per unit return to capital or to labour. While this seems to be as good an assumption as any, it is certainly possible that some of the increase in cost would be reflected in lower returns to capital. Because of the presence of unions, however, it seems unlikely that producers would be able to lower their wage costs. Another assumption implicit in our analysis is that all price changes apply to imports as well as to domestically produced goods. While this is probably not unrealistic for the petroleum sector, for Ontario buys virtually all its petroleum and natural gas from western Canada, it is not a very satisfactory assumption for all the other intermediate goods in the model. In other words, the indirect effects of the petroleum price increase are assumed to affect all the purchases of these products by Ontario industry, and if similar petroleum price rises are not occurring in the rest of the world there is no reason to believe that the costs of imported intermediate products will rise. Unfortunately, since the Ontario input-output table does not distinguish imports of intermediate products, we have no alternative but to make this assumption. We also recall that constant returns to scale have been implicitly assumed, for we have assumed that the unit costs are the same regardless of the level of operation of the industries. Of course even if increasing returns to scale exist this assumption will not be particularly harmful unless significant changes in the level of output occur. We should note, however, that if there are increasing returns to scale and if increases in the costs of fuels do result in a reduction in over-all output, then we shall be underestimating the commodity price changes. Finally, we note that crude petroleum and natural gas are included along with petroleum products in the Ontario input-output table, so that, in effect, the price change we have assumed applies not only to crude oil and natural gas but to refined petroleum products as well. Since the Canadian input-output table does distinguish between crude petroleum and petroleum products we will be able to check the effect of this aggregation when the Canadian data are presented. Columns I and 2 of Table I give the direct and the total price effects of an increase in the price of hydrocarbon fuels.2 For the purposes of our calculations 2 We will refer, throughout, to petroleum and coal as fuels. For a number of industries, however, particularly industries 39 to 43, petroleum, and to some extent coal, are used not as fuels but as goods in process.

Analysis of energy costs 27 we have assumed that the price of these products doubles, that a 100 per cent tax has been imposed. The figures in brackets of columns 1 and 2 are the rank of the price changes, while column 3 is the ratio of the total effect to the direct effect. Column 4 gives the difference in the ranks for the direct and total effects. The first thing that we observe from this table is that, on average, the direct effects tend to be very small. Of the 49 industries, only ten indicate a direct price effect of more than 1 per cent for a 100 per cent increase in the price of hydrocarbon fuels. Of these, the two highest are industry 39 (Petroleum and coal) and industry 43 (Other chemical industries). This is what we would expect, for 39 is the industry on which the tax was applied, and 43 is very closely related to the petroleum industry. Somewhat surprising is the fact that the price change in the Agriculture, forest, and fishing sector is almost as high as that in Chemicals. But after these three industries the price changes fall quite quickly and, in fact, the median price change is only 0.3 per cent. Thus, for over half of the industries tripling the price of hydrocarbon fuels would result in a product price increase of less than 1 per cent. What we are really interested in is not the direct effects of an increase in the price of hydrocarbon fuels, but rather the total effect that has on prices, and this information is presented in column 2. From this column, and from column 3, which presents the ratios of the total to the direct effects, we can see that the total effects are significantly higher than the direct effects. However, even after taking account of all the direct effects there is only one industry where the price change is over 10 per cent, and that is the petroleum and coal industry itself. The median price change is 2.1 per cent. One of the interesting aspects of the comparison of the direct and total changes is the degree to which the ranking of these two price changes differs. We note, for example, that Tobacco goes from 47 to 8, and that Meat and poultry moves from 38 to 5. The actual price changes are worthy of note as well, Tobacco moving from less than 0.1 per cent to 4, and Meat and poultry moving from less than 0 .2 per cent to 5.3. At the other end of the scale we have Utilities moving from 18 to 46, Communications moving from 31 to 48, and Miscellaneous machinery manufacturing moving from 19 to 36 . Indeed, as we look down the table we observe that most of the positive rank changes occur in food processing industries and the textile industries, while most manufacturing sectors have negative rank changes. For the agricultural and the food processing industries, not only are the changes in rank large, but the actual percentage price changes are also high . Note that for industries 1 (Agriculture, forestry, and fishing), 3 (Meat and poultry), and 4 (Dairy products), the total price effect is in all cases greater than 5 per cent. Only Petroleum and coal, Plastics, and Other chemicals industries have higher total price changes. Agriculture and the related sectors, of course, are heavy users of

28 The effects of energy price changes TABLE 2 Ontario net export activity

Industries

1 2 3 4 5 6 7 8 9 IO

ll

12 13 I4 I5 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

Agriculture, forestry, and fishing Mining Meat and poultry Dairy products Grain mills Biscuits and bakeries Sugar and confectioneries Other food Soft drinks Distilleries, breweries, and wineries Tobacco and tobacco products Rubber products Leather and leather products Cotton yarn and cloth Synthetic textiles Knitting mills Clothing Other textile mills Sawmills Furniture and fixtures Other wood Pulp and paper mills Paper products Printing and publishing Iron and steel mills Other primary metals Fabricated and structural metals Metal stamping, pressing, and coating Other metal fabricating Miscellaneous machinery Motor vehicles and aircraft Other transportation equipment Electrical appliances Electrical industrial equipment Communication equipment Other electrical products Clay, lime, and cement Other non-metallic mineral products Petroleum refineries and coal products Plastics and syn the tic resins

Total price effect

7 .102 2.468 5.283 5.189 4.221 2.697 1.749 3.495 2.112 l.321 4.001 2.553 2.396 l.727 3.503 1.777 l .404 l.87 l 2.676 I.SOS l .921 l.910 2.695 0.904 l.137 3.438 0.689 l.208 l.514 1.551 l.342 l.418 l.846 1.419 l.956 2.096 2.675 3.090 10.970 6.355

(3) (19) (5) (6) (7) (15) (32) (I 0) (23) (42) (8) (20 (22) (35) (1 l) (33) (39) (29) (16) (37) (30) (31) (17) (47) (45) (12) (49) (44) (38) (36) (43) (40) (34) (41) (27) (25) (18) (13) (I)

(4)

2 Net trade balance 406 ,606 137,481 - 93,465 165,905 - 88,382 61 ,723 - 42,933 91,977 - 3,443 132,207 101,560 139,758 38 ,359 - 88,679 - 62,256 46,501 -210,516 18,662 - 87 ,960 73,132 - 28,642 207,997 30,882 148,792 -337 ,578 -398 ,879 27 ,590 223,744 332,413 287,403 1,057 ,445 12,264 189,587 239,096 234,258 25 ,195 3,755 -l,o28 - 85,614 - 35 ,698

3 Trade balance Total output 0.238 0.166 -0.146 0.350 -0.299 0.266 -0.297 0.115 -0.042 0.458 0.589 0.361 0.215 -l.510 -0.305 0.383 -0.803 0.053 -1 .045 0.278 -0.166 0.355 0.071 0.255 - 0.321 -0.535 0.138 0.537 0.203 0.588 0.303 0.075 0.593 0.655 0.429 0.133 0.012 -0.004 -0.150 -0.203

Analysis of energy costs 29 TABLE 2 continued

Industries

Total price effect

41 42 43 44 45 46 47 48 49

3.869 2.027 7.946 2.071 2.548 2.999 1.104 0.860 2.108

Paint and varnish Pharmace uticals and medicines Other chemical Miscellaneous manufacturing Construction, maintenance, and repair Transportation, storage, and trade Utilities Communications and other services Unallocated sector

(9) (28) (2) (26) (21) (14) (46) (48) (24)

2 Net trade balance

3 Trade balance Total o utput

- 17,115 2,875 211,609 123,316 630 ,224 - 2,000,831 - 256,344 - 391,030 - 229,675

- 0.142 0.023 0.213 0.228 0.198 - 0.482 - 0.529 - 0.057 - 0.121

petroleum products, but the very large price rises relative to most other sectors is noteworthy. Among other things it suggests that the recent petroleum price rises will result in further increases in the cost of food . Of the basic manufacturing industries, on the other hand, for only one of them , industry 26 (Other primary metals), do we find a total price change in excess of3 per cent. Most are less than 2 per cent. Table I clearly suggests that other than those industries directly related to the petroleum industry, price changes in the manufacturing sector will be quite moderate, with much more significant price changes in those industries relating to agriculture. We also note that industries 45 and 36 (Construction and Transportation and trade) have higher than average direct price effects. These observations suggest that industries which tend to be domestic in nature and which tend to produce for the local market will be harder hit than those industries which export significant amounts of their outputs to the rest of Canada. While Ontario producers of food products do export, these exports in the main tend to be to other countries rather than to other parts of Canada, while a much more significant part of the exports of manufactures go to other parts of Canada. Unfortunately, there is little direct evidence on the composition of Ontario trade as between Canada and the rest of the world; the Ontario input-output table does not show exports to the rest of Canada, and for the full table gives only net exports to the rest of the world. Exports and imports are shown separately for a more aggregated classification, but exports to the rest of Canada are not distinguished from exports to the rest of the world. Furthermore, this net export number is simply a residual, being the difference between total Ontario production and total Ontario consumption. Nevertheless, some indication of the relationship between exports and price changes due to increases in hydrocarbon fuel prices can be seen from Table 2. Here we show the net trade balance by sector and the ratio of

30 The effects of energy price changes exports to total production for the 49 sectors of the Ontario input-output table. We note that for the agriculture-related industries, industries l to 13 (excluding 2), although Ontario has a substantial trade surplus, this surplus is with the rest of the world and not with the rest of Canada. The price changes in these industries tend to be relatively high, and thus the effect on exports will depend on the price change in petroleum products in the rest of the world. Petroleum prices have risen by at least as much in Canada's major trading partners as they have in Ontario. In the textile sector, industries 14 to 18, Ontario has a significant trade deficit, but again this deficit is with the rest of the world and not with other Canadian provinces. In any case, price changes are relatively small in this group, so that it seems unlikely that there will be much change in trade patterns. Ontario is also a net exporter of the forest products group, industries 19 to 24, but of the major exports the price changes are small, so that major changes in exports seem unlikely. There might be some increase in imports, but for these industries it certainly will not be from Alberta or Saskatchewan. There is a surprisingly large trade deficit in 25 (iron and steel) and 26 (Other primary metals) and if prices rise only in Ontario imports of these industries could increase significantly. Again, however, the imports are not from the oilproducing provinces. In the manufacturing sector, industries 27 through 36, price changes are small and there is an export surplus for all these sectors. Only in the petroleum-related sectors, 39 to 42, where there is a trade deficit and large price changes, would one expect a large increase in imports. And even in this sector note that for 43 (Other chemicals), with the second highest price change, Ontario's trade surplus is larger than the deficit for all of industries 39 to 42. In summary there does not seem to be much reason for concern from the trade point of view. A COMPARISON WITH CANADA AND THE US

In this section we examine the implications to Canada of price rises in hydrocarbon fuels and then compare these with similar price increases in the United States. We want to make these comparisons for three reasons. First, Ontario trades mainly with the rest of Canada and the United States, and some information on the relative importance of commodity price changes due to increases in the costs of fuels can be gained from these comparisons. Second, although the Ontario input-output table provides a great deal of information on the structure of Ontario industry, it does have its drawbacks, one of the most important of which is the absence of any detailed export or import figures. It is hoped that further insights into Ontario trade patterns, and perhaps even into

Analysis of energy costs 31 employment and income changes, can be obtained from an examination of the Canadian data. An examination of the Canadian input-output table also allows us to distinguish between the petroleum industry and the coal industry, something we could not do for Ontario. One final reason for examining comparable statistics for Canada and the United States is that we can examine the implications of a uniform price rise everywhere. It is often assumed that petroleum price increases will be neutral as long as the same price rise occurs in all markets; in this section we present evidence that this is an oversimplification. Using the 1963 Canadian 65-sector input-output table we have calculated the direct and total effects for both coal and petroleum and then for the sum of these two sectors, and these are reported in Table 3. As with Table 1, these figures represent the direct and total price effects of a 100 per cent increase in the price of these two fuels. The effects of petroleum price increases are generally much larger than for coal price increases; this, of course, is what we would expect and simply reflects the predominant use of petroleum as a fuel source rather than coal. The only industry in which coal price increases result in a significant price increase is industry 31 (Iron and steel mills). Again, this comes as no surprise, for it is well known that this industry is still a major coal consumer. Other sectors where coal is an important input are industry 28 (Pulp and paper mills), industry 46 (Clay, lime, and cement), and industry 52 (Other chemicals). There are also several industries where the total effects of price increases in coal are significant. These are the industries in the group 32 (Smelting and refining), to 41 (Other transport equipment). The significant total effects here result, not from the fact that these industries use coal directly, most of them having very small direct effects, but from their using significant quantities of the output of industry 31 (Iron and steel mills). As one specific example, we note that for industry 34 (Structural metals), doubling the price of coal would result in a 2.31 per cent increase in the price of this industry's output, even though the industry itself uses virtually no coal; the direct effect is only 0.04 per cent. Turning to the petroleum industries, columns 3 and 4 in Table 3, the direct effects for Canada are, on the average, very small relative to Ontario. In fact, of the sixty-five industries listed, fifty-three have direct effects of less than 0.1 per cent. This, however, is due, not to any fundamental difference in the structure of the Ontario and Canadian economies, but rather to the different aggregation of the two tables. In the Canadian table, as well as a sector for Petroleum and natural gas, industry 7, we have industry 48, the Petroleum products industry, while in the Ontario table these two, along with coal, were aggregated into sector 39. As we would expect, the direct price effect on sector 49 (Petroleum products) of an increase in the price of petroleum and natural gas, sector 7, is

TABLE 3 Canadian petroleum and coal price effects

Industries I 2 3 4 5 6 7 8 9

l0 11 12 13 14 15 16 17 18 I9 20 21 22 23

Agriculture Forestry Fishing, hunting Metal mines Non-metal mines Coal mines Petroleum, natural gas Meat, poultry processing Dairy factories Fruit, vegetable canners Feed, flour, cereal Biscuits, bakeries Sugar, confectionery Other food Soft drink manufacturing Alcoholic beverage mfg Tobacco products Rubber products mfg Leather products mfg Synthetic textile mills Cotton, yarn, cloth Knitting mills Clothing

2

3

4

Direct effects, coal

Total effects, coal

Direct effects, petroleum

Total effects, petroleum

0.12 0.01

0.30 0.19 0.15 0.42 0.63 0.57 0.13 0.31 0.42 0.85 0.36 0.30

0.01

3.87 2.57 3.30 3.02 3.52 0.98 4.31 2.74 3.11 1.90 2.76 2.12 1.29 2.15 1.34 1.29 2.17 1.59 1.19 1.86 0.86 1.11 0.96

0.15 0.41 0.40 0.06 0.13 0.33 0.05 0.03 0.24 0.10 0.06 0.21 0.03 0.37 0.18 0.08 0.44 0.18 0.02

0.44

0.40 0.31 0.46 0.31 0.80 0.52 0.52 0.72 0.52 0.35

0.01 2.15 1.14

3.23 0.02

0.o2 0.01 0.01 0.06 0.05

0.o3 0.01

0.o2 0.10 0.01 0.01

5 Total effects, petroleum and coal

6 Total effects, petroleum industries

4.17 2.75 3.45 3.44 4.15 1.54

6.32 4.19 5.41 1.40 3.82 1.56 1.63

4.44

3.05 3.53 2.75 3.12 2.41

l.73 2.55 1.65 1.75 2.48 2.39 1.71 2.38 1.58 1.63 1.31

4.41

5.02 2.95 4.46 3.27 1.94

3.39 2.10 1.99 3.47 2.16 1.80 2.75 1.30 1.68

1.47

TABLE 3 continued

Industries 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47

Other textile Sawmills Furniture, fixtures Other wood Pulp and paper mills Other paper Printing and publishing Iron and steel mills Smelting and refining Other primary metal Structural metal Metal stamping Other metal mfg Machinery mfg Aircraft and parts mfg Motor vehicle mfg Vehicle parts mfg Other transport equipment Electrical appliance Electrical industrial equipment Communications equipment and wire Other electrical products Clay, lime, cement Non-metallic minerals

2

3

4

Direct effects, coal

Total effects, coal

Direct effects, petroleum

0.25

0.61 0.19 0.61 0.29 1.40 0.77 0.40 6.11 1.40 1.73 2.31 2.09 1.46 1.00 0.48 1.00 1.51 1.08 0.44 0.58 0.49 0.92 1.84 0.44

0.01

0.11 0.06 1.15 0.05

O.o2

5.79 0.73 0.10 0.04 0.05 0.08 0.10 0.05 0.15 0.34 0.15 0.10 0.17 0.04 0.04 1.65 0.14

0.01 0.01 0.51 0.01 0.01 0.07 0.08 0.04 0.01 0.04 0.04 0.02 0.01 0.01 0.03 0.01 0.02 0.01 0.01 0.01 0.24 0.14

To tal effects, petroleum

5 Total effects, petroleum and coal

6 Total effects, petroleum industries

1.38 1.96 1.09 1.86 2.35 2.77 1.00 2.02 2.52 1.78 1.22 1.30 1.19 1.06 1.04 1.07 1.18 1.07 4.37 3.05 1.58 1.56 0.95 1.34

1.98 2.15 1.70 2.14 3.75 3.54 1.40 8.12 3.91 3.51 3.53 3.39 2.65 2.06 1.52 2.07 2.69 2.15 4.81 3.63 2.07 2.47 2.79 1.78

2.07 3.18 1.62 2.96 2.88 4.10 1.39 2.52 1.84 1.98 1.64 1.71 1.58 1.40 1.55 1.52 1.60 1.49 1.56 1.30 1.23 1.44 3.24 2.75

TABLE 3 continued

Industries 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

Petroleum products Plastic synthetic resin Paint and varnish mfg Pharmaceuticals, soap, toiletries Other chemicals Miscellaneous mfg Construction Trade, wholesale, and retail Transport, storage Communications Utilities Finance, insurance, real estate Business services Hotel, restaurant Other services Office supplies Advertising and travel Operating supplies

Arithmetic average Weighted average

2

3

4

Direct effects, coal

Total effects, coal

Direct effects, petroleum

0.1 I 0.22 0.03 0.14 1.07 0.09

0.30 0.52 0.30 0.34 1.59 0.60 0.62 0.16 0.25 0.07 0.77 0.10 0.08 0.18 0.16 0.44 0.24 0.62

58.19 0.09 0.01 0.01 0.66 0.01

0.74 0.52

I.JO 0.96

0.10 0.55

0.01

0.28 0.17

0.03 0.02 0.06 0.02 0.01 0.01

Total effects, petroleum

5 Total effects, petroleum and coal

6 Total effects, petroleum industries

59.16 1.63 3.32 1.39 5.24 1.41 1.88 1.74 2.94 0.47 0.99 0.71 0.48 1.41 1.01 1.37 1.24 1.78

59.46 2.15 3.62 1.73 6.84 2.02 2.50 1.90 3.19 0.54 1.76 0.81 0.56 1.59 J.J 7 1.81 1.49 2.40

6.95 5.72 5.67 2.43 7.02 2.01 2.89 2.76 4.79 0.75 1.46 1.09 0.75 2.25 1.60 1.98 1.92 1.06

2.85 2.74

3.59 3.26

2.67

Analysis of energy costs 35 very large, and in fact we see that a doubling in the price of petroleum would result in a 58 per cent increase in the price of the output of industry 48. Although the direct effects in the Canadian table are small, the total effects are quite significant, reflecting the fact that while very few industries use crude petroleum or natural gas directly many of them do use the output of industry 48, the petroleum products industry. It is therefore not meaningful to compare the direct effects from the Ontario table with those for the Canadian table, since the aggregation is quite different. The indirect effects are more comparable, although even here caution is required, for it is clear that more is included in the Ontario sector 39 than in the Canadian sectors 6 and 7. Most of the Canadian sector 48, Petroleum products, must be included in Petroleum refineries and coal products, sector 39 of the Ontario table. Thus, since we are applying the same tax to the crude petroleum and natural gas sector in the Canadian table as to the petroleum refineries and coal products sector of the Ontario table, we would expect the total ·effects to be generally higher for Ontario than for Canada. To give some idea of the difference, we show in column 6, Table 3, the total price effect of a 100 per cent increase in the price of the Petroleum products industry in Canada, sector 48, rather than assuming that the increase was in crude petroleum and natural gas. The difference between columns 5 and 6 is somewhat less than we might have expected a priori, in part because column 5 contains an increase in coal prices as well as in petroleum prices, but also reflecting the extent to which an increase in the price of crude petroleum and natural gas is passed on through the refining stage to other products. To a considerable extent, this is owing to the relatively small value added in the petroleum refining sector. In any case, the comparison of columns 5 and 6 in Table 3 suggests that the total effects for Ontario from Table 2 can be compared with the total effects for Canada from column 5 of Table 3. To facilitate this comparison Table 4 arranges the sectors from the Ontario forty-nine-industry table and the Canadian sixty-five-industry table so that the comparable industries are side by side. Columns I and 3 of this table are simply reproduced from Tables I and 3, and column 2 represents the difference between the two; a positive number shows that the total effect is larger for Ontario and a negative number that the total effect is larger for Canada. This middle column can be thought of as giving an estimate of the price advantage or disadvantage that Ontario would experience if petroleum and coal price increases were uniform throughout Canada. For those sectors in the Canadian table which are disaggregated more than in the Ontario table we have constructed an average which is used to find the difference. These are weighted averages, the weights being the total sales of the industries.

TABLE 4 Ontario and Canada, price change comparisons

Ontario ( 49 industries) I Agriculture, forestry, and fishing

Total effect

2

3

Difference

Total effect

Canada (65 industries)

4.17 2.75 3.45

Agriculture Forestry Fishing, hunting

I 2 3

- l.10

3.44 4.15

Metal mines Non-metal mines

4 5

7.10 3.35

2 Mining

2.47

3 Meat and poultry

5.28

2.23

3.05

Meat, poultry processing

8

4 Dairy products

5.19

1.66

3.53

Dairy factories

9

5 Grain mills 6 Biscuits and bakeries 7 Sugar and confectioneries

4.22 2.70 1.75

1.10 0.29 0.02

3.12 2.41 1.73

Feed, 0our, cereal Biscuits, bakeries Sugar, confectionery

11 12 13

8 Other food

3.50

2.75 2.55

Fruit, vegetable canners Other food

IO

0.90 0.46 - 0.43 l.52 0.16 0.69 0.15 l.l2

1.65 1.75 2.48 2.39 1.71 l.58 2.38

Soft drink manufacturing Alcoholic beverage mfg Tobacco products Rubber products mfg Leather products mfg Cotton, yarn, cloth Synthetic textile mills

15 16 17 18 19 21 20

9 Soft drinks IO Distilleries, breweries, and wineries 11 Tobacco and tobacco products 12 Rubber products 13 Leather and leather products 14 Cotton yarn and cloth I 5 Synthetic textiles

2.11 l.32 4.00 2.55 2.40 1.73 3.50

14

TABLE 4 continued 2 Ontario (49 industries)

Total effect

16 17 18 19 20 21 22 23 24

1.78 1.40 1.87 2.68 1.51 1.92 1.91 2.70 0.90

Knitting mills Clothing Other textile mills Sawmills Furniture and fixtures Other wood Pulp and paper mills Paper products Printing and publishing

25 Iron and steel mills

1.14

26 Other primary metals

3.44

27 28 29 30

Fabricating and structural metals Metal stamping, pressing, and coating Other metal fabricating Miscellaneous machinery

31 Motor vehicles and aircraft

0.69 1.21 1.51 1.55

3 Total effect

Canada (65 industries)

0 .15 0.09 0.11 0 .53 0.19 0.22 1.84 0.84 0.50

1.63 1.31 1.98 2.15 1.70 2.14 3.75 3.54 1.40

Knitting mills Clothing Other textile Sawmills Furniture, fixtures Other wood Pulp and paper mills Other paper Printing and publishing

22 23 24 25 26 27 28 29 30

- 6.98

8.12

Iron and steel mills

31

- 0.26

3.91 3.51

Smelting and refining Other primary metal

32 33

-

3.53 3.39 2.65 2.06

Structural metal Metal stamping Other metal mfg Machinery mfg

34 35 36 37

1.52 2.07 2.67 2.15 4.81

Aircraft and parts mfg Motor vehicle mfg Vehicle parts mfg Other transport equipment Electrical appliance

38 39 40 41 42

Difference

-

2.84 2.18 1.14 0.51

1.34 - 0.85

32 Othe r transport equipment 33 Electrical appliances

1.42 1.85

- 0.73 - 2.96

TABLE 4 continued

Ontario ( 49 industries)

34 35 36 37 38

Total effect

Electrical industrial equipment Communication equipment Other electrical products Clay, lime, and cement Other non-metallic mineral products

1.42 l.96 2.10 2.68 3.09

39 Petroleum refining and coal production

10.87

40 41 42 43 44 45

Plastics and synthetic resins Paint and varnish Pharmaceuticals and medicines Other chemical Miscellaneous manufacturing Construction, maintenance, and repair

46 Transport, storage, and trade

6.36 3.87 2.03 7.95 2.07 2.55

2

3

Difference

Total effect

-2.21 -0.ll -0.37 -0.ll 1.31

3.63 2.07 2.47 2.79 l.78

-21.69

1.54 4.44 58.30

4.21 0.25 0.30 1.11 0.05 0.05

Electrical industrial equipment Communications equipment and wire Other electrical products Clay, lime, cement Non-metallic minerals

43 44 45 46 47

Coal mines Petroleum natural gas Petroleum products

6 7 48

2.15 3.62 l.73 6.84 2.02 2.50

Plastic, synthetic resins Paint and varnish mfg Pharmaceutical soap, toilet Other chemicals Miscellaneous mfg Construction

49 50 51 52 53 54

1.11

l.90 3.19

Trade, wholesale, and retail Transport, storage

55 56

-0.66

l.76

Utilities

58

-0.03

0.54 0.81 0.56

Communications Finance insurance, real estate Business services

57 59 60

3.00

47 Utilities

1.10

48 Communications and other services

0.86

Canada (65 industries)

TABLE 4 continued 2 Ontario ( 49 industries)

49 Unallocated sector

Total effect

Difference

2.11 0.42

3 Total effect

Canada (65 industries)

1.59 1.17

Hotel, restaurant Other services

61 62

1.81 1.49 2.40

Office supplies Advertising and travel Operating supplies

63 64 65

40 The effects of energy price changes The first thing that strikes one in an examination of Table 4 is the pattern of total price effect differences. Up to sector 18 the total price effects for Ontario are larger in all but three of the sectors and in two of these, sectors 10 and 18, the differences are very small. Only in Mining, sector 2, is the price change significantly smaller for Ontario. Thus, in the agriculture and agricultural products, textiles, and clothing industries, the total price changes are larger for Ontario than for Canada. In the wood and wood products industries, 19 to 24, the total effects for Canada tend to be slightly higher than for Ontario, but the differences on average are not great. For the basic metals and manufacturing sectors, 25 to 38, however, significantly higher total effects appear for Canada, except in sector 38, Non-metallic mineral products. The total effects are higher in the petroleum and coal sector in Canada, but are higher in Ontario for the petroleum derivative sectors, industries 40 to 43. The figures are almost identical for industries 44 and 45 and slightly higher in Ontario for 46, Transportation, storage and trade. The differences in the remaining two sectors are not great, and in any case it is not clear what particular significance can be attached to these numbers, particularly for 49, the unallocated sector. Table 4 presents a clear picture of how Ontario would fare with a uniform price increase for petroleum across Canada . It should also be remembered that since the Canadian figures include Ontario, the differences between Ontario and the rest of Canada would be significantly greater than those shown in column 2. With regard to what we might call the Ontario balance of trade with the rest of Canada, an over-all increase in petroleum and coal prices seems likely to be beneficial to Ontario. Ontario will be at a disadvantage in the agriculture and the agricultural products sectors, but this is not an area in which Ontario exports large quantities to the rest of Canada. Nor would we expect the relative increase in prices in Ontario to result in significant increases in imports of these products from the rest of Canada. Ontario tends to be fairly self-sufficient in these areas. Table 2 shows that of the first seventeen sectors Ontario is a net importer in only seven, and of these most are clearly imports from the rest of the world rather than from other provinces in Canada. The only significant exception to this is sector 5, Feed grains, where the imports are largely from western Canada. In the basic metals and manufacturing industries, industries 25 to 38, Ontario is a net exporter in all but two, and for one of these, 38, the net trade sector is almost zero. These are the main sectors from which we would expect exports to the rest of Canada to be generated, and it seems clear that in these industries shipments from Ontario to the rest of Canada would increase. In sectors 40 to 45 Ontario is currently a net exporter. One might expect the higher total effect in these sectors to have a detrimental effect on production in these industries.

Analysis of energy costs 41 However, in these sectors price differences are really very small except for sector 40. It seems unlikely that there will be significant trade changes in the remaining four sectors. It seems pretty clear, then, that a uniform increase in the price of petroleum and coal would be beneficial to Ontario vis-a-vis the rest of Canada. We have considered the case where the price increases were assumed to occur only in Ontario and the case where the price rise was assumed to be uniform across Canada, and it would be useful to consider the possibility of the price rise occurring in all of Canada except the oil-producing provinces. This case is difficult to analyse, however, for we have no information on the industrial structure of western Canada, nor do we have trade statistics specifically relating the eastern and western provinces. As we argued earlier, if Ontario were the only province in which price changes took place, then the total effects of Table 1 would give us some estimate of the price disadvantage that Ontario producers would face . We have already observed that for Ontario's basic export industries these price changes are not very large, generally are less than 2 per cent. Even this, however, is likely to be a significant overestimate of the price difference. These estimates assume not only that the original price change in petroleum takes place only in Ontario but that all the secondary price changes take place only in Ontario as well. In fact, of course, many industries in the rest of Canada buy intermediate products from Ontario, and since the price of these goods will have risen, production costs will rise throughout Canada and not just in Ontario. Unfortunately, we are unable to make any estimate about what these effects might be. It does seem clear, however, that even if Ontario were the only province in which petroleum price rises occur, the over-all price effect for most industries would be very moderate. And if Ontario enjoys a comparative advantage in the production of a wide range of manufactured goods, as is commonly supposed, it seems very unlikely that such a petroleum price increase would have much of an effect on this position. We now turn to a brief analysis of comparable data for the United States, using the 1965 input-output table. Again, we have assumed a 100 per cent price increase, first in crude petroleum and natural gas, then in coal, and finally in both. These results are given in Table 5. The first column presents the direct effects for coal, the second the total effects for coal, the third the direct effects for petroleum, the fourth the total effects for petroleum, and the fifth the total effects for petroleum and coal combined. So that Table 5 will be completely analogous to Table 4 we have, in column 6, represented the effect of a I 00 per cent price increase in industry 31, Petroleum products. As in the Canadian table, the direct effects are, in general, very low. In fact, of the eighty-one industries

TABLE 5 United States petroleum and coal price effects

Industries 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Livestock and livestock products Other agricultural products Forestry and fishery products Agricultural, forestry, and fishery services Iron and ferroalloy ores mining Non-ferrous metal ores mining Coal mining Crude petroleum and natural gas Stone and clay mining and quarrying Chemical and fertiliser mineral mining New construction Maintenance and repair, construction Ordnance and accessories Food and kindred products Tobacco manufacturers Broad and narrow fabrics, yarn & thread mills Miscellaneous textile goods and floor coverings Apparel Miscellaneous fabric and textile products Lumber and wood products, except containers

Direct effects, coal 0.02

2

3

4

Total effects, coal

Direct effects, petroleum

2.42

Total effects, petroleum

5 Total effects, petroleum and coal

6 Total effects, petroleum industries

1.80 2.58 1.66 1.40 1.09 1.09 1.16 3.25 2.07 1.43 1.65 2.04 0.65 1.35 0.73

1.98 2.73 1.78 1.59 1.74 1.52 19.86 3.36 2.70 1.90 2.10 2.29 0.94 1.63 0.86

3.54 5.21 3.43 2.76 1.82 1.56 1.83 1.31 3.69 1.93 3.16 4.11 1.07 2.50 1.40

0.02 0.05 0.02

0.18 0.16 0.12 0.19 0.65 0.43 18.70 0.11 0.63 0.47 0.46 0.25 0.30 0.28 0.13

0.10

0.47

1.56

2.03

2.75

0.04 0.01 0.01

0.40 0.26 0.84

1.40 0.88 1.39

1.80 1.14 2.22

2.52 1.53 1.97

0.02

0.20

1.07

1.27

1.96

0.36 0.09 15 .56 0.23 0.09

TABLE 5 continued

Industries 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39

Wooden containers Household furniture Other furniture and fixtures Paper and allied products except containers and boxes Paperboard containers and boxes Printing and publishing Chemicals and selected chemical products Plastics and synthetic materials Drugs, cleaning, and toilet preparations Paints and allied products Petroleum refining and related industries Rubber and miscellaneous plastics products Leather tanning and industrial leather products Footwear and other leather products Glass and glass products Stone and clay products Primary iron and steel mfg Primary non-ferrous metal mfg Metal containers

2

3

4

Direct effects, coal

Total effects, coal

Direct effects, petroleum

Total effects, petroleum

0.01 0.04 0.04

0.42 0.35 0.56

0.94 0.91 0.87

1.35 1.26 1.43

1.59 1.56 1.44

0.58 0.02

1.09 0.56 0.32

1.45 1.35 0.86

2.54 1.91 1.17

2.43 2.40 1.48

0:41 0.38 0.04 0.01 0.04

0.93 0.92 0.33 0.47 0.26

5.01 3.12 1.42 3.87 50.74

5.94 4.04 1.75 4.34 51.00

9.32 5.87 2.61 4.34 109.53

0.09

0.48

1.34

1.82

2.33

0.21 0.01 0.10 0.86 1.91 0.09 0.01

0.44 0.25 0.45 1.52 3.14 0.48 1.41

0.78 0.67 1.18 13.06 1.28 1.30 1.05

1.22 0.92 1.63 14 .58 4.42 1.78 2.46

1.32 1.13 1.60 2.80 1.96 1.94 1.70

0.19

44.94

5 Total effects, petroleum and coal

6 Total effects, petroleum industries

TABLE 5 continued

Industries 40 Heating, plumbing and fabricating structural metal products 41 Screw machine products, bolts, nuts , etc. , and metal stampings 42 Other fabricating metal products 43 Engines and turbines 44 Farm machinery 45 Construction, mining, oil field machinery equipment 46 Materials handling machinery and equipment 4 7 Metalworking machinery and equipment 48 Special industrial machinery and e4uipment 49 General industrial machinery and equipment 50 Machine shop products 51 Office , computing and accounting machines 52 Service industry machines 53 Electrical transmission and distribution e4uipment and electrical industrial apparatus 54 Household appliances

2

3

4

Direct effects, coal

Total effects, coal

Direct effects, petroleum

0.01

Total effects, petroleum

5 Total effects, petroleum and coal

6 Total effects, petroleum industries

0.98

0.98

1.96

1.57

0.02 0.02 0.09 0.06

0.89 0 .80 0 .69 0.79

0.95 0.99 0 .87 0 .84

1.84 1.79 1.57 1.64

1.54 1.61 1.4 7 1.38

0.05

0.82

0.83

1.65

1.34

0.03 0.02

0.66 0.53

0.86 0 .79

1.52 1.33

1.46 1.30

0.02

0.55

0.97

1.52

1.70

0.02 0.01

0.63 0.50

0.93 0.71

1.56 1.22

1.58 1.11

0.02 0.02

0.23 0.58

0.52 0 .90

0.74 1.47

0.86 1.48

0.03 0.04

0.47 0.60

0 .92 0.91

1.39 1.52

1.56 1.48

TABLE 5 continued

Industries 55 Electric lighting and wiring equipment 56 Radio, telephone, and communications equipment 57 Electronic components and accessories 58 Miscellaneous electrical machinery, and e4uipment, and supplies 59 Motor vehicles and equipment 60 Aircraft and parts 61 Other transport e4uipment 62 Professional, scientific, and controlling instruments and supplies 63 Optical, ophthalmic and photographic equipment and supplies 64 Miscellaneous mfg 65 Transport and warehousing 66 Communications, except radio and television broadcasting 67 Radio and television broadcasting 68 Electric, gas, water and sanitary services 69 Wholesale and retail trade 70 Finance and insurance 71 Real estate and rental 72 Hotels and lodging places; personal and repair services, except automobile repair

2

3

4

Direct effects, coal

Total effects, coal

Direct effects, petroleum

Total effects, petroleum

0.02

0.49

1.26

1.75

2.26

0.01 O.Ql

0.22 0.32

0.60 0.81

0.82 1.13

1.00 1.31

0.03 0.04 0.01 0.04

0.38 0.70 0.29 0.73

0.81 0.84 0.71 0.88

1.20 1.54 1.00 1.61

1.31 1.39 1.18 1.45

0.01

0.32

0.72

1.04

1.20

0.11 0.02 0.02

0.37 0.36 0.19

0.79 0.93 2.61

1.16 1.28 2.79

1.36 1.61 5.20

0.57 0.43 9.43 0.98 0.62 0.80

0.65 0.52 12.92 1.12 0.79 0.90

0.99 0.59 1.90 1.68 0.81

0.97

1.13

1.69

0.01

0.07 0.09 3.48 0.14 0.17 0.10

0.01

0.16

2.15

0.05

6.57 O.Ql 0.10

5 Total effects, petroleum and coal

6 Total effects, petroleum industries

1.36

TABLE 5 continued

Industries 73 75 76 77 78 79 80 81 82

Business services Automobile repair and services Amusements Medical, educational services, and non-profit organizations Federal Government enterprises State and local government enterprises Gross imports and goods and services Business travel, entertainment and gifts Office supplies

Arithmetic average Weighted average

Direct effects, coal

0.03 1.14 I.I 2

Total effects, petroleum

5 Total effects, petroleum and coal

6 Total effects, petroleum industries

0.19 0.20 0.12

0.71 1.04 0.62

0.90 1.24 0.74

1.10 1.92 0.96

0.20 1.44 1.76

0.78 0.65 2.40

0.98 2.08 4.16

1.05 J.10 2.08 3.31

1.72 1.00

1.92 1.49

2.12 2.12

2.90 2.55

2

3

4

Total effects, coal

Direct effects, petroleum

0.53

0.20 0.49 0.33 0.13

0 .78 0.43

0.70 0.75

1.72 2.01

Analysis of energy costs 4 7 only eighteen have direct effects greater than O.l per cent. 3 This, of course, is due mainly to the greater disaggregation of the US table . For as we disaggregate more and more we would expect to observe more industries with very small direct effects. It is clear, after all, that very few industries use crude petroleum or natural gas directly in their production process, except for the use of natural gas as a fuel. The significant point to be observed from the us table is that the total effects on average tend to be smaller than for the Canadian table. Indeed, there are only a half a dozen industries in which the total effect in Canada is less than it is in the United States. This fact is also reflected in both the unweighted and weighted averages presented at the end of both the Canadian and us tables. The conclusion to be drawn from these facts is that an over-all increase in the price of petroleum will have a larger effect on commodity prices in Canada than in the United States. Among the reasons to explain this difference in total price effects are, first, the fact that Canada is a colder country where it takes a considerable amount of coal and petroleum simply to heat factories and offices and, second, because of the abundance of hydrocarbon fuel Canadians may not economize on these products. The second reason is correctable, but the first is not, and a two-price system for Canadian petroleum might thus have some justification. The data of Tables 3 and 5 would suggest that if Canada adopts world prices for petroleum, as petroleum prices continue to rise Canada will find itself at more and more of a price disadvantage relative to the United States. Of course, as we have already observed, the over-all price effects are not large, and our evidence suggests that a 100 per cent increase in the price of petroleum in both Canada and the United States would be about equivalent to a 1 per cent price increase in the Canadian exchange rate. The conclusion to be drawn from this would seem to be that any industrial shifts generated by the petroleum crisis are likely to be because of either real or imagined shortages and not because of possible commodity price effects that increases in petroleum prices would generate. INCOME AND EMPLOYMENT: SOME PRELIMINARY RESULTS

We have shown doubling the price of petroleum and coal would affect commodity prices in Ontario, and we have indicated how these price changes would compare with price changes under the assumption that the initial hydrocarbon-fuel price changes were uniform throughout Canada. We have also 3 Note that there is no industry 74. This resulted from a change between the US 1958 and 1965 tables.

48 The effects of energy price changes shown how the price changes in Canada would compare with price changes in the United States under the same assumptions. In this section the calculated price changes will be used to make some preliminary estimates on the resulting effects on income and employment. We want to stress the fact that these are very preliminary results, and our modest hope is that they will give us some crude estimate of the changes that we might expect. Obviously a careful analysis of the many variables which would enter into a detailed calculation of income and employment effects is well beyond the scope of this study. In order to make estimates of income and utility changes which could result from the increase in the price of petroleum and coal products, we must make some assumption about demand elasticities. As no estimates of elasticities of demand by industry are available for Ontario, we are forced to make some arbitrary assumption, and the one most convenient for our purposes is to assume that utility functions are Cobb-Douglas. Thus we assume that the utility function for Ontario can be represented as

U = X I a1 X2 a2

· · ·

X

n

an

,

(19)

where n ), 49 and where ~ ai = 1. This utility function is useful for our purposes because, as is well known, it implies cross-price elasticities of demand which are zero and own price elasticities of demand which are unity . The demand functions derived from this utility function for any good Xi can be written

(20) or in other words total spending on any commodity is a constant proportion of income M. It thus follows that the percentage change in the consumption of any commodity X; is equal and opposite in sign to the percentage change in its price . Table 6 presents estimates of the consumption changes for Ontario for 1971. To construct these estimates we first need consumption figures by sector, and since these are available only for 1965, the year for which the input-output table was constructed, the figures for 1971 have been derived assuming that the ratio of expenditure by consumers on the output of the various industries has not changed since 1965. We have also included in our consumption figure that component of government expenditure which is not allocated to either imports or investment. Although the provincial accounts define G to include all expenditure by government, the input-output structure does not, so that using the 1965 data on government spending by sector and the 1965 and 1971 total government spending figures we created a government 'consumption' vector which we added to the personal consumption vector. This component of G

Analysis of energy costs 49 presumably adds to the welfare of society and thus should be included in C. The figures in column 1 have been constructed by multiplying personal consumption plus government expenditure by sector for 1965 by the ratio of total provincial consumption in 1971 divided by total provincial consumption in 1965. Note that this assumption is consistent with our assumption about the utility function only if relative prices have not changed since 1965. Column 2 of Table 6 simply reproduces the price change vector from Table I. Column 3 is a calculation of the consumption change found by multiplying columns l and 2. In the last row of the table in column 2 we show the average price change weighted by the consumption by sector. This number has two interpretations. First of all, it is an estimate of the real income change associated with the price changes of column 2. Secondly, for those who are not reluctant to think in terms of cardinal util.ity, this number represents the percentage change in the utility index of equation (I 9). While this real income change is certainly significant, it does not seem large relative to the recent increases in the Canadian consumer price index. As we shall see , however, this figure is seriously downward-biased because it relates only to first-round changes. The calculations reported in Table 6 assumed that money income was unchanged . This is a reasonable initial assumption because implicit in our argument have been the assumptions that the only primary factor price which has risen has been petroleum and coal and that the return to capital and the wage rate have not changed. Thus, with the same level of production, factor incomes will be unchanged. The point, of course, is that we would not expect production to be unchanged, for we have just shown in column (3) of Table 6 a vector of consumption changes all of which are negative . These consumption changes will obviously imply production changes and of course the production changes will not only be the changes shown in column 3 but would also include the changes in the use of intermediate products. What we now want to calculate is the total production changes which would be attributable to the vector of price changes shown in column 2 of Table 6. In order to calculate the total change in production we need to know the total change in aggregate demand by sector. In particular, we must make some assumption about changes in personal consumption, government spending, investment, exports, and imports. From the assumption that the utility functions have the form shown in equation (I 9), personal consumption expenditure will decrease by the same percentage that price increases. It seems reasonable to suppose that investment expenditure will not be much affected by the price of investment goods but will depend more on exogenous factors such as the interest rate, and thus we assume that there is no change in investments. It is difficult to formulate any hypothesis about how government expenditure will

50 The effects of energy price changes TABLE 6 Consumption changes for Ontario, 1971

Industries

l 2 3 4 5 6 7 8 9 10 11

12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

Agriculture, forestry, and fishing Mining Meat and poultry Dairy products Grain mills Biscuits and bakeries Sugar and confectioneries Other food Soft drinks Distilleries, breweries, and wineries Tobacco and tobacco products Rubber products Leather and leather products Cotton yarn and cloth Synthetic textiles Knitting mills Clothing Other textile mills Sawmills Furniture and fixtures Other wood Pulp and paper mills Paper products Printing and publishing Iron and steel mills Other primary metals Fabricated and structural metals Metal stamping, pressing, and coating Other metal fabricating Miscellaneous machinery Motor vehicles and aircraft Other transportation equipment Electrical appliances Electrical industrial equipment Communication equipment Other electrical products Clay , lime, and cement Other non-metallic mineral products Petroleum refineries and coal products Plastics and synthetic resins

Consumption 1971

336.3 47 .9 712 .7 238.2 232.6 207.0 179.5 630.4 115 .9 210.9 80 .2 41.l 155.9 38.6 83 .8 104.5 803 .2 133.5 27.9 236 .9 11.4 63 .3 44.4 87.3 5.0 102.3 17.7 262.8 24.5 1,838 .9 77.7 140.7 13.5 104.6 20.3 103.8 43 .5 438.7 27.l

2 Price change 7.10 2.47 5.28 5.19 4.22 2.70 1.75 3.49 2.11 1.32 4.00 2.55 2.40 1.73 3.50 1.78 1.40 1.87 2.68 1.50 1.92 1.91 2.70 0.90 1.14 3.44 0.69 1.21 1.51 1.55 1.34 1.42 1.85 1.42 1.96 2.10 2.68 3.09 10.87 6.35

3 Consumption change -23 .9 -1.2 -37 .7 -12.4 -9 .8 -5 .6 -3.l -22 .0 -2.4 -2.8 -3 .2 -1.0 -3.7 -0.7 -3 .3 -1.9 -11.3 -2.5 -0.7 -3.6 -0.2 -1.2 -1.2 -0.8 -0.l -2.2 -0.7 - 0.2 - 4.0 -0.4 -24.7 -1.l -2.6 -0.2 -2 .0 -0.4 -2.8 -1.3 -47 .7 -1.7

Analysis of energy costs 51 TABLE 6 continued

Industries

Consumption 1971

2 Price change

3 Consumption change

41 42 43 44 45 46 47 48 49

24 .7 161.2 140.6 241.6 384.4 5,662.2 662.1 8,512.4 2.9

3.87 2.03 7.95 2.07 2.55 3.00 1.10 0.86 2.11

-1.0 -3.3 - 11.2 -5.0 -9 .8 - 91.4 -7.3 -137.5 -0.1

Paint and varnish Pharmaceuticals and medicines Other chemical Miscellaneous manufacturing Construction, maintenance, and repair Transportation, storage, and trade Utilities Communications and other services Unallocated sector

Arithmetic average Weighted average

2.76 2.95

change, so that we have assumed that government expenditure by sector will remain constant in dollar value. This can be interpreted as the assumption that each government agency is allocated a money budget and that it spends it all. Thus any price changes will simply result in corresponding quantity changes. This assumption implies exactly the same behaviour by governments as was assumed for consumers, but for quite different reasons. For the trade sector, the assumptions we make about how exports and imports change will clearly depend on what we assume about prices in the rest of the country. Two extreme cases are possible; first, that no other prices anywhere in Canada change and, second, that all prices in Canada change by the same amount as in Ontario. For each of these assumptions we have the further problem of deciding how exports and imports will respond to the price changes. In other words, we must make some assumptions about the elasticities of import demand for Ontario and for the rest of the world. It is obvious from the importance of the trade sector in the Ontario economy that the choice of assumptions will make a significant difference to our final results. We have therefore done our calculations on three different assumptions about import behaviour, in the hope that the range of results so generated will capture the actual situation. Table 7 assumes that prices have increased only in Ontario and that individuals in the rest of the country have the same preference functions as Ontario residents. Thus, since the prices of Ontario exports have increased, the consumption of these goods by the rest of world will decrease, and thus exports will fall. Note that in Table 7 all numbers are negative. The prices of Ontario

TABLE 7 Production and employment effects for Ontario, 1971 , no import changes 3

4

Consum .

2 Dom. Consum. Ch .

Imports Ch.

I 2 3 4 5 6 7

73.0 22.3 59.2 25 .2 11.5 8.9 4.5

73.0 22.3 59 .2 25 .2 11.5 8.9 4.5

8 9 10 12 13 14

39.7 3.5 5.1 3.5 5.6 4.8 3.0

15 16 17 18 19 20 21

14.9 3.4 20.8 4.7 2. 1 5.9 2.1

Tot.

11

Exports Ch.

5 Trade Ch .

6 Fin . Tot. Dem . Ch.

7 Dom. Prod. Ch .

8 Emp . Effects

0.0 0.0 0.0 0.0 0.0 0.0 0.0

37.7 33.8 29.3 17.4 2.3 4.5 1.9

37 .7 33.8 29 .3 17.4 2.3 4.5 1.9

110.7 56.1 88.5 42 .6 13 .8 13.4 6.4

279 .8 212 .1 138.8 60.8 51.1 15.7 15.1

6.9 5.2 2.2

39 .7 3.5 5.1 3.5 5.6 4.8 3.0

0.0 0.0 0.0 0.0 0.0 0.0 0.0

24 .0 1.4 3.1 8.2 12.6 3.0 1.8

24 .0 1.4 3.1 8.2 12.6 3.0 1.8

63.7 4.9 8.2 11.8 18.2 7.9 4.8

107.6 6.5 11.6 13.8 37.4 11.9 20.9

2.4 0.1 0.2 0.1 0.7 0.7 0.4

14

14 .9 3.4 20.8 4.7 2.1 5.9 2.1

0 .0 0.0 0.0 0.0 0.0 0.0 0.0

9.3 1.3 5.5 1.7

9.3 1.3 5.5

1.2

1.2 2.2 1.8

24 .2 4.7 26.2 6.4 3.3 8.1 3.9

46.0 6.0 28.4 29 .2 13.6 8.9 I 3.4

0.9 0.4 1.7 0.6 0.6 0.4 0.6

15 16 17 18 19 20 21

2.2 1.8

1.7

1.1 I.I

0.8 0.3

Tot. 1 2 3 4 5 6 7 8 9 10 11

12 13

TABLE 7 continued 3

4

Imports Ch .

Exports Ch.

5 Trade Ch.

6 Fin. Dem. Tot. Ch .

7 Dom. Prod. Ch .

8 Emp. Effects

Tot.

Tot.

Consum.

2 Dom. Consum . Ch.

22 23 24 25 26 27 28

3.3 4.2 2.0 27.9 12.3 I.I 2.3

3.3 4 .2 2.0 27 .9 12.3 I. I 2.3

0.0 0 .0 0.0 0.0 0 .0 0.0 0.0

7.5 10.7 4.4 15 .8 25 .9 1.2 6.0

7.5 10.7 4.4 15.8 25.9 1.2 6.0

10.8 14.9 6.5 43.7 38.2 2.3 8.3

45 .6 59.6 54.8 144.9 144.0 5.8 29 .3

1.4 1.9 2.3 3.8 5.2 0.2 0.8

22 23 24 25 26 27 28

29 30 31 32 33 34 35

8.4 4.7 41.8 2.1 9.9 13.0 20.4

8.4 4.7 41.8 2.1 9.9 13 .0 20.4

0.0 0.0 0.0 0.0 0.0 0.0 0.0

11.4 11.2 51.1 30.5 3.5 6.0 8.6

11.4 11.2 51.1 30 .5 3.5 6.0 8.6

19 .7 15 .9 93 .0 32 .6 13.4 19.0 29 .0

87 .6 22.6 147.9 42.2 15.6 22 .7 43 .2

4.3 0.7 2.0

1.7

29 30 31 32 33 34 35

36 37 38 39 40 41 42

29 .1 3.8 8.1 87 .1 9.2 3.4 5.2

29 .1 3.8 8.1 87 .1 9.2 3.4 5.2

0.0 0 .0 0.0 0.0 0.0 0.0 0.0

13 .5 1.5 9 .5 41.5 3.6 4.2 3.3

I 3.5 1.5 9 .5 41.5 3.6 4 .2 3.3

42.6 5.3 17 .6 128 .6 12 .8 7.6 8.5

52.5 18.1 35 .1 190.9 38 .3 18.1 I 3.5

2.0 0.5 0.9 2.6 0.7 0.5 0.4

36 37 38 39 40 41 42

1.3

0.6 0.9

TABLE 7 continued

Tot.

Consum.

2 Dom. Consum. Ch.

43 44 45 46 47 48 49

52 .7 11.7 9.8 145 .3 8.6 154.1 23.6

52 .7 11.7 9.8 145 .3 8.6 154.1 23.6

Total Empl. Eff. 98 .8

3

4

Imports Ch . 0.0 0.0 0.0 0.0 0.0 0.0 0.0

Total Trade Eff. 579.2

Exports Ch.

5 Trade Ch .

6 Fin. Dem. Tot. Ch.

7 Dom. Prod. Ch.

8 Emp. Effects

Tot.

69.4 17.0 0.0 0.0

69.4 17.0 0.0 0.0

122.1 28 .7 9.8 145.3 9.7 168.3 26 .2

255.1 55.7 66 .5 388.4 58.9 385.9 248.6

5.6 3.1 1.6 9.5 1.4 9.4 6.1

43 44 45 46 47 48 49

1.1

I.I

14.2 2.5

14.2 2.5

Tot. Cons. Ch. 1028 .63

NOTE: All figures are in millions of dollars except employment, which is in thousands of persons. All numbers are negative.

Analysis of energy costs 55 imports, although unchanged in dollar terms, are now relatively lower than before, and thus the change in imports will depend on what substitution assumption we make. But this qualitative information is not of much help to us, for we must have some estimate of what the changes will actually be. If we assume that there is no competition between imports and domestic production for both Ontario residents and consumers in the rest of Canada and the world, 4 and since we have assumed that the rest of the world has the same preferences as Ontario residents, our task is simplified significantly. Under these assumptions we can assume that exports fall in the same proportion as domestic consumption falls, for consumers in the rest of Canada will reduce their purchases from Ontario by the same percentage as prices have risen. Imports, on the other hand, will not change, since the same amount will be spent on these commodities as before. Note that we are still assuming a constant level of money income. The changes in the various components of aggregate demand are shown in Table 7. Column 1 is the total change in consumption expenditure in Ontario, and is made up of the change in the consumption of domestic production, column 2, plus the change in imports, column 3. Since in this table we have assumed no change in imports, columns 1 and 2 are identical. Column 4 is the export change, column 5 is the trade balance change, and column 6 is the change in domestic demand, the sum of columns 2 and 4. Column 6 represents the direct change in total demand, and to find the final change in production we multiply this demand change vector by the [/ -A 1- 1 matrix to obtain the total production change vector shown in column 7. Finally, to estimate what this production change would imply for employment in Ontario we have taken employees by sector as a ratio of total sales and multiplied by the production change vector in order to get employment changes. These changes are shown in column 8, with the total employment change being shown in the last row of the table. Note that the changes in consumption and production are in millions and the employment changes are in thousands. In making the calculations in Table 7 and all subsequent tables, an adjustment was made for the consumption change_ in two of the industries. It was felt that for industry 46, Transportation, storage, and trade, and for industry 48, Communications and other services, it would not be appropriate to assume that the change in demand would be, in percentage terms, equal and opposite in sign 4 By no competition we mean that imports are different commodities in the sense of being

different arguments of equation ( 19), and not that there would be no substitution between commodities if relative prices were to change. This implies that there are more arguments in equation (19) than there are sectors in the input-output table.

56 The effects of energy price changes to the price change. For these industries the change in activity is much more closely related to changes in over-all business activity than to the change in the price of these services. Thus an initial estimate of total change in production was made, and this change, as a percentage of the initial production level, was used as the change in demand for these two sectors. Before moving on to a consideration of alternative import assumptions, several aspects of Table 7 deserve some comment. First, the employment reduction of 98.8 thousand represents approximately 3 per cent of the 1971 Ontario labour force, so that the employment effect is not insignificant. 5 For the individual industries the employment changes are very much what we would expect. Industry 1, Agriculture, shows a large change due to both the size of the sector and the relatively large price change. Too much significance should not be attached to this figure, for it is unlikely that our demand assumption is particularly appropriate for this industry. The relatively large employment change for Mining, industry 2, is interesting because this is usually considered to be a very capital-intensive industry. The main reason for the large employment effect is attributable to the large production change generated by the interindustry flows. Note that the ratio of total production change (column 7) to final demand (column 6) is among the highest in the table. Other relatively large employment changes due primarily to intermediate effects occur in industries 25 and 26, the primary metals industries. The table also emphasizes the large employment/production ratio differences which exist in the economy. Compare, for example, industry 17, Clothing, with industry 40, Plastics and synthetic resins. Let us now assume that prices in the rest of the economy have changed by the same amount as in Ontario. Again we must settle the question of how exports and imports would be expected to change. Under our no-competition assumption there is no reason to expect exports to change from the level shown in Table 7. Imports, however, will now be more expensive than previously, so that imports will fall by the amount of the price rise, assumed to be the same for all of Canada as for Ontario. These imports would thus be equal to those shown in column 3 of Table 8. The import changes shown here, however, are quantity changes and do not represent any change in the value of imports. At the same time, while the physical quantity of imports has fallen, because of our no-competition assumption, this change in imports does not represent any change in production in Ontario, for demand for domestically (Ontario) produced goods will not have changed. Thus under our present assumptions the 5 for an alternative approach to the estimation of employment and price changes due to the increase in the price of petroleum see Jump and Wilson (1975).

TABLE 8 Production and employment effects for Ontario, 1971, imports reduced 3

4

Consum . Ch.

2 Dom. Consum. Ch.

Imports Ch .

Exports Ch.

2 3 4 5 6 7

73.0 22.3 59.2 25 .2 11.5 8.9 4.5

23.9 1.2 37 .7 12.4 9.8 5.6 3.1

49 . l 21.1 21.6 12.8 1. 7 3.3 1.4

37.7 33.8 29.3 17.4 2.3 4.5 1.9

8 9 10 11 12 13 14

39.7 3.5 5.1 3.5 5.6 4.8 3.0

22 .0 2.4 2.8 3.2 1.0 3.7 0.7

17.7 1.1 2.4 0.3 4.5 1.1 2.3

15 16 17 18 19 20 21

14.9 3.4 20.8 4.7 2.1 5.9 2.1

3.3 1.9 11.3 2.5 0.7 3.6 0.2

11.6 1.6 9.5 2.2 1.3 2.4 1.9

Tot.

6 Fin. Tot. Dem. Ch.

7 Dom. Prod. Ch .

8 Emp. Effects

- 11.4 12.7 7.7 4.6 0.6 1.2 0.5

61.6 35.0 66 .9 29.8 12. 1 10. l 5.0

184.8 139 .2 103.9 42.7 37.5 11.7 11.4

4.5 3.4 1.7 0.8 0.8 0.6 0.2

I 2 3 4 5 6 7

24.0 1.4 3. 1 8.2 12.6 3.0 1.8

6.3 0.3 0.8 7.9 8.1 1.9 - 0.5

46.0 3.8 5.9 11.4 13.6 6.8 2.5

77 .1 4.9 8.0 13.4 26.8 9.8 12.7

1.7 0.1 0.2 0.1 0.5 0.6 0.2

8 9 10 11 12 13 14

9.3 1.3 5.5 1.7

- 2.3 - 0.3 - 4.0 - 0.4 - 0.1 - 0.2 -0. l

12.6 3.1 16.7 4.2 2.0 8.7 11.9

26.8 3.9 18.1 19.4 9.1 6.3 8.7

0.5 0.2 1.1 0.4 0.4 0.3 0.4

15 16 17 18 19 20 21

1.2

2.2 1.8

5 Trade Bal. Ch.

Tot.

TABLE 8 continued

Exports Ch.

5 Trade Bal. Ch.

6 Fin. Tot. Dem. Ch.

2.1 3.0 1.2 27 .8 10.0 0.4 2.1

7.5 10.7 4.4 15 .8 25 .9 1.2 6.0

5.4 7.8 3.2 -12. 1 15.9 0.8 3.8

8.7 11.9 5.2 15.8 28 .2 1.9 6.2

4.0 6.4 24.7 J.J 2.6 0.2 2.0

4.4 4.3 17.2 1.0 7.3 12.8 18.3

11.4

11.2 SJ.I 30.5 3.5 6.0 8.6

7.0 6.9 33.9 29.5 -3 .9 -6.7 -9.7

0.4 2.8

28.7 1.0 6.7 39.4 7.5 2.5 2.0

13.5 1.5 9.5 41.5 3.6 4.2 3.3

-JS.I 0.4 2.8 2.1 -3.9 1.7 1.3

3

4

Tot.

2 Dom. Consum. Ch. Consum. Ch.

Imports Ch .

22 23 24 25 26 27 28

3.3 4.2 2.0 27 .9 12.3 J.J 2.3

1.2 1.2 0.8 0. 1 2.2 0.7 0.2

29 30 31 32 33 34 35

8.4 4.7 41.8 2.1 9.9 13.0 20.4

36 37 38 39 40 41 42

29.J 3.8 8. 1

87.1 9.2 3.4 5.2

1.3

47.7 1.7

1.0 3.3

8 Emp. Effects

Tot.

32.6 42.1 35 .2 81.6 93.9 4.4 20.7

1.0 1.3 1.5 2.2 3.4 0.2 0.6

22 23 24 25 26 27 28

15.3 11.6 75 .8 31.6 6.0 6.2 10.7

6J.J 16. I 117.7 40.0 7.7 8.4 18.J

3.0 0.5 1.6 1.3 0.3 0.3 0.7

29 30 31 32 33 34 35

14.0

19.6 12.9 22.3 130.6 21.7 12.0 9.9

0.8 0.4 0.5 1.8 0.4 0.3 0.3

36 37 38 39 40 41 41

4.2

10.9 89 .2

5.3 5.1 6.5

7 Dom. Prod. Ch.

TABLE 8 continued

Tot.

Consum. Ch.

2 Dom. Consum. Ch.

43 44 45 46 47 48 49

52 .7 11.7 9.8 145 .3 8.6 154. 1 23 .6

11.2 5.0 9.8 91.4 7.3 137.5 0.1

3

4

Imports Ch . 41.5 6.7 0.0 53 .8 1.3 16.7 23 .6

Total empl. eff. 66.0 Total trade eff. 65.3 NOTE: See note for Table 7.

Exports Ch.

5 Trade Bal. Ch .

6 Fin. Tot. Dem. Ch.

7 Dom. Prod. Ch.

8 Emp. Effects

Tot.

69.4 17.0 0.0 0.0 I.I 14.2 2.5

28.0 10.3 0.0 -53 .8 -0.2 -2.4 -21.0

80.6 22 .0 9.8 91.4 8.4 151.7 2.6

164.9 39.2 49 .8 251.3 41.2 295 .6 151 .2

3.6 2.2 1.2 6.1 1.0 7.2 3.7

43 44 45 46 47 48 49

Tot. cons. ch. 1028.63

60 The effects of energy price changes final production changes and the employment changes shown in Table 7 hold whether or not prices in the rest of Canada rise. Of course the assumption that imports and domestically produced goods represent different commodities in the sense that they are separate arguments of the utility function of equation (I 9) is not a very realistic one. While some imports do not compete with domestic produce, many do, and a consideration of the case where there is 'competition' is clearly necessary. To take the other extreme case we assume that there is no difference between imports and domestic production, and thus that imports are a component of each of the arguments of the utility function. In this case, when the price increases are the same throughout the world, Ontario residents will reduce their purchases of imports by the same proportionate amount that they reduce their spending on Ontario production. These changes are shown in column 3 of Table 8. These import changes will result in a change in domestic production in the sense that the import change, shown in column 3, plus the domestic consumption change, shown in column 2, must again sum to column I, which is assumed unchanged. The shift in assumptions from no-competition to full competition will not affect our assumption about exports, for the change in the Ontario production of goods for foreign purchases will be the same in either case. Thus the new vectors of total production change and employment change can be found as before by using the same export vector, and these are shown in Table 8. Since consumption of domestic production is higher in Table 8 than in Table 7 by an amount equal to the import change, we have a smaller reduction in total production and a smaller change in employment. For Table 7 the employment change was 98,800 while for Table 8 it is 66,000, a reduction of almost exactly one-third. And of course there is a substantial improvement in the trade balance. It is noteworthy, however, that the trade balance change is negative even in Table 8, indicating that imports have increased by more than exports. The employment changes for the individual industries show no great surprises and tend to be simply scaled-down versions of Table 7. In Table 8 we assumed that all imports had Ontario-produced substitutes, and that prices in the rest of Canada and the rest of the world rose by the same amount as in Ontario. We have not, however, considered the case where, with this 'competition' assumption, prices are assumed to rise only in Ontario. The difficulty with this case, of course, is that with perfect substitution assumed between imports and domestic production, there is nothing in our model to prevent all Ontario demand from being satisfied by imports, and this is not a very interesting case. Nevertheless, since there is a possibility of prices rising in Ontario relative to at least some other provinces, some estimate of the effect of this case would seem appropriate. For lack of a better alternative, we have, in Table 9

TABLE 9 Production and employment effects for Ontario, 1971 , imports increased

Exports Ch.

5 Trade Bal. Ch.

6 Fin. Tot. Dem. Ch.

7 Dom. Prod. Ch.

8 Emp. Effects

-49.1 -21.l -21.6 -12 .8 -1.7 -3.3 -1.4

37 .7 33 .8 29 .3 17.4 2.3 4.5 1.9

86 .8 54 .9 50.9 30.2 3.9 7.8 3.3

159.8 77 .2 110.1 55.4 15.4 16.7 7.8

374.9 285 .0 173.6 78.8 64 .7 19.6 18.8

9.2 7.0 2.8 1.4 1.0 0.4

l 2 3 4 5 6 7

57 .3 4.6 7.5 3.9 IO.I 5.9 5.3

-17 .7 -1.l -2.4 -0.3 -4 .5 -I.I -2 .3

24 .0 1.4 3.1 8.2 12.6 3.0 1.8

41.7 2.5 5.5 8.6 17.1 4.1 4.2

81.3 6.0 10.6 12.1 22 .7 8.9 7.1

138.2 8.1 15.2 14.1 48.1 14.0 29 .1

3.1 0.2 0.3 0.1 0.9 0.9 0.6

8 9 10 11 12 13 14

26.5 5.0 30.3 6.8 3.4 8.3 4.0

-11.6 -1.6 -9 .5 -2.2 -1.3 -2.4 - 1.9

9.3 1.3 5.5 1.7

20 .9 2.8 15.0 3.9 2.6 4.6 3.7

35.8 6.2 35 .7 8.6 4.7 10.5 5.8

65 .2 8.0 38.7 39.l 18.2 11.5 18.2

1.3 0.5 2.3 0.8 0.8 0.5 0.8

15 16 17 18 19 20 21

3

4

Consum.

2 Dom. Consum. Ch.

Imports Ch.

l 2 3 4 5 6 7

73.0 22 .3 59.2 25 .2 11.5 8.9 4.5

122.0 43.4 80.8 38.0 13.2 12.2 5.9

8 9 10 II 12 13 14

39.7 3.5 5.1 3.5 5.6 4.8 3.0

15 16 17 18 19 20 21

14.9 3.4 20 .8 4.7 2.1 5.9 2.1

Tot.

1.2

2.2 1.8

1.5

Tot.

TABLE 9 continued 2 Dom. Consum. Ch .

3

4

Tot.

Consum.

Imports Ch .

Exports Ch.

5 Trade Bal. Ch.

6 Fin. Tot. Dem. Ch.

7 Dom. Prod. Ch.

8 Emp. Effects

22 23 24 25 26 27 28

3.3 4.2 2.0 27.9 12.3 I.I 2.3

5.4 7.1 3.2 55 .7 22 .3 1.5 4.4

-2.1 - 3.0 - 1.2 - 27.8 -10 .0 -0.4 -2.1

7.5 10.7 4.4 15.8 25 .9 1.2 6.0

9 .6 I 3.7 5.7 43 .6 36.0 l.6 8.1

Tot.

12.9 17.8 7.7 71.5 48.2 2.7 10.4

58 .5 77.2 74.4 208.2 194.0 7.2 37 .9

l.8 2.4 3.1 5.5 7.0 0.2 l.O

22 23 24 25 26 27 28

29 30 31 32 33 34 35

8.4 4.7 41.8 2.1 9.9 I 3.0 20.4

I 2.7 9.0 59.0 3.1 17.2 25.7 38.7

-4.4 - 4 .3 - 17.2 - 1.0 - 7.3 -12.8 - 18.3

l l.4 11.2 51.1 30.5 3.5 6.0 8.6

15.7 15 .5 68 .3 31.5 10.8 18.8 27.0

24 .1 20 .2 110.1 33.6 20.7 31.8 47 .3

114.1 29 .0 178.0 44.4 23 .6 37.0 68 .3

5.6 0.9 2.4 1.4 0.9 1.5 2.7

29 30 31 32 33 34 35

36 37 38 39 40 41 42

29.1 3.8 8. 1 87.1 9.2 3.4 5.2

57.8 4.8 14.8 126.5 16.7 5.9 7.2

- 28.7 - 1.0 - 6.7 - 39.4 -7 .5 - 2.5 -2.0

I 3.5 l.5 9.5 41.5 3.6 4 .2 3.3

42 .2 2.5 16.2 80 .9 II.I 6.7 5.2

71.3 6.3 24 .3 168 .0 20.3 IO . I 10.5

85.4 23 .3 47 .9 251.2 54 .9 24.2 17. l

3.3 0.7 1.2 3.4 0.9 0.7 0.5

36 37 38 39 40 41 42

TABLE 9 continued

Tot. 43 44 45 46 47 48 49

Consum. 52.7 11.7 9.8 145.3 8.6 154.1 23 .6

2 Dom. Consum. Ch. 94.1 18.4 9.8 199.1 9.9 170.8 47.2

3

4

Imports Ch.

Exports Ch.

- 41.5 - 6.7 0.0 - 53 .8 - 1.3 - 16.7 - 23 .6

Total Empl. Eff. 131.6 Total Trade Eff. 1093.1 NOTE: See note to Table 7.

69.4 17.0 0.0 0.0 1.1

14.2 2.5

5 Trade Bal. Ch .

6 Fin. Tot. Dem . Ch.

7 Dom. Prod. Ch .

8 Emp. Effects

Tot.

110.9 23 .7 0.0 53 .8 2.4 30.9 26.1

163.6 35.4 9.8 199.1 10.9 185 .0 49 .8

345 .3 72.1 83 .3 525.5 76 .6 476 .1 346.0

7.5 4.0 2.0 12.9 1.9 11.7 8.5

43 44 45 46 47 48 49

Tot. Cons. Ch. I 028.63

64 The effects of energy price changes assumed that with price increases for Ontario only, imports have increased by an amount equal to the price increase . The calculations are exactly the same as before, and, as we would expect, the employment reduction and the trade balance deterioration are substantially higher. Before comparing our results here with those for Canada there are several limitations of the data used to derive Tables 7, 8, and 9. Those familiar with the 1965 Ontario input-output table will be aware that exports and imports are not shown for the forty-nine sectors and that only a trade balance is reported. Subsequently, however, exports and imports were reported, but on a much more aggregated basis. 6 The figures used for imports and exports have been created from these by disaggregating according to total production . The figures must therefore be regarded as only approximations. Furthermore the exports and imports reported in the 1973 version were not exports and imports of final consumer goods, but total exports and imports, so that in our export and import figures we have included intermediate products. Of course, for imported and exported intermediate products it is not , in general , appropriate to assume that quantities purchased will be sensitive to own prices as we have implicitly done, for clearly the demand for intermediate goods will be derived from the demand for final output. Unfortunately no alternative seems available . Our procedure will thus be expected to overestimate both export and import changes, and what the net effect of these biases will be is unclear. The calculations which we presented above assume that the rest of Canada and the rest of the world have the same industrial structure as does Ontario. We know from our earlier comparison of the Ontario and Canadian input-output tables, however, that significant differences do exist, and a better estimate of the over-all effect on Ontario can be found from combining the effects calculated from the Ontario and from the Canadian tables. First, for comparison, we show in Table 10 the calculations for the Canadian economy analogous to Table 7. This table assumes an increase in the price of both petroleum and coal. As with the Ontario table, we have calculated the total consumption change (which includes the change in the government sector), the import change, and the export change, and this allows us to calculate the total change in final demand shown in column 6 . Then, using the input-output table, we have calculated the total change in domestic production shown in column 7. Using employment per dollar of output we have calculated the employment change by sector and have shown this in column 8. In the last row of the table we have shown the total employment change, the total trade effect, and the total consumption change . 6 See the Ontario Statistical Review: I 9 73, 34 . In this revision only twenty-three sectors are included.

TABLEI0 Production and employment effects for Canada, 1971, imports zero

Exports Ch.

5 Trade Bal. Ch.

6 Fin. Dem. Tot. Ch.

7 Dom. Prod. Ch.

8 Emp. Effects

0.0 0.0 0.0 0.0 0.0 0.0 0.0

93.2 3.4 2.4 49.6 18.1 0.3 25.7

93.2 3.4 2.4 49.6 18.1 0.3 25.7

148.2 6.0 3.8 57.4 23.2 4.9 67.2

320.9 84.8 77.8 203.5 50.3 74.8 147.8

37.8 3.7 3.5 5.2

66.3 45.3 22.0 5.5 17.6 7.2 20.7

0.0 0.0 0.0 0.0 0.0 0.0 0.0

6.1 2.4 0.7 6.0 0.2 0.1 11.7

6.1 2.4 0.7 6.0 0.2 0.1 11.7

72.4 47.7 22.7 11.4 17.9 7.3 32.4

89 .2 71.9 46.3 52.0 67.3 98.3 62.5

1.4 1.5 1.5 0.8 3. 7 2.6 1.7

8 9 10 11 12 13

4. 1 12.0 10.7 7.2 8.8 3.5 5.3

0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.0 4.3 1.9 0.5 0.8 1.4 0.5

0.0 4.3 1.9 0.5 0.8 1.4 0.5

4.1 16.3 12.6 7.7 9.6 4.8 5.8

10.1 20.5 34.9 32.5 18.2 466.9 26.5

0.4 0.4 0.6

15 16 17 18 19 20 21

3

4

Consum. Ch.

2 Dom. Consum . Ch.

Imports Ch.

2 3 4 5 6 7

55.1 2.6 1.5 7.8 5.1 4.6 41.5

55.l 2.6 1.5 7.8 5.1 4.6 41.5

8 9 10 11 12 13 14

66.3 45.3 22.0 5.5 17.6 7.2 20.7

15 16 17 18 19 20 21

4. I 12.0 10.7 7.2 8.8 3.5 5.3

Tot.

1.7

3.9 1.3

1.2

1.2 I 9.1 1.2

Tot. 1 2 3 4 5 6 7

14

TABLE 10 continued 3

4

Imports Ch.

Tot.

Consum. Ch.

2 Do m. Consum. Ch.

Exports Ch.

5 Trade Bal. Ch.

6 Fin . Dem. Tot. Ch .

7 Dom . Prod . Ch.

8 Emp. Effects

Tot.

22 23 24 25 26 27 28

7.1 20.6 13.9 1.8 12.0 1.7 6.5

7 .1 20.6 13.9 1.8 12.0 1.7 6.5

0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.1 0.3 0.8 2 1.2 0.1 2.4 116.0

0. 1 0.3 0.8 21.2 0.1 2.4 116.0

7. 1 20.9 14.7 22.9 12. 1 4.1 122.5

10.5 24.5 35.5 41.4 15.3 28.0 162.5

1.2 1.6 1.6 1.5 0.4 1.2 4.6

22 23 24 25 26 27 28

29 30 31 32 33 34 35

7.3 11.1 17.8 5.4 6.5 1.6 6.1

7.3 11.1 17.8 5.4 6.5 1.6 6.1

0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.5 0.7 25.7 90.7 5.9 0.5 0.4

0.5 0.7 25.7 90.7 5.9 0.5 0.4

7.7 11.8 43.5 96 .1 12.4 2.1 6.5

42.6 51.4 I 18.5 182.7 56.7 18.4 37.5

1.4 2.6 3.3 6.4 1.4 0.7 1.2

29 30 31 32 33 34 35

36 37 38 39 40 41 42

18.2 70.9 14.7 49.0 21.6 8.5 34.3

18.2 20.9 14.7 49 .0 21.6 8.5 34.3

0.0 0.0 0.0 0.0 0.0 0.0 0.0

2.6 10.6 5.6 1.7 1.5 1.1 0.9

2.6 10.6 5.6 1.7 1.5 1.1 0.9

20 .7 81.5 20.3 50.7 23 .1 9.6 35 .2

91.6 111.0 30.4 60.9 67.1 23.7 50.6

3.9 4.5 1.5 0.8 1.8 0.9 1.8

36 37 38 39 40 41 42

TABLE 10 continued

Tot .

Consum. Ch.

2 Dom. Consum. Ch.

3

4

Imports Ch.

43 44 45 46 47 48 49

15 .8 10.9 12.0 4.9 3.8 720.8 2.6

15.8 10.9 12.0 4.9 3.8 720.8 2.6

0.0 0.0 0.0 0.0 0.0 0.0 0.0

1.8 2.5 0.5 0.9 1.5 17.3 1.8

1.8 2.5 0.5 0.9 1.5 17.3 1.8

17.6 13.5 12.5 5.7 5.3 738.1 4.4

40.5 55 . 1 40.9 44.8 21.7 813.4 35.0

2.1 2.8 1.5 1.4 1.0 6.0 0.7

43 44 45 46 47 48 49

50 51 52 53 54 55 56

1.5 11.7 42.0 32.3 264.2 242.3 61.9

1.5

11.7 42.0 32.3 264.2 242.3 61.9

0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.1 0.6 33.0 2.9 0.0 8.7 43 .2

0.1 0.6 33.0 2.9 0.0 8.7 43.2

1.6 12.3 75.0 35 .1 264.2 251.0 105.1

30.3 25 .5 146.4 65.3 322.4 427.8 253.3

1.0 0.7 4.0 2.6 10.2 10.2 IS . I

50 51 52 53 54 55 56

57 58 59 60 61 62 63

4.4 15.5 74.8 2.7 35.4 31.5 0.0

4.4 15.5 74.8 2.7 35.4 31.5 0.0

0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.4 0.9 0.4 0.1 0.0 0.2 0.0

0.4 0.9 0.4 0.1 0.0 0.2 0.0

4.8 16.4 75 .2 2.8 35.4 31.7 0.0

56. 1 95 .8 186.1 35.3 71.1 80.4 23.2

12.0 9.2 6.4 12. 1 24.4 27 .6 0.0

57 58 59 60 61 62 63

Exports Ch.

5 Trade Bal. Ch.

6 Fin . Dem. Tot. Ch.

7 Dom. Prod. Ch.

8 Emp. Effects

Tot.

TABLE 10 continued

Tot.

Consum . Ch.

2 Dom. Consum. Ch.

64 65

0.0 0.0

0.0 0.0

Total Empl. Eff. 288.8

6 Fin. Dem. Tot. Ch.

7 Dom. Prod. Ch.

8 Emp . Effects

Tot.

0.0 0.0

0.0 0.0

0.0 0.0

87 .9 181.7

0.0 0.0

64 65

4

Imports Ch. 0.0 0.0

Total Trade Eff. 635.4

NOTE: see note to Table 7.

Exports Ch.

5 Trade Bal. Ch.

3

Tot. Cons. Ch. 2281.40

Analysis of energy costs 69 As in the Ontario tables, we have for convenience assumed a decrease rather than an increase in price, so that all figures in Table IO should be interpreted as having a negative sign. To show the effect of exports and imports Table 11 repeats our calculations, but this time assuming that the price changes observed in Canada are also experienced in the rest of the world. We have further assumed that consumer behaviour in the purchase of imports is exactly the same as with domestically produced goods. In other words, we are assuming that an increase in price causes a proportionate reduction in imports, just as in Ontario in Table 8, so that Tables 8 and 11 are comparable. This results in significant differences in the employment and balance-of-trade effects, but these differences are not nearly as large as in the Ontario table. While the employment change for Ontario was 331/J per cent between Tables 7 and 8, for Canada between Tables IO and 11 it is less than 2 per cent, and the trade change for Ontario is about twice as large in percentage terms as for Canada. This is what we would expect and simply reflects the fact that the Ontario economy is much more dependent on the trade sector than is the Canadian economy as a whole. As our final calculation, we have, in Table 12, repeated our calculations for Ontario, this time assuming that the petroleum price increase is the same throughout Canada and that the net price effects for Ontario are represented by the difference in the price effects for Ontario and Canada, the figure shown in column 2 of Table 4. This table implicitly assumes that only price differences between Ontario and Canada are relevant and thus we are implicitly assuming that the rest of the world has the same structure of prices and the same technology as Canada. We have assumed that imports have not changed, since this is both the intermediate case and the one which seems likely to be closest to reality for this particular calculation. The calculations have been derived in the same way as for the previous tables. If the industrial technology were the same for both Ontario and the rest of Canada, we would expect to observe no shifts in any of the variables. The increase in production and employment shown in Table 12 due to a uniform price change throughout Canada is surprisingly large; perhaps even suspiciously large. The increase of employment of 63,000 represents almost 2 per cent of the Ontario labour force in 1971. To some extent at least these rather large figures reflect the difficulties of matching the Ontario and Canadian input-output structures, and this is particularly evident for industry 39, which for Canada includes the very large price increase associated with the Petroleum products industry. But while little significance can be attached to the quantitative aspects of Table 12, it does suggest that Ontario will very probably gain from uniform petroleum price increases throughout Canada. Furthermore,

TABLE II Production and employment effects for Canada, I 971, imports reduced

Consum. Ch.

2 Dom. Consum. Ch.

I 2 3 4 5 6 7

55 . 1 2.6 1.5 7.8 5.1 4.6 41.5

36.3 1.9 0.4 0.0 0.9 1.3 12.3

7.8 4.2 3.3 29.3

93.2 3.4 2.4 49 .6 18. I 0.3 25 . 7

8 9 10 11 12 14

66.3 45 .3 22.0 5.5 17.6 7.2 20.7

62.2 44 .6 16.5 5.0 17.1 6. 1 14.3

4.2 0.8 5.5 0.4 0.5 1.0 6.4

6.1 2.4 0.7 6.0 0.2 0. 1 11.7

1.9 1.6 -4.7 5.5 -0.3 -0.9 5.4

68 .2 46.9 17.3 11.0 17.4 6.2 26.0

81.0 66.2 35.8 46.8 53.2 77.8 51.7

1.3 1.4 I.I 0.7 3.0 2. I 1.4

15 16 17 18 19 20 21

4. 1 12.0 10.7 7.2 8.8 3.5 5.3

3.9

0. 1 1.9 0.4 2.6 1.4 2.8 4.1

0.0 4.3 1.9 0.5 0.8 1.4 0.5

- 0.1 2.4 1.5 -2.0 -0.6 - 1.5 -3.6

4.0 14.4 12.2 5.1 8.2 2.0

8.8 17.9 29.9 24 .9 14.9 370.8 18.0

0.3 0.3 0.5 0.9 1.0 15 .1 0.8

Tot.

13

IO. I

10.3 4.6 7.4 0.6 1.2

3

4

Imports Ch . 18.8 0.7 1.1

Exports Ch.

5 Trade Bal. Ch.

6 Fin. Dem . Tot. Ch .

7 Dom. Prod. Ch.

8 Emp. Effects

74.4 2.6 1.3 41.9 13.9 - 3.0 - 3.5

129.4 5.2 2.7 49.6 19.0 1.6 38.0

286.2 75 .5 60.1 168.8 40.7 56.9 104.4

33.8 3.3 2.7 4.3 1.4 3.0 0.9

I 2 3 4 5 6 7 8 9

1.7

Tot.

10

11 12 13 14 15 16 17 18 19 20 21

TABLE 11 continued 3

4 Exports Ch .

5 Trade Bal. Ch.

6 Fin. Dem. Tot. Ch.

7 Dom. Prod. Ch.

8 Emp. Effects

Imports Ch.

-I.I -1.1 - 6.2 19.7 -- 1.0 1.2 111.7

6.0 19.5 7.7 21.4 11.0 2.8 118.2

8.7 22.4 25.8 38.4 I 3.7 24.5 151.5

0.5 1.5 1.4 0.4 I.I 4.2

22 23 24 25 26 27 28

Tot.

Consum. Ch.

2 Dom. Consum. Ch .

22 23 24 25 26 27 28

7.1 20.6 I 3.9 1.8 12.0 1.7 6.5

5.9 19.2 6.9 0.3 10.9 0.4 2. 1

1.2 4.4

0.1 0.3 0.8 21.2 0.1 2.4 116.0

29 30 31 32 33 34 35

7.3 JI.I 17.8 5.4 6.5 1.6 6.1

4.5 7.5 0.0 0.0 0. 1 0.1 2.4

2.7 3.6 17.8 5.4 6.3 1.5 3.8

0.5 0.7 25 .7 90 .7 5.9 0.5 0.4

- 2.3 -2.9 7.9 85 .3 - 0.4 -1.0 -3.4

5.0 8.2 25 .7 90 .7 6.0 0.5 2.8

35.2 42.2 82.3 160.5 42.0 14.4 27.4

1.2 2.1 2.3 5.6 1.0 0.6 0.9

29 30 31 32 33 34 35

36 37 38 39 4041 42

18.2 70 .9 14.7 49.0 21.6 8.5 34.3

6.0 36.0 8.9 40.5 3.9 6.9 25 .2

12.1 34.9 5.8 8.6 17.7 1.6 9.0

2.6 10.6 5.6 1.7 1.5 I.I 0.9

-- 9.6 --24.3 -0.2 - 6.9 -16.2 - 0.5 -8.1

8.6 46 .6 14 .5 42.2 5.4 8.0 26 .2

66.8 75.0 22.4 49.5 40.9 18.9 37.9

2.8 3.0 I.I 0.7 I.I 0.7 1.3

36 37 38 39 40 41 42

1.2 1.4 7.0 1.5 I.I

I.I

Tot.

TABLE 11 continued 2 3 Dom. Consum. Ch. Imports Ch.

4 Exports Ch.

5 Trade Bal. Ch.

6 Fin. Dem. Tot. Ch.

7 Dom. Prod. Ch.

8 Emp. Effects

1.8 2.5 0.5 0.9 1.5 17.3 1.8

- 4.0 -2.1 - 3.6 - 2.6 - 1.2 - 129.2 - 0.7

11.8 8.8 8.4 2.3 2.6 591.6 1.9

29.8 42.4 31.3 36.1 16.5 655.7 25 .2

1.5 2.1 I.I I.I 0.8 4.8 0.5

43 44 45 46 47 48 49

Tot.

Consum. Ch.

43 44 45 46 47 48 49

15 .8 10.9 12.0 4.9 3.8 720.8 2.6

574 .2 0.0

3.5 2.7 146.6 2.5

50 51 52 53 54 55 56

1.5 11.7 42.0 32.3 264.2 242.3 61.9

1.0 9.9 9.7 19.2 264.2 242.3 57 .8

0.5 1.8 32.3 13.0 0.0 0.0 4.0

0.1 0.6 33.0 2.9 0.0 8.7 43.2

-0.3 -1.2 0.7 --10.2 0.0 8.7 39.2

I.I 10.5 42.7 22.1 264.2 251.0 101.1

24. 1 21.4 100.0 47.6 314.9 397.4 225 .7

0.8 0.6 2. 7 1.9 10.0 9.5 I 3.4

50 51 52 53 54 55 56

57 58 59 60 61

4.4 15.5 74.8 2.7 35.4 31.5 0.0

4.3 15.0 73.6 2.0 35.4 29.0 0.0

0.2 0.5 1.2 0.7 0.0 2.5 0.0

0.4 0.9 0.4 0.1 0.0 0.2 0.0

0.2 0.4 -0.8 -0 .6 0.0 - 2.3 0.0

4.6 15 .9 74.0 2. 1 35.4 29 .2 0.0

46.8 80.2 166.9 29.9 64.9 69.3 19.6

10.0 7.7 5.8 10.2 22.2 23.8 0.0

57 58 59 60 61 62 63

62 63

10.0 6.3 7.9 1.4 1.1

5.8 4.6 4. 1

Tot.

TABLE 11 continued

Tot.

Consum. Ch.

2 Dom. Consum. Ch.

64 65

0.0 0.0

0.0 0.0

Total Empl. Eff. 244.5

6 Fin. Dem. Tot. Ch.

7 Dom. Prod . Ch.

8 Emp. Effects

Tot.

0.0 0.0

0.0 0.0

0.0 0.0

75 .2 149.0

0.0 0.0

64 65

4

Imports Ch. 0.0 0.0

Total Trade Eff. 163.1

NOTE: See note to Table 7.

Exports Ch.

5 Trade Bal. Ch.

3

Tot. Cons. Ch. 2281 .40

TABLE12 Production and employment effects for Ontario, 1971 , uniform Canadian price increase and no import change 3

4

Consum. Ch.

2 Dom. Consum. Ch.

Imports Ch.

2 3 4 5 6 7

- 25 .0 15.5 --49.5 -- 15.6 -6.3 - 2.7 - 0.2

- 25 .00 15 .5 - 49.9 - 15.6 -6.3 -2.7 - 0.2

8 9 10 II 12 13 14

- 17.4 -0.0 1.8 -2.5 --1.4 - 3.1 -- 0.5

-- 17.4 0.0

15 16 17 18 19 20 21

- 3.9 - 0.6 -4.2 0.0 -0.3 2.7 0.3

Tot.

Exports Ch.

5 Trade Bal. Ch.

6 Fin. Dem. Tot. Ch.

7 Dom . Prod. Ch.

8 Emp. Effects

0.0 0.0 0.0 0.0 0.0 0.0 0.0

-12.9 23.5 - 24.5 - 10.8 -1.2 - l.4 -0.1

-12.9 23.5 -24.5 -10.8 - 1.2 --1.4 -0.l

-38.0 39.l -74.0 - 26.3 -7.5 - 4.l -0 .2

- 123.6 318.6 - 96.7 -32.0 -22.l -3.2 - 2.0

- 3.0 7.8 -1.5 -0.6 - 0.5 - 0.2 -0.0

l 2 3 4 5 6 7

--10.5 0.0

-2.5 -1.4 -3.1 -- 0.5

0.0 0.0 0.0 0.0 0.0 0.0 0.0

- 5.8 -3.2 -2.0 - 0.3

-10.5 0.0 1.1 - 5.8 - 3.2 - 2.0 - 0.3

- 27 .9 0.0 2.8 -8.3 - 4.6 -5.1 - 0.9

-35.8 0.5 4.5 - 9.7 5.4 -5.l - 1.4

-0.8 0.0 0.1 -0.l 0.1 - 0.3 0.0

8 9 10 II 12 l3 14

- 3.9 -- 0.6 -4.2 0.0 - 0.3 2.7 0.3

0.0 0.0 0.0 0.0 0.0 0.0 0.0

- 2.4 - 0.2 -1.l 0.0 - 0.2 1.0 0.3

-2.4 - 0.2 -1.l 0.0 -0.2 1.0 0.3

-6.3 -0.8 - 5.2 0.0 -0.4 3.7 0.6

- 4.4 -1.0 -5.2 2.9 4.3 4.0 4.7

- 0.l - 0.1 -0.3 0.1 0.2 0.2 0.2

15 16 17 18 19 20 21

1.8

1.1

Tot.

TABLE 12 continued 2 Dom. Consum. Ch.

3

4

Imports Ch.

Exports Ch.

5 Trade Bal. Ch.

6 Fin. Dem. Tot. Ch.

7 Dom. Prod. Ch.

8 Emp. Effects

Tot.

10.9 7.1 3.8 124.1 15.0 10.8 30.0

10.9 7.1 3.8 124. l 15.0 10.8 30.0

15.6 9.8 5.5 344.0 22.0 20.9 41.7

31.8 24.4 31.6 563.9 147.1 23.5 48.4

1.0 0.8 1.3 15.0 5.3 0.8 1.3

22 23 24 25 26 27 28

Tot.

Consum. Ch.

22 23 24 25 26 27 28

4.8 2.7 1.7 219.9 7.1 IO. I 11.7

219.9 7.1 IO. I 11. 7

0.0 0.0 0.0 0.0 0.0 0.0 0.0

29 30 31 32 33 34 35

18.9 8.4 74.3 4.7 30.0 34.4 10.7

18.9 8.4 74.3 4.7 30.0 34.4 10.7

0.0 0.0 0.0 0.0 0.0 0.0 0.0

25.7 20.1 90.8 67.5 10.5 16.0 4.5

25.7 20.1 90.8 67.5 10.5 16.0 4.5

44.5 28.5 165.1 72.2 40.5 50.3 15.2

110.1 30.0 245.4 89.1 43.4 55.4 26.1

5.4 0.9 3.4 2.8 1.7 2.2 1.0

29 30 31 32 33 34 35

36 37 38 39 40

22.2 1.2 - 2.6 199.1 --6.1 - 0.l -- 0.1

22.2 1.2 -2.6 199.1 --6.1 - 0.1 -0.1

0.0 0.0 0.0 0.0 0.0 0.0 0.0

10.3 0.5 - 3.0 94.9 -2.4 -0.1 -0.1

10.3 0.5 -3.0 94.9 -2.4 - 0.1 - 0.1

32.5 1.6 -5.6 294.0 - 8.5 -0.1 - 0.2

43.2 17. 7 - 0.7 327.9 -1.8 9.2 0.5

1.7 0.5 0.0 4.4 0.0 0.3 0.0

36 37 38 39 40 41 42

41 42

4.8 2.7 1.7

TABLE 12 continued

Tot.

Consum. Ch.

2 Dom. Consum. Ch.

43 44 45 46 47 48 49

2. 1 0.3 3.9 - 53.2 8.2 7.4 0.7

2.1 0.3 3.9 -53.2 8.2 7.4 0.7

Total Empl. Eff. 63.1

3

4

Imports Ch. 0.0 0.0 0.0 0.0 0.0 0.0 0.0

Total Trade Eff. 491.l

Exports Ch.

5 Trade Bal. Ch.

6 Fin. Dem. Tot. Ch.

7 Dom. Prod. Ch.

8 Emp. Effects

Tot.

2.8 0.4 0.0 0.0

2.8 0.4 0.0 0.0

0.7 0.1

0.7 0.1

5.0 0.6 3.9 -53.2 9.2 0.8 0.8

49.7 11.5 32.4 84.7 55 .3 117.0 146.2

l.l 0.6 0.8 2.1 1.4 2.9 3.6

43 44 45 46 47 48 49

LO

LO

Tot. Cons. Ch. 509 .53

NOTE: All figures in millions of dollars, except employment, which is in thousands of persons. All figures have the correct sign.

Analysis of energy costs 77 the relative magnitudes of the employment changes in column 8 may have significance even if the absolute levels do not. Partiuclarly noteworthy is the very large increase in industry 25 , Iron and steel mills. The industry employment figures also suggest the agriculture and textile industries will suffer relative to all other sectors. With employment and output in Ontario increased , a uniform increase in the price of petroleum would be expected to be beneficial to Ontario relative to the rest of Canada. From Ontario's point of view, policies which stabilize prices across Canada at higher levels thus represents not a neutral policy but an expansionary one. Of course, it must be recalled that this takes account only of Canadian price increases and no attempt has been made to investigate the effects vis-a-vis the rest of the world. It does not seem likely, however, that the over-all effects of petroleum price increases will be larger in Canada than in Canada's major trading partners, for even though the total effects of a given price change may be larger in Canada than, for example, in the United States, it seems likely that the over-all direct effect will be enough larger in the United States to eliminate this difference. We are thus supposing that Canadian petroleum prices will be kept lower than in the rest of the world. Moreover, our discussion takes no account of what is done with the tax revenue , which we have assumed has been collected and which has caused this increase in petroleum prices. Which level of government gets this revenue and how the revenue is spent are clearly going to make a significant difference with regard to the total effect on output and employment.

4

Provincial policy options

INTRODUCTION

We have examined the effects on Ontario of petroleum price increases first for the case where the increase was assumed to affect only Ontario, and second for the case where the price increases were uniform throughout Canada. In this chapter we examine the possible policies which the province of Ontario could formulate in order to minimize the undesirable consequences of such price increases. We have already seen that for the case where prices were assumed to rise uniformly throughout Canada, Ontario was seen to benefit, at least with regard to the rest of Canada. Clearly no further action on the part of Ontario is either necessary or desirable in this case, except in so far as trade patterns with the rest of the world are affected. Indeed, it could be argued that one major policy conclusion could be that if petroleum price rises are inevitable, then Ontario should make every effort to ensure that these price rises are uniform throughout the country. Since uniform price rises present no problem for Ontario our major concern here will be with the possibility of price rises which occur in Ontario but not in other provinces. In practice, price rises would likely occur, not just in Ontario, but in all provinces other than the major oil producers. An analysis of this situation would require an input-output table for the oil-producing province and an input-output table for the rest of Canada; unfortunately, no such data are available. Our analysis must therefore proceed under the unrealistic assumption that price rises take place only in Ontario. It should be noted that this assumption will introduce a significant bias into our calculations. For example, a

Provincial policy options 79 significant proportion of the trade between Ontario and the rest of Canada is with Quebec and, to a lesser extent, the Atlantic provinces, and there is no reason to believe that this trade would be affected by a two-price system for Canadian oil. The price of petroleum in Quebec and the Atlantic provinces will be at least as high as it is in Ontario. Indeed, although no figures are available, one suspects that a relatively small proportion of Ontario's trade is done with Alberta and Saskatchewan. Unfortunately, no alternatives to our procedure seem to be available at the present time. Our discussion considers two kinds of policy actions: first, those which attempt to remove the competitive disadvantage associated with the higher petroleum prices that Ontario has been assumed to pay; second, more aggressive policies which would try, through changing the terms of trade, to recapture some of the income loss associated with the higher petroleum prices being imposed. In short, we investigate the possibilities of both non-retaliatory and retaliatory action. Our aim is not to suggest that retaliatory policies should be undertaken; quite the opposite, by drawing attention to the possibilities of retaliation we hope to make clear the disadvantages of such actions and the counteractions they cause. It is a well-known proposition in international trade that tariffs in one country, followed by retaliatory tariffs in another country, followed by further tariffs in the first, and so on, will inevitably result in a reduction in the over-all level of welfare. Admittedly it is possible, al though not likely, that one of the countries could, in the end, be better off. These arguments are equally true for trade restrictions between provinces, for it is clear that the welfare of the entire economy would be enhanced by free interprovincial trade. This is not to say, of course, that free interprovincial trade will necessarily maximize the welfare of each region or each province. But in so far as certain provinces or regions are disadvantaged by the over-all economic structure of the economy, these problems can best be dealt with by the federal government; attempts by provincial governments to capture a larger share of the Canadian GNP through the imposition of trade restrictions are likely to result in a lower level of welfare for everyone. NON-RETALIATORY POLICIES

In the previous chapter we calculated the total price effect of a I 00 per cent increase in the price of petroleum and coal and reported the results in Table 1. If the Ontario government were to re-establish the prices which existed before the increase in the price of hydrocarbon fuels, some policy would have to be found to counteract these price increases. Theoretically this can be done at two levels. First, a policy could be formulated which would attempt to remove the direct

80 The effects of energy price changes effect of the petroleum price increases, presumably by operating at the level of petroleum as an input. Second, an attempt could be made to lower the over-all price of the product, that is, the province could attempt to re-establish the original commodity prices directly. The first of these could take the form of a policy to subsidize the use of petroleum products; the second would be some policy to directly subsidize output of the final products. Both of these methods have advantages and disadvantages. First, subsidization of the use of petroleum products would be administratively simple. It would require only a subsidy equal to the amount of the increase in the price of petroleum and coal. The disadvantage of such a scheme is that the subsidy would be passed on, not only to Ontario producers, but also to all other producers who buy intermediate goods from Ontario. Thus the Ontario government would be subsidizing all producers in Canada who buy inputs from Ontario firms. Subsidizing output of the firms directly has the advantage of applying only to Ontario producers, but the disadvantage of being extremely difficult to administer, particularly if it was thought necessary to remove the entire price change for each sector. Although we have presented a vector of total price effects these must be regarded as only approximations of the price changes which would actually occur, and it seems unlikely that any very good estimates of the total effect would be possible. We have not, for example, taken account of any possible substitution among various kinds of fuels, and although we are using disaggregated data it seems very likely that, from a commodity point of view, in any one sector there could be large variance with respect to the actual price increases. We have, for example, calculated a 7. I per cent increase in the price of the output of the agriculture, forestry, and fishing industries; but this sector comprises a large range of commodities, and it is unlikely that this figure will be a good approximation for any large number of them. There will also be differences in the costs of these two programs. The cost for the subsidization of coal and petroleum products is easily found by observing from the Ontario input-output table the total value of the use of these products as an intermediate good, for our assumption has been that there will be a l 00 per cent rise in price . The total increase in the cost of petroleum and coal is thus found to be $401 million for 1965. Assuming that petroleum and coal purchases in I 971 are in the same ratio to gross provincial expenditure as in I 965, expenditure would be $700 million in I 97 l. The cost of subsidizing the final output can be found by multiplying the price rise for each industry, as found in column 2 of Table 1, by the total sales of these products to consumers and then summing; this is found to be $1,029 million. It is clear from these two figures that although the subsidization of the use of petroleum and coal products also aids firms not in Ontario, the total cost of this method is still substantially less

Provincial policy options 81 than the subsidization of the final output. Both methods, however, are relatively expensive. For 1965 the subsidization of coal and petroleum would have required an increase of 11 per cent in the government budget, while the subsidization of the final products would have required an increase of 16 per cent in government expenditure . And in considering such a policy it must be remembered that taxes would have to increase by this amount if deficits were to be avoided. Such tax increases could not help having a substantial contractionary effect on the Ontario economy, and it is therefore not clear that much would be gained from such a policy. Another method of improving the competitive position of Ontario industry which is aimed at the cost side but not at the cost of petroleum would be the removal of any existing Ontario taxes which affect production. The main candidate here would be the Ontario corporate income tax; we shall discuss this tax in detail below. RETALIATORY POLICIES

As we saw in chapter 2, to increase the welfare of the province or country imposing it, a tariff must raise the prices of the export commodities. Thus, in theory, one province can impose a tariff on the rest of the country through any method which raises the price of the commodities exported by that province. For example, if the provinces were allowed to impose origin-type taxation, then in order to improve the terms of trade Ontario could impose a tax similar to the Canadian manufacturers' sales tax, either uniformly for all products or selectively for those products which are the major exports to the rest of Canada. That the income loss due to the higher prices of petroleum and coal could easily be recovered by a modest tax on some of Ontario's major exports can easily be seen from Table 2, where we have shown net exports for Ontario for major commodity groups. As we can see, imports of petroleum and coal are small, relative, for example, to the exports of the major manufacturing industries. Specifically, imports of petroleum products are only about 1 per cent of total exports of the manufactured goods from industries 27 through 36. The income loss to Ontario of a 100 per cent tax. on petroleum imposed by petroleumproducing provinces could thus be regained by a 1 per cent manufacturers' sales tax on these sectors. Ontario would not be collecting all of the tax from the same provinces which imposed the tax on petroleum, for Ontario's exports go to other countries as well as to other provinces of Canada. To the extent that these exports go to other countries, caution would have to be exercised in increasing their prices to avoid reducing exports. The same argument obviously applies to exports to other parts of Canada, except that in this case many Ontario industries have a clear competitive advantage , so that small increases in price

82 The effects of energy price changes would have little effect on sales to other provinces. However, as previously noted, all this is hypothetical because provinces are not allowed to tax on the origin principle. Taxes can be imposed in other ways which would ultimately have the effect of raising commodity prices. A tax on any input which is widely used in the manufacturing sector would have this result, assuming that the tax is passed along to producers. Inputs used in all industries and therefore best suited for such taxation are the primary factors capital and labour. And of these two the factor which is usually used as a source of tax revenue is capital. We shall turn now to a brief analysis of corporate income tax and its effect on commodity prices. There has been a good deal of discussion of the corporate income tax and its effect on output prices and returns to the factors of production. Originally the corporate income tax was meant to be a tax on corporate profits, profits, that is, in the economic sense of a return above the normal rate of return. A pure profits tax has no effect on price, output, or the return to the factors of production, but simply lowers excess profits. However, the corporate income tax does not act in this way. First of all, it is calculated on the return to all capital, or at least on the return to all equity capital, and not just on some concept of excess profits. This being the case, it must affect production conditions, either by increasing output prices or by reducing the return to the factors, just as would a tax on any factor of production . There has been a good deal of discussion on who bears the burden of the corporate income tax . In a path-breaking work Harberger (I 962), using the traditional neoclassical production model with two sectors and two factors of production which is so widely used in international trade, analysed the corporate income tax and concluded that by and large the tax was borne by the owners of capital. Another important piece of research in this area was conducted by Krzyzaniak and Musgrave (1963), who argued that the tax was not only fully reflected in commodity prices but in fact shifted forward by significantly more than I 00 per cent. They argued, in effect, that any tax on capital actually increased the return to capital and this increase was passed on to consumers in the form of higher prices. It should be noted that these two studies used quite different methodologies. The Harberger study was primarily a theoretical analysis based on the assumption that the corporate income tax applied only to one of the two sectors of the economy. The tax therefore represented a distortion on the production side of the economic system . The Krzyzaniak and Musgrave study , on the other hand, was an empirical analysis using time-series data for the period 1930-59. In Canada a study similar to that of Krzyzaniak and Musgrave was carried out by Spencer (1969), and he also concluded that the tax was shifted forward by more than I 00 per cent.

Provincial policy options 83 All of these studies, particularly the empirical work of Krzyzaniak and Musgrave and Spencer, have been subjected to a considerable amount of criticism. Many have argued that their sample time period was inappropriate since it included the years of the depression as well as the war years, and their choice of the non-tax variables has been criticized. For example, when Cragg, Harberger, and Mieszkowski (l 967) added a cyclical variable and dummy variables for the war years, they found that the estimate of shifting dropped to 60 per cent and the tax variable was no longer significant. It is not our intention, however, to carry on a lengthy discussion of the controversy surrounding the shifting of the corporate income tax (this issue is fully discussed in Melvin, 1975a, 1975b). We are only interested in the question whether or not it can be assumed that the tax is in fact reflected in commodity prices. Strangely enough, on this question both the Harberger and the Krzyzaniak and Musgrave results agree. Harberger, for example, shows that in the long run the corporate income tax must be reflected in commodity prices, and that prices will rise in proportion to the amount which capital contributes to the per unit cost of production. The argument on the shifting of the corporation income tax, then, is basically about whether or not capital is disadvantaged by the tax, not about whether the tax is passed on to consumers. While the first of these arguments is no doubt important, it is not of particular interest for our study. We thus proceed on the assumption that the corporate income tax is entirely reflected in commodity prices. The model we use is exactly analogous to the one we described in chapter 3. 1 Again we assume constant returns to scale and a fixed-coefficient production model. But now the tax is applied to the value added by capital rather than to the direct input of petroleum so that instead of equation (17) we have

p

= A

p + vt,

(21)

where v is a vector of value added and t is the corporate tax rate assumed to be constant for all firms. The solution to this system is completely analogous to (18): p =

(J-Ar 1 vt.

(22)

There are, however, difficulties associated with the estimation of the vector p. First, the value added vector in equation (22) is the value added for all capital, For a full description of this model and for the price and employment effects for the United States see Melvin (l 975c).

84 The effects of energy price changes TABLE 13 Commodity price effects of the corporation income tax, Ontario

Industries I 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

Agriculture, forestry, and fishing Mining Meat and poultry Dairy products Grain mills Biscuits and bakeries Sugar and confectioneries Other food Soft drinks Distilleries, breweries, and wineries Tobacco and tobacco products Rubber products Leather and leather products Cotton yarn and cloth Synthetic textiles Knitting mills Clothing Other textile mills Sawmills Furniture and fixtures Other wood Pulp and paper mills Pa per products Printing and publishing Iron and steel mills Other primary metals Fabrica ted and structural metals Metal stamping, pressing, and coating Other metal fabricating Miscellaneous machinery Motor vehicles and aircraft Other transportation equipment Electrical appliances Electrical industrial equipment Communication equipment Other electrical products Clay, lime, and cement Other non-metallic mineral products Petroleum refineries and coal products Plastics and synthetic resins

2

3

Direct effect

Total effect

Ratio

4 Total effect , petroleum

0.21 2.06 0.90 1.58 1.35 1.36 4.48 4.48 2.96 9.11 4.68 4.63 I.SO 2.09 2.49 1.91 1.32 3.24 1.20 1.26 1.20 1.02 2.42 3.47 1.45 1.45 1.02 3.97 2.27 5.12 3.35 2.62 1.50 2.75 2.89 2.57 3.14 3.12 5.30 3.5 I

2.64 4.42 3.90 4.71 4 .90 5.12 8.76 8.78 6.51 11.76 7.85 9 .18 4 .83 6.03 6.61 6.01 4 .63 7.13 3.31 4 .18 3.61 3.35 6.01 5.86 3.81 4.59 3.13 6.74 5.21 7.49 7.34 5.86 4 .69 5.51 6.09 6.19 6.02 6.64 9.55 10.61

12.550 2,146 4.336 2.984 3.622 3.768 1.953 1.960 2.201 1.290 1.679 1.982 3.220 2.886 2.660 3.142 3.515 2.202 2.755 3.323 3.003 3.286 2.488 1.687 2.636 3.174 3.065 1.699 2.294 1.464 2.187 2.238 3.128 2.006 2.102 2.408 1.915 2.126 1.803 3.019

7.10 2.47 5.48 5.19 4.22 2.70 1.75 3.49 2.11 1.32 4.00 2.55 2.40 1.73 3.50 1.78 1.40 1.87 2.68 I.SO 1.92 1.91 2.70 .90 1.14 3.44 .69 1.21 1.51 1.55 1.34 1.42 1.85 1.42 1.96 2.10 2.68 3.09 10.87 6.35

Provincial policy options 85 TABLE 13 continued

Industries 41 42 43 44 45 46 47 48 49

Paint and varnish Pharmaceuticals and medicines Other chemical Miscellaneous mfg Construction, maintenance , and repair Transport, storage, and trade Utilities Communications and other services Unallocated sector

Arithmetic average Weighted average

2

3

Direct effect

Total effect

Ratio

4 Total effect, petroleum

1.39 7.50 3.52 3.44 .81 1.17 4.80 3.02 1.78

6.17 11.50 7.81 7.63 3.34 2.74 6 .71 4.28 6 .63

4.441 1.534 2.217 2.220 4.146 2.349 1.399 1.417 3.719

3.87 2.03 7.95 2.07 2.55 3.00 I.LO .86 2.11

2.76 1.78

6.04 4.10

2,76 2.95

whereas the corporate income tax is applied only to equity capital. Second, even if we could sort out equity capital from debt capital, there would still be a problem in identifying this value added figure from the input-output table. The value-added-for-capital figure in input-output analysis is traditionally just a residual containing many things other than the real return to capital. It contains, for example, elements of rent which may exist in various industries. To solve both of these problems we have decided to estimate the vector directly rather than obtaining it from the input-output table. Under our assumption that the corporate income tax is passed along entirely in the form of higher commodity prices, in order to find out what the price effect of the corporate income tax is we need only find how much tax each sector pays and then calculate the ratio of total taxes collected to total sales. If the tax is entirely passed along in the form of higher commodity prices, then the amount that prices must rise must be exactly reflected in the amount of corporate taxes actually collected. The commodity price effects of the corporate income tax are presented in Table 13. The first column represents the direct effect of the corporate income tax, this being the ratio of total taxes to total sales or, in terms of equation (22), vt. Column 2 is the total effect of the corporate income tax and represents the calculations of p through the use of the input-output table as indicated in equation (22). Column 3 is the ratio of the total effect to the direct effect. For comparison , we have presented in column 4 the total effects of the petroleum price increase calculated in Table 1. The last two rows of the table give the

86 The effects of energy price changes arithmetic averages of the direct and total effect of the corporate income tax as well as these averages weighted by sales. There are two important limitations of the data presented in Table I 3. First, the direct effect of the corporate income tax, that is, the ratio of taxes collected to sales, were calculated using Canadian rather than Ontario data. We were not able to obtain corporate taxes collected for Ontario for I 971 and are therefore assuming that the Canadian counterparts will be a reasonable approximation of the Ontario figures. It is clear that for many industries this approximation will not be far wrong, for in the manufacturing sector, for example, much of the activity is carried on in Ontario, and thus Ontario has a heavy weight in the Canadian figures. Second, the tax figures used are total federal plus provincial taxes collected, not simply provincial taxes. The Ontario provincial corporate income tax is IO per cent, and thus, in I 971, represented approximately one-fifth of the total corporate tax collected. Therefore, even if the provincial corporate tax were removed and this tax saving passed along to producers, prices would fall only by 20 per cent of the figures shown in column 2. In fact, however, the provincial corporate tax is a deduction against the federal corporate income tax up to a maximum of 10 per cent, so that the removal of the provincial corporate income tax would simply mean that the federal government would be collecting this tax revenue . Thus, without a change in legislation, Ontario cannot lower the amount by which the corporate income tax increases commodity prices. It is possible for Ontario to increase the corporate income tax and thus the price of the products exported to other Canadian provinces, hoping thereby to improve the terms of trade. The degree to which this can be done is limited by the fact that corporate income tax cannot exceed 100 per cent, and from a practical point of view, no province would consider such an extreme change. A l 0 per cent increase in the tax, an increase in total corporate taxes from, say, 50 per cent to 60 per cent, would be about the maximum we could expect. As we can see from column 2 of Table 13, this would have a relatively moderate effect on increasing prices, even relative to the total effects of the petroleum price increase. There are, for example, only three sectors where this would result in a price increase of more than I per cent. We must conclude, then, that while some retaliation through increases in the provincial corporate income tax is possible, any increase large enough to bring about a significant price change is likely to have serious consequences with regard to the location of industry. It seems almost certain that major changes in the Ontario corporate income tax relative to other provinces would prompt many firms to leave Ontario for other parts of Canada. In Tables 14, I 5, and I 6 we present estimates of the production and employment changes which would occur if the corporate income tax were

TABLE14 Effect of the removal of the corporate income tax: no import change

Tot.

Consum. Ch.

Exports Ch.

5 Trade Bal. Ch.

6 Fin. Dem. Tot. Ch.

7 Dom. Prod. Ch.

8 Emp. Effects

2 Dom. Consum. Ch.

3

4

Imports Ch.

Tot.

2 3 4 5 6 7

27.1 39 .9 43.8 22.8 13.3 16.9 22.7

27 . 1 39 .9 43.8 22.8 13.3 16.9 22.7

0.0 0.0 0.0 0.0 0.0 0.0 0.0

14.0 60.6 21.7 15.8 2.6 8.5 9.5

14.0 60.6 21.7 15 .8 2.6 8.5 9.5

41.1 100.5 65.4 38 .6 16.0 25.4 32.1

259.5 329.7 132.9 65 .9 64.6 30.6 56.0

6.4 8.1 2.1 1.2 1.4 1.5 1.2

I 2 3 4 5 6 7

8 9 10 11 12 13 14

99.7 10.8 45 .7 6.9 20.I 9.7 10.4

99.7 10.8 45 .7 6.9 20.1 9.7 10.4

0.0 0.0 0.0 0.0 0.0 0.0 0.0

60 .3 4.3 27.7 16.2 45.3 6. 1 6.5

60 .3 4.3 27.7 16.2 45.1 6.1 6.5

160.0 15.1 73.4 23.1 65 .3 15.8 16.9

246.7 I 8.6 83.2 27.0 124.1 25.6 65.4

5.5 0.4 1.8 0.3 2.2 1.6 1.2

8 9 10 II 12 13 14

15 16 17 18 19 20 21

28.1 I 1.6 68.5 17.8 2.6 16.5 4.0

28 . 1 11.6 68.5 17.8 2.6 16 .5 4.0

0.0 0.0 0.0 0.0 0.0 0.0 0.0

17.5 4.2 18.1 6.6 1.5 6.2 3.3

17.5 4.2 18.1 6.6 1.5 6.2 3.3

45.6 15.8 86.6 24.4 4.1 22.6 7.3

111. 9 20.0 92.5 88.3 27.6 24.7 28.4

2.2 1.2 5.5 1.7 1.2 1.0 1.2

15 16 17 18 19 20 21

TABLE 14 continued

Tot.

Consum. Ch.

2 Dom. Consum. Ch.

3

4

Imports Ch.

Exports Ch.

5 Trade Bal. Ch.

6 Fin . Dem. Tot. Ch .

7 Dom. Prod. Ch.

8 Emp. Effects.

Tot.

22 23 24 25 26 27 28

5.8 9.3 13.0 93.5 16.4 5. 1 13.0

5.8 9.3 13.0 93.5 16.4 5. 1 13.0

0.0 0.0 0.0 0.0 0.0 0.0 0.0

13.2 23 .9 28.8 52.8 34.6 5.5 33.2

13.2 23.9 28.8 52.8 34.6 5.5 33.2

19.0 33.2 41.8 146.3 51.0 10.6 46.2

98.2 134.9 158.0 474.3 351.3 19.1 90. 1

3.0 4.3 6.7 12.6 12.7 0.6 2.4

22 23 24 25 26 27 28

29 30 31 32 33 34 35

28.7 22.7 228.8 8.7 25 .2 50.3 63 .3

28.7 22.7 228.8 8.7 25 .2 50.3 63.3

0.0 0.0 0.0 0.0 0.0 0.0 0.0

39.1 54. 1 279 .5 126.1 8.8 23.4 26.9

39.1 54.1 279.5 126.1 8.8 23.4 26 .9

67 .8 76.7 508 .2 134.8 33.9 73 .8 90.2

244.3 89.7 758.5 169.1 42.4 83.9 130. 1

12.0 2.7 10.4 5.3

29 30 31 32 33 34 35

36 37 38 39 40 41 42

86.0 8.6 17.3 76.5 15.4 5.5 29.7

86.0 8.6 17.3 76.5 15.4 5.5 29.7

0.0 0.0 0.0 0.0 0.0 0.0 0.0

40.0 3.3 20.4 36.5 5.9 6.7 18.5

40.0 3.3 20.4 36.5 5.9 6.7 18.5

126.0 11.8 37.8 113.0 21.3 12.2 48 .2

154.2 42.3 88.1 206.5 88.3 40.0 57 .5

6.0 1.3 2.2 2.8

1.7

3.3 5.1

1.5

I.I 1.8

36 37 38 39 40 41 42

TABLE 14 continued

Tot.

Consum . Ch.

2 Dom. Consum. Ch .

43 44 45 46 47 48 49

51.8 43.0 12.9 317.9 52.2 337.3 74.3

51.8 43.0 !2.9 317.9 52.2 337.3 74 .3

Total Empl. Eff. 229 .5

3

4

Imports Ch . 0.0 0.0 0.0 0.0 0.0 0.0 0.0

Total Trade Eff. 1414.2

Exports Ch.

5 Trade Bal. Ch.

6 Fin. Dem. Tot. Ch.

7 Dom. Prod. Ch.

8 Emp. Effects

Tot.

68.3 62.7 0.0 0.0 6.6 3 I. I 8.0

68.3 62.7 0.0 0.0 6.6 31.1 8.0

120.0 105.7 12.9 317.9 58 .8 368.4 82.3

370.2 176.0 128.9 839.9 169.8 859.9 593.7

8.1 9.8 3.2 20.6 4.2 21.1 14.5

43 44 45 46 47 48 49

Tot. Cons. Ch. 2250.90

TABLE15 Effects of the removal of the corporate income tax: imports reduced 7 Dom. Prod. Ch.

8 Emp. Effects

22.9 62 .7 49.5 27.0 14.0 I 9. I 25.2

I 83.1 212.9 98.8 46.5 48.9 22.9 43.2

4.5 5.2 1.6 0.9 1.0 1.2 0.9

I 2 3 4 5 6 7

15.9 I.I 6.7 15.5 29.0 3.9 -1.6

115. 7 11.9 52.5 22.5 49.0 13.6 8.8

177 .5 14.2 58.7 26.3 90.8 20.9 39.9

4.0 0.3 1.3 0.3 1.6 1.3 0.8

8 9 10 II 12 13 14

-4.4 -1.1 -13.3 -1.6 -0.1 - 0.4 -0.2

23.7 10.5 55.2 I 6. I 2.5 I 6.1 3.7

68.2 I 3.2 59.0 59.5 I 9.3 17.5 I 8.7

1.3 0.8 3.5 I.I 0.8 0.7 0.8

15 16 17 18 19 20 21

3

4

Consum. Ch.

2 Dom. Consum. Ch.

Imports Ch.

Exports Ch.

I 2 3 4 5 6 7

27. I 39.9 43.8 22.8 I 3.3 16.9 22.7

8.9 2. I 27.8 11.2 11.4 10.6 15.7

18.2 37.8 15.9 11.6 1.9 6.3 7.0

14.0 60.6 21.7 15.8 2.6 8.5 9.5

- 4.2 22.8 5.7 4.2 0.7 2.2 2.5

8 9 10 11 12 13 14

99.7 10.8 45.7 6.9 20.1 9.7 10.4

55.4 7.5 24.8 6.3 3.8 7.5 2.3

44.4 3.3 20.9 0.6 16.3 2.2 8.1

60.3 4.3 27.7 16.2 45.3 6.1 6.5

15 16 17 18 19 20 21

28.1 11.6 68.5 17.8 2.6 16.5 4.0

6.2 6.3 37.2 9.5 0.9 9.9 0.4

21.9 5.3 31.3 8.2 1.6 6.6 3.5

17.5 4.2 18.1 6.6 1.5 6.2 3.3

Tot.

5 Trade Bal. Ch.

6 Fin. Dem. Tot. Ch.

Tot.

TABLE 15 continued

Tot.

Consum. Ch.

2 Dom. Consum. Ch.

3

4

Imports Ch .

Exports Ch.

5 Trade Bal. Ch.

6 Fin. Dem. Tot. Ch. 15.3 26.6 33.9 53.0 37.6 8.7 34.4

7 Dom. Prod . Ch.

8 Emp. Effects

Tot.

69 .9 95.2 105.0 273.6 225.6 14.9 65.3

2.1 3.0 4.4 7.3 8.1 0.5 1.7

23 24 25 26 27 28

22

28

5.8 9.3 13.0 93 .5 16.4 5.1 13.0

2.1 2.7 5 .1 0.2 3.0 3.2 1.2

3.6 6.6 7.9 93 .3 13.4 1.9 11.8

13.2 23.9 28.8 52 .8 34.6 5.5 33.2

9.6 17.3 20.8 - 40 .5 21.2 3.5 21.4

29 30 31 32 33 34 35

28.7 22.7 228.8 8.7 25.2 50 .3 63.3

13.7 1.8 134.9 4.6 6.6 0.7 6.4

15.0 20.8 93.9 4.2 18.6 49.6 57.0

39.1 54.l 279 .5 126.1 8.8 23.4 26.9

24.0 33.3 I 85.6 121.9 -9.8 - 26.2 -30.1

52.8 55.9 414 .3 130.7 15.4 24.2 33.3

174.8 65.2 611.4 161.7 21.7 30.2 53 .6

8.6 1.9 8.4 5.1 0.9 1.2 2.1

29 30 31 32 33 34 35

36 37 38 39 40 41 42

86.0 8.6 17.3 76.5 15.4 5.5 29 .7

1.3 6.3 2.9 41.9 2.9 1.5 18.5

84.7 2.3 14.4 34.6 12.5 4.0 II.I

40.0 3.3 20.4 36.5 5.9 6.7 18.5

- 44.7 1.0 6.0 1.9 -6.6 2.7 7.4

41.3 9.5 23.3 78.4 8.8 8.2 37. I

5 7. l 29.7 57 .2 141.1 52 .5 26.8 43.4

2.2 0.9 1.4 1.9 0.9 0.7 1.4

36 37 38 39 40 41 42

22 23 24 25 26 27

TABLE 15 continued

Tot.

Consum . Ch.

2 3 Dom. Consum. Ch . Imports Ch .

43 44 45 46 47 48 49

51.8 43 .0 12.9 317.9 52.2 337.3 74.3

I 1.0 18.4 12.9 200.l 44.4 300.8 0.2

Total Empt. Eff. 154.4

40.8 24.6 0.0 117.8 7.8 36.5 74.1

Total Trade Eff. 278.2

4 Exports Ch.

5 Trade Bal. Ch.

6 Fin. Dem. Tot. Ch.

68 .3 62 .7 0.0 0.0 6.6 31.1 8.0

27.5 38 .1 0.0 -117.8 -1.1 -5.3 -66.1

79 .3 81.1 12.9 200.1 51.0 331.9 8.2

Tot. Cons. Ch. 2250.90

7 Dom . Prod. Ch .

8 Emp . Effects

Tot.

239 .2 127.0 96.1 544.1 127.0 657.8 354.3

5.2 7.0 2.4 13.3 3.1 16.1 8.7

43 44 45 46 47 48 49

TABLE16 Effects of the removal of the corporate income tax: imports increased

Tot.

Consum. Ch.

2 3 Dom. Consum. Ch. Imports Ch.

4 Exports Ch.

5 Trade Bal. Ch.

6 Fin. Dem. Tot. Ch .

7 Dom. Prod. Ch.

8 Emp. Effects

Tot.

I 2 3 4 5 6 7

27 .1 39.9 43 .8 22.8 13.3 16.9 22.7

45 .3 77 .7 59 .7 34.5 15 .3 23.1 29.6

-18.2 -37 .8 -15 .9 -11.6 - 1.9 -6.3 -7.0

14.0 60.6 21.7 15 .8 2.6 8.5 9.5

32.2 98.4 37.6 27.4 4.6 14.8 16.4

59.3 138.3 81.3 50 .2 17.9 31.7 39.1

335 .9 446.5 167.1 85.2 80.4 38.3 68.8

8.2 10.9 2.7 1.6 1.7 1.9 1.5

I 2 3 4 5 6 7

8 9 10 11 12 13 14

99.7 10.8 45 .7 6.9 20.1 9.7 10.4

144.1 14.1 66.7 7.6 36.3 11.9 18.5

-44.4 -3 .3 -20.9 -0.6 -16.3 -2.2 -8.1

60.3 4.3 27.7 16.2 45.3 6.1 6.5

104.7 7.6 48.6 16.8 61.5 8.3 14.5

204.4 18.4 94.3 23.7 81.6 18 .0 24 .9

316.0 22.9 107.7 27.8 157.4 30.3 91.0

7.1 0.5 2.3 0.3 2.8 1.9 1.7

8 9 10

15 16 17

28 .1 11.6 68 .5 17.8 2.6 16.5 4.0

50.0 16.9 99.9 26.0 4.2 23 .1 7.5

-21.9 -5 .3 -31.3 -8 .2 -1.6 - 6.6 -3 .5

17.5 4.2 I 8.1 6.6 1.5 6.2 3.3

39.4 9.5 49.4 14.8 3.2 12.7 6.9

67 .5 21.1 117.9 32.6 5.8 29.2 10.8

155 .6 26.8 126.0 117.1 35 .9 32.0 38.1

3.0 1.6 7.5 2.3 1.5 1.3 1.6

15 16 17 18 19 20 21

18

19 20 21

11

12 13 14

TABLE 16 continued

Tot.

Consum . Ch.

2 3 Dom. Consum . Ch . Imports Ch .

4 Exports Ch.

5 Trade Bal. Ch.

6 Fin. Dem. Tot. Ch.

7 Dom. Prod. Ch .

8 Emp. Effects

Tot.

22 23 24 25 26 27 28

5.8 9.3 13 .0 93 .5 16.4 5.1 13.0

9.4 15.8 21.0 186.8 29 .8 7.0 24 .8

-3 .6 - 6.6 -7 .9 - 93.3 - I 3.4 - 1.9 - 11 .8

16.8 23.9 28 .8 52.8 34.6 5.5 33 .2

16.8 30.5 36.7 146.1 48.0 7.4 45.0

22.6 39.8 49.8 239.5 48 .0 12.5 58 .0

126.4 174.7 21 I.I 675.0 477.1 23.3 I 14.8

3.8 5.5 8.9 17.9 17.2 0.8 3.1

22 23 24 25 26 27 28

29 30 31 32 33 34 35

28.7 22.7 228.8 8.7 25 .2 50 .3 63.3

43 .8 43 .5 322.7 12.9 43 .7 99 .9 120.3

- 15 .0 -20 .8 -93.9 - 4.2 -18.6 - 49.6 - 57.0

39.1 54 .1 279.5 126.1 8.8 23 .4 26 .9

54.1 74.9 373.4 130.3 27.3 73.0 83.9

82 .9 97 .6 602 .1 139 .0 52 .5 123.4 147.2

313.7 114.2 905.5 176.4 63 .1 137.6 206.5

15.4 3.4 12.5 5.6 2.5 5.4 8.2

29 30 31 32 33 34 35

36 37 38 39 40 41 42

86.0 8.6 I 7.3 76 .5 15.4 5.5 29 .7

170.7 10.9 31.8 111.2 27 .9 9.5 40.8

-84.7 -2.3 - 14.4 -34 .6 -12.5 - 4.0 -II.I

40 .0 3.3 20.4 36.5 5.9 6.7 18.5

124.8 5.6 34.9 71.1 18.4 10.6 29 .7

210.8 14.1 52.2 147.7 33 .8 16.1 59.3

251.3 54.8 I I 9.0 271.9 124.1 5 3.1 71.5

9.7 1.6 2.9 3.6 2. 1 1.4 2.3

36 37 38 39 40 41 42

TABLE 16 continued

Tot.

Consum. Ch.

2 3 Dom . Consum. Ch. Imports Ch.

43 44 45 46 47 48 49

51.8 43.0 12.9 317.9 52.2 337.3 74.3

92.5 67.6 12.0 435.7 59.9 373.7 148.5

Total Empl. Eff. 304.6

-40 .8 -24.6 0.0 -117.8 -7.8 -36.5 -74.1

Total Trade Eff. 2550.1

Exports Ch.

5 Trade Bal. Ch.

6 f-in. Dem. Tot. Ch.

7 Dom. Prod. Ch.

8 Emp. Effects

Tot.

68.3 62.7 0.0 0.0 6.6 3 I. I 8.0

109.0 87.2 0 .0 117.8 14.4 67.6 82.1

160.8 130.3 12.9 435.7 66.6 404.8 156.5

501.2 225.0 161.6 1135. 7 212.6 1062.0 833.l

11.0 12.5 4.0 27 .8 5.2 26.0 20.4

43 44 45 46 47 48 49

4

Tot. Cons. Ch. 2250.90

96 The effects of energy price changes removed. Again, these changes are in terms of both the federal and the provincial corporate income tax and should be interpreted as the over-all effect on production and employment if the corporate income tax were to be entirely removed. We are assuming, as before, that the corporate income tax is entirely reflected in commodity prices, and that if the tax were removed prices would fall by the amount shown in column 2 of Table 13. The methodology used in generating these tables is exactly the same as that used in the previous chapter, and Tables 14, 15, and 16 are comparable, respectively, to Tables 7, 8, and 9. We have constructed a vector of changes in final demand by industry, used the input-output table to generate the total change in production, and then multiplied this production change by the employment coefficients in order to arrive at figures for the change in employment. These employment changes are about what we would expect given the size of the corporate price effect relative to the petroleum change price effect, and the employment effects which we found in chapter 3. Again, changing the provincial corporate income tax from I 0 to 20 per cent would result in an employment change of 20 per cent of those shown in the last column. If one takes 20 per cent of the calculated employment changes of the final column, the results are significant in aggregate, because the largest change, which occurs when imports increase, represents 2 per cent of the total Ontario labour force. 2 These employment reductions must be added to the reductions already calculated in chapter 3. In other words, an attempt by Ontario to improve its terms of trade relative to the rest of Canada by increasing the corporate income tax would further reduce Ontario employment. Thus, while it is possible to produce moderate terms-of-trade effects through increases in the corporate income tax, this can be done only at considerable cost in employment. We must conclude, then, that retaliatory action carried out through changes in the corporate income tax is not a very attractive alternative for the Ontario government. As before, of course, we have not taken account of the effect on the Ontario economy of the expenditure of this increase in government revenue. Indeed, if a higher corporate income tax were imposed and used to generate production in Ontario, it is possible that there need be no employment reductions at all. Even so, however, the over-all benefits of such a corporate tax increase seem dubious. The discussion here has been concerned entirely with economic options open to the Ontario government, but it is clear that other kinds of action are likely to be of more significance. Alternatives would include political pressuring, 2 The Ontario labour force in 1971 was approximately 3.25 million; see Ontario Statistical Review 1973, 31.

Provincial policy options 97 bargaining in federal-provincial conferences, court proceedings, and various kinds of inducement (economic or otherwise) to influence the location of Canadian industries. None of these issues have been dealt with here, not because they are unimportant but simply because they are outside the scope of this study.

5

Conclusions

SUMMARY OF THE RESULTS

The purpose of this study has been to investigate the effects of petroleum and natural gas price increases on commodity prices and to make some preliminary estimates of the effects of these price changes on trade and employment. We began with a discussion of interprovincial trade restrictions and argued that because taxes can be used to duplicate a tariff structure, any province, through the use of taxes or price controls, can effectively impose tariffs on the rest of Canada. A diagrammatic analysis was used to show the effects of taxes imposed by provinces, and it was argued that, just as there are optimal tariffs for countries, so there are optimal tax structures for provinces. Although, in general, the imposition of such an optimal tax structure would require the use of origin-type taxes, which in Canada are prohibited, for petroleum this difficulty can be overcome since petroleum-producing provinces have some control over the price at which petroleum will be sold. It was argued that the main reason that petroleum price increases can have such pervasive effects on commodity prices is because petroleum and natural gas are used extensively as intermediate inputs, both as fuels and as raw materials. It thus seemed useful to consider an intermediate good model, and after a brief review of the basic structure of such models the implications for tariffs and terms-of-trade changes were examined. It was argued that because of the intermediate-good nature of petroleum, the gains from trade and the potential gains from tariff imposition were likely to be greater than for goods destined for final consumption only.

Conclusions 99 In chapter 3 we calculated exactly how much a petroleum price increase would raise commodity prices. After a brief discussion of the fixed-coefficient model to be employed, the algebraic form of the model was constructed and the equation for finding the vector of price changes derived. Using the Ontario input-output table, the vectors of direct and total price changes associated with a l 00 per cent increase in the price of petroleum products, natural gas, and coal were calculated. Although the effect of interindustry flows was seen to result in significant differences between the direct and total price effects, the median increases being over six times the direct effect, the total effects were still relatively small. Other than in the petroleum sector itself no price increased by more than 8 per cent, and the median increase was 2.1 per cent. It was found that the agriculture and food-producing industries were among those with the highest total price increases, suggesting that petroleum price increases may have a larger influence on food prices than has been traditionally assumed. At the same time it was found that the price increases for the basic manufacturing industries were lower than average, being generally less than 2 per cent, suggesting that Ontario's competitive position vis-a-vis the rest of Canada would not be much affected even if such a large petroleum price increase occurred only in Ontario. We next examined the price effects of petroleum, natural gas, and coal price increases for all of Canada using the Canadian input-output table. The main reasons for examining the comparable Canadian figures were to check whether differences in industrial structure existed between Ontario and the rest of Canada, to compare the trade sectors for Ontario and Canada, and to check whether the disaggregation of the all-inclusive petroleum-products industry used in the Ontario table into three separate sectors ( coal, petroleum and natural gas, and petroleum products) much affected the calculated price changes. It was found that the aggregation used in the Ontario table did not seem to distort significantly the calculations of commodity price changes. Interesting differences in the industrial structures were found, and the earlier indication that Ontario exports would not be much affected was reinforced. When the price changes for Ontario were compared with those for Canada as a whole, an interesting pattern of price differences emerged. It was found that in Ontario prices for agricultural and food products rose more, but in the basic manufacturing sectors the price increases were significantly lower. The obvious conclusion is that in a situation where petroleum prices rise uniformly across Canada, Ontario will benefit compared with the rest of Canada. This is of particular interest in the light of recent legislation introduced by the federal government. House of Commons Bill C-32 essentially provides that the prices of

I 00 The effects of energy price changes natural gas and petroleum, exclusive of transportation and service costs, be uniform throughout Canada. We next increased the range of our comparisons by making the same set of calculations for the United States economy. It was found that the commodity price increases associated with the same increase in petroleum prices were generally lower than those found for both Ontario and Canada. Although the differences were not great, it is important to note that uniform price increases for petroleum products in North America would favour United States industry. This suggests that there may be an economic justification for maintaining a petroleum price differential between Canada and the United States. We then turned to a consideration of the extent to which these price changes would be expected to change production and employment. To do so we assumed a Cobb-Douglas utility function which implies unitary own-price elasticities of demand, and zero cross-price elasticities. Changes in quantities demanded are thus equal but opposite in sign to the percentage changes in price . The main problem in calculating the production changes was deciding on an appropriate assumption about the response of imports and exports to the price changes. In order to present a range of production changes, and as a check on the sensitivity of the results, three different assumptions about imports were used. For these three assumptions a range of employment reductions of from 2 per cent to 4 per cent of the Ontario labour force was found. The trade balance was found to vary from a reduction of $65.3 million to a reduction of $1093.1 million. The same calculations for the Canadian economy were made, and as expected the results were not as sensitive to the assumptions about imports and exports as were the Ontario results. While for Ontario changing the import assumption resulted in a 331/3 percentage difference in the employment change, the same assumed change in imports for Canada resulted in an increase of only 2 per cent in the employment change . The trade-balance change was also much less for Canada than for Ontario, reflecting the more open nature of the Ontario economy. In chapter 4 the policy options open to Ontario were considered in the event that it was thought desirable to offset the effects of an increase in the price of petroleum and natural gas. Ontario could subsidize either the use of petroleum or the commodities directly. Though subsidizing petroleum use would also aid producers outside of Ontario who buy intermediate products from Ontario, it was still found to be cheaper than subsidizing the commodities directly. Both options would be expensive, however, and would result in increases in the government budget of 11 per cent for petroleum subsidization and I 6 per cent for commodity subsidization. Policies which attempted to retaliate rather than to simply offset the commodity price effects were next considered. It was shown that if origin-type

Conclusions l Ol taxation, such as a manufacturers' sales tax, were allowed for provinces, it would be a relatively simple matter to impose a 'tariff on the rest of Canada that would offset the I 00 per cent price increase for petroleum and natural gas. The BNA Act, however, specifically prohibits this type of taxation by provinces. An allowable form of provincial taxation that may have the same kind of effect as an origin-type tax is the corporate income tax. If this tax is passed along to consumers in the form of higher prices for commodities, it has much the same effect as a tax on production. After a brief comment on the history of the controversy surrounding the question of the shifting of the corporate income tax it was concluded that this study could assume that the tax is entirely reflected in commodity prices. A model was then developed to estimate the commodity price effects of the corporate income tax, and these were compared to the commodity price effects of the petroleum price increase. It was argued that although Ontario could improve its terms of trade with the rest of Canada by increasing the provincial corporate income tax, the scope for this sort of action is limited by the fact that the provinces control only a relatively small percentage of the tax collected. It was concluded that increases in the corporate income tax were not an attractive policy alternative . Indeed, it seems that by far the best policy for Ontario to pursue is to attempt to make sure that any petroleum and natural gas price increases are applied equally to all industries in Canada, and that a domestic two-price system not be instituted. To conclude, and in order to caution the reader against attaching undue confidence to these conclusions, the major assumptions made in the study will be commented on briefly . First of all, the entire study is based on input-output models, and thus we have implicitly assumed both that such a model is an appropriate description of the real world and that the coefficients of the input-output table used are sufficiently accurate for the purposes of the calculations performed. Regarding the first of these, there is no clear evidence whether fixed coefficient models are a sufficiently close approximation to reality to justify their use. In the short run they probably are but in the long run the question is very much an open one . To the degree that substitution is possible, our estimates should be thought of as upper limits of the 'real' values. If one accepts the input-output methodology there would seem to be no particular reason to be concerned about the accuracy of the tables, because the results are not very sensitive to small changes, nor even to moderately large changes in a small enough number of coefficients. Admittedly, the Ontario input-output table is probably less accurate than the Canadian table, because of the greater difficulty associated with collecting data on a provincial basis. But this fact probably does not account for any significant amount of the difference found between Ontario and Canada. Even if differences in

102 The effects of energy price changes accuracy exist there is no reason to expect them to bias our estimates systematically in any particular direction . Another assumption implicit in input-output models is constant returns to scale. While in contexts where production changes are large, such as in the analysis of tariff removals, constant returns to scale may be unjustified, the assumption is much more defensible in this study where production changes are small. Nevertheless, for those industries with the larger production changes the absence of constant returns to scale could produce biases. A final difficulty associated with the use of input-output tables is that one must use the tables available. For Ontario the only table available uses 1965 data, so that it has been implicitly assumed that this table was appropriate for use with 1971 data . While this may not present serious problems for the calculation of the price changes, in later stages it was necessary to assume that such things as the patterns of consumption and government expenditure did not change between 1965 and 1971 , and this is much more difficult to accept . For the Canadian data the 1963 table was used, and this could produce further biases. It is possible, for example, that some of the differences found for Ontario and Canada were due to the fact that the input-output tables were for different years. While for the estimation of price changes the only major assumptions concerned the input-output tables, when employment changes were estimated, several other assumptions were required, the most important being that utility functions were Cobb-Douglas. No particular justification can be given for this assumption, and as has already been admitted it is employed more because of the simplifications it produces than for any deep belief in its ability to characterize the world. Any use of the employment effects calculated in the study must be made with this assumption in mind. SUGGESTIONS FOR FURTHER RESEARCH

There are, of course, many ways in which the results of this study could be improved. One major problem has been the inability to produce regional price-change estimates on account of the absence of regional input-output tables. While such estimates must await the construction of such tables, several provinces other than Ontario do have input-output tables, and the comparison of the results derived from these with the Ontario results could produce some interesting conclusions. In particular, an examination of comparable estimates for Quebec would be of interest, particularly if the historical price differences which have existed across the Ottawa Valley are perpetuated. It seems quite clear, for example, that past federal policy, which resulted in higher petroleum

Conclusions I 03 prices in Ontario than in Quebec, was an implicit subsidy to Quebec industry . It would be of interest to know exactly how much this subsidy affected relative commodity prices, and how much effect the removal of this subsidy would have for both Ontario and Quebec. As was pointed out, the assumptions about the elasticities of demand have been made as much for convenience as for realism, and of course if actual estimates become available they could easily be substituted into the model. On the other hand, the tremendous problems involved in making such estimates would not be worthwhile simply for the purposes of this study. While these and other refinements could be made , one wonders how much additional information would be forthcoming. Perhaps the most important conclusion has been that even relatively large increases in the price of petroleum and natural gas lead to rather modest commodity price increases, even when the indirect effects are taken into account.

Bibliography

Beigie, C.E. (I 974) Oil prices, inflation, and economic growth. HRI Observations (Montreal: C.D. Howe Research Institute) Canadian-American Committee ( 1975) Keeping options open in Canada-US oil and natural gas trade. (Montreal: C.D. Howe Research Institute) Cragg, J.G., A.C. Harberger, and P. Mieszkowski (I 967) 'Empirical evidence on the incidence of the corporation income tax.' Journal of Political Economy 75, 811-21 Gainer, W.P. and T.L. Powrie (1975) 'Public revenue from Canadian crude petroleum production.' Canadian Public Policy l, 1-12 Grubel, H.G. and S. Sydney Smith (I 975) 'The taxation of windfall gains on stocks of natural resources.' Canadian Public Policy I, 13-29 Harberger, A.C . ( 1962) 'The incidence of the corporate income tax.' Journal of Political Economy 70, 215-40 Jump, G.V. and T.A. Wilson (I 975) 'Macro-economic effects of the energy crisis 1974-1975 .' Canadian Public Policy I, 30-8 Koopmans, T.C., ed. ( 195 l) Activity Analysis of Production and Allocation (New York: John Wiley and Sons) Krzyzaniak, M. and R.A. Musgrave (I 963) The Shifting of the Corporation Income Tax (Baltimore: Johns Hopkins) Martin, F. ( I 975) 'Incidence de la crise de I' energie sur le developpement regional canadien.' Canadian Public Policy l, 39-46 Maxwell, J. (I 973) Observation (Montreal: C.D. Howe Research Institute) McDougall, I. (I 975) 'Canada's oil and gas : an eleventh hour option that must not be ignored.' Canadian Public Policy I, 47-57

106 Bibliography McKinnon, R.E. (I 966) 'Intermediate products and differential tariffs: a generalization of Lesner's symmetry theorem.' Quarterly Journal of Economics 80, 584-615 McRae, J.J. (I 975) 'Economic theory and non-replenishable resources.' Canadian Public Policy I, 58-65 Melvin, J.R. (I 969) 'Intermediate goods, the production possibility curve, and gains from trade.' Quarterly Journal of Economics 83, 141-51 Melvin, J.R. (I 975a) The Tax Structure and Canadian Trade: A Theoretical Analysis. Economic Council of Canada Special Study (Ottawa: Information Canada) Melvin, J.R. (I 975b) 'Some general equilibrium consequences of the corporate income tax.' University of Western Ontario (mimeo.) Melvin, J.R. (I 975c) 'The effect of the corporate income tax on commodity prices, international trade, and employment.' University of Western Ontario (mimeo.) Mundell, R.A. (I 960) 'The pure theory of international trade.' American Economic Review 50, 67-110 Ritchie, R.S. (I 975) 'Public policies affecting petroleum development in Canada.' Canadian Public Policy 1, 66-75 Spencer, B.G. ( 1969) 'The shifting of the corporation income tax in Canada.' Canadian Journal of Economics 2, 21-34 Vanek, J. (1963) 'Variable factor proportions and interindustry flows in the theory of international trade.' Quarterly Journal of Economics 77, 129-42 Warne, R.D. (I 97 I) 'Intermediate goods in international trade with variable proportions and two primary inputs.' Quarterly Journal of Economics 85, 225-36 Waverman, L. (I 975) 'The two price system in energy : subsidies forgotten.' Canadian Public Policy 1, 76-88