MEXICO offers a unique opportunity for studying the effects of devaluation on the international accounts of an underdeveloped country, as it is the only underdeveloped country without exchange restrictions in which there has been an exchange devaluation in the postwar period. The part played by the United States in every sector of the Mexican goods and services account is so predominant that a study confined to U.S.-Mexican transactions will give a fair enough picture of Mexico’s international position as a whole. The present study necessarily stops short of a complete account of both the long-term and the short-term effects of devaluation. Some of these long-term consequences undoubtedly had not become fully effective at the time when the outbreak of hostilities in Korea began a new series of price movements which became the dominant factor in the Mexican balance of payments in the second half of 1950 and in 1951. It is extremely difficult to distinguish statistically between the effects of devaluation and the effects of these other influences which were determining the course of world prices in the second half of 1950 and in 1951. Moreover, though attention is given to the hypothetical record of changes that would have developed if there had not been a devaluation, the statistical information needed for this purpose is necessarily inadequate for some of the major items in the Mexican balance of payments.
During World War II, the gold and foreign exchange reserves of the Bank of Mexico increased greatly, from $54 million at the end of 1941 to $344 million at the end of 1945. The rise was due primarily to a substantial expansion of tourist receipts and unilateral transfers; the value of imports increased by about the same amount as the value of exports. In 1946 the situation changed drastically. Though tourist receipts were still high, receipts from unilateral transfers fell and the increase in the value of imports far outstripped the increase in exports. Receipts from unilateral transfers came mainly from Mexican contract labor in the United States. With the end of the war and the demobilization of the U.S. armed forces, the demand for Mexican labor slackened and their transfers to Mexico fell correspondingly. Wartime shortages of both consumers’ and investment goods in Mexico had built up a considerable backlog of pent-up demand, which was only partially satisfied by the large increase in imports in 1946. Also, the removal of price controls in the United States and the attendant rise in prices resulted in rapidly rising prices for Mexican imports. The volume of exports appears to have increased little in 1946, and export prices rose less rapidly than import prices.
All these changes together produced serious balance of payments difficulties. Between the end of 1945 and the end of 1946, the Bank of Mexico’s gold and foreign exchange reserves fell by $121 million. The situation continued to deteriorate in 1947, and the rate at which the Bank of Mexico was losing reserves in the early part of the year reached such proportions that the Government was forced to take a series of drastic steps, including prohibitions against “nonessential and luxury” imports, a downward revision of export duties, and an upward revision of import duties. These measures, however, failed to halt the drain on reserves, which continued throughout the first half of 1948, and the fixed rate of exchange of 4.86 pesos per U.S. dollar was abandoned on July 22.
In September 1948, President Aleman, in a message to Congress, announced a 23-point program to combat the effects of exchange depreciation.1 The exchange rate was to be allowed to seek its “natural” level; the federal budget was to be balanced; credit was to be provided only to meet the legitimate needs of business; an embargo and taxes were to be levied on certain exports, to increase the domestic supply and to hold domestic prices down; and measures were to be introduced to improve productivity and increase output. A 15 per cent ad valorem duty had already been imposed on all exports “to prevent articles of export from going up in price within Mexico and so that a portion of the foreign exchange gain (by the peso devaluation) be absorbed by the State as it has to meet unusual expenditures.”2
After some fluctuation, the exchange rate was held for a considerable period at between 6.85 and 6.89 pesos per U.S. dollar, with the help of market intervention by the Bank of Mexico. In February 1949, however, the rate began to fluctuate again, falling at times to 7.90 pesos. Intervention by the Bank of Mexico restored it to around 6.97 pesos. Toward the end of April 1949, the rate again lost ground and, despite frequent intervention by the Bank, it fluctuated for some time between 8 pesos and 8.60 pesos. (The 8.60 peso rate was exceptional, however, and was quoted for only part of one day.) Beginning June 17, 1949, a new de jure par value of 8.65 pesos per U.S. dollar was established.
At the 6.85–6.89 peso rate, the Bank of Mexico first gained and then lost reserves. After the adoption of the new par value in June 1949, reserves increased fairly substantially, but then declined a little in the first six months of 1950.
Thus, in order to bring her international exchange accounts into balance, Mexico in a two-year period tried, successively, import restrictions, import tariff increases, and export tariff reductions, and, in effect, two exchange devaluations. In addition, the Government adopted a disinflationary fiscal and restrictive credit policy, balancing the federal budget in 1949 and 1950. By the end of 1949 there was a surplus in Mexico’s international goods and services account (Table 1).
|Exports (f.o.b.) including nonmonetary gold.||365.2||481.8||471.9||434.8||531.5|
|Total goods and services||–168.6||–212.6||–76.5||10.1||40.1|
Between 1947 and 1950 there was a net change of $252.7 million in the Mexican goods and services account. Exports had increased by $49.7 million, and net receipts from foreign travel by $73.4 million. Imports had fallen by $121.8 million, net payments for investment services by $26.6 million, and receipts on account of “other services” by $18.8 million. The purpose of this paper is to assess quantitatively the part played in these changes by the various measures taken in 1947–49 to attain balance. The analysis begins with their effects upon the nontrade items, and then discusses the effects on exports and on imports.
Foreign travel is the most important of the three main nontrade items in the Mexican international goods and services account. It is also the item most likely to be affected by such a corrective measure as devaluation. The fall in the value of the peso relative to other currencies was expected to increase the number of tourists and the volume of their expenditures in Mexico, and to reduce the number and expenditures of Mexicans going abroad.
Any estimate of the effects of the devaluations on the Mexican travel account is necessarily somewhat arbitrary, and yields at best only a rough approximation. On both the receipts and the payments side, the picture is completely dominated by U.S. transactions. According to Mexican data, the United States received in 1946 and 1947 more than 90 per cent of Mexican travel payments, while more than 97 per cent of Mexican travel receipts came from the United States. There is, therefore, no risk of serious error if the analysis is confined to U.S. transactions.
A very large proportion of Mexican travel payments to the United States (88 per cent in 1946 and 89 per cent in 1947) and of Mexican travel receipts from the United States (45 per cent in 1946 and 46 per cent in 1947) arise in connection with border visits, which are largely a consequence of the economic interdependence between the towns on either side of the boundary line, and unlikely to be much affected by devaluation. Mexican travel to the interior of the United States is predominantly for business, health, or family reasons, and it might also be expected that devaluation would have no substantial effects upon this expenditure. Mexican travel payments with the United States had indeed been rising irregularly before the devaluations, and, but for the devaluations, might have been expected to rise somewhat further in 1949 and 1950.
The more important question is the effect of the devaluations on U.S. tourist traffic in Mexico. In seeking an answer to this question, it is convenient to use U.S. data, which are available on a quarterly basis and, therefore, make it possible to trace in detail the movements of travel payments and receipts throughout the period affected by the devaluations.3 The use of U.S. data should make it possible to isolate the effects of the devaluations on Mexican travel receipts by comparison with the record in relation to U.S. travel in other contiguous areas. In fact, only the Canadian record can be studied with advantage for this purpose, as certain special factors which affect the U.S. travel accounts with Cuba and the Caribbean area during 1948–50 destroy the usefulness of any comparison between these accounts and the Mexican account.4
Although the records of U. S. travel expenditures in Mexico and Canada between 1940 and 1947 show some divergence (Table 2), there is sufficient similarity to justify the view that the difference between the trends in 1948–50 can be accepted as a fair indicator of the effects of the devaluations upon travel in Mexico.
U.S. travel expenditures in Canada, which had fallen between 1941 and 1943, principally because of the gasoline and tire shortages, rose rapidly in the period 1944–48. The subsequent leveling-off in 1949 and 1950 was to be expected, for it could not reasonably be assumed that the rate of expansion of automobile traffic that had been taking place would be maintained. Also, the leveling-off appears in line with the leveling-off of recreational service expenditures and the increased expenditures on consumers’ durable goods in the United States.
U.S. travel to Mexico, on the other hand, had been affected by the gasoline and tire shortages in 1942 and 1943 much less than travel to Canada, and the rise in travel expenditures after 1943 was not so steep. Also, the subsequent leveling-off began somewhat earlier than in Canada; in both 1947 and 1948, U. S. travel expenditures in Mexico were less than in 1946. After the devaluation of July 1948, however, in contrast with the Canadian record, U.S. travel expenditures in Mexico increased from an annual rate of $114 million in the first half of 1948 to an annual rate of $128 million in the first half of 1949; and after the devaluation of June 1949, they rose further, to an annual rate of $142 million in the first half of 1950. These two successive increases of $14 million, or $28 million in all, cannot indeed be regarded as measuring accurately the effects of the devaluations on travel expenditures in Mexico. Even if there had been no devaluations, some increase in these expenditures might have been expected in 1949 and 1950 in response to the continuing improvement of transportation and other facilities in Mexico, and the Mexican Government’s intensified campaign for U.S. tourist dollars. But some increase in Mexican travel payments to the United States might also have been expected. These payments had been rising rather irregularly up to 1946, and, except for the effects of the import restrictions of July 1947 and of the devaluations, might have been expected to continue to rise. The increase of $28 million may therefore, perhaps, be accepted as a fair approximation of the net effects of the devaluations on the Mexican travel account as a whole.
Mexico is a debtor country, and during 1947–50 there were two dominant forces—the devaluation of the peso and the recession in world raw material prices in 1949—that influenced its payments on investment services account. Certain investment income payments, such as interest, are dollar liabilities, and although the devaluations increased the peso burden of these interest charges, they did not change their dollar value. Interest payments, however, account for only a small part of Mexico’s total payments on investment services account (Table 3). Of much greater importance are the payments resulting from direct, or equity, investment, which are, in effect, peso liabilities. The dollar value of these dividend payments 5 was strongly affected by both the devaluations and the price recession. This dollar value is related to the exchange rate in two ways: There is the direct effect of the recalculation of peso dividends at a new rate of exchange; given the peso value of the dividend payments, their dollar value will change in proportion to changes in the exchange rate. Also, there is an indirect effect resulting from the increased peso prices of exports whose dollar prices are set in a world market. In Mexico, for example, an increase in the peso prices of minerals means higher peso profits for the mining industry, in which much of Mexico’s foreign capital is invested, and larger peso dividend payments on investment services account. In the case of the Mexican devaluations, however, this indirect effect was considerably weakened by the imposition in August 1948 of a 15 per cent ad valorem export tax, which practically halved the effect of the devaluation on export prices. The transmission of profits, however, was subject to the full direct effect of the change in the exchange rate.
|Payments resulting from direct investment1||42||38||13||34|
Any estimate of the effects of the devaluations on these payments can be only a rough approximation. The data are very limited and do not permit the effects on payments of world price movements to be distinguished statistically from the effects of the devaluations. The only data available (shown in Table 3) are on an annual basis, so that the shorter periods used in estimating the effects of the devaluations on travel account cannot be used here. However, the differences between the payments in 1947, 1948, and 1949 may be taken as approximate measures of the combined influence of the devaluations and of world price movements on investment income payments.
As a first approximation, an estimate of the effect of the devaluations may be confined to their direct effects. These effects may be calculated by revaluing the 1947 dividend payments at the new rates of exchange, on the further assumption that, if world prices had remained constant, the 1947 level of dividends in pesos would have been maintained in the two succeeding years. This calculation yields $12 million as the estimated decline in dividend payments resulting from the fall in the exchange rate to the 6.85–6.89 rate, and $18 million as the combined effect of the two devaluations, i.e., the total decline in the exchange rate from 4.86 pesos per dollar to 8.65 pesos per dollar. These estimates should represent an upper limit to the effect of the devaluations on investment income payments. Alternative calculations, which estimate the effects of the devaluations as residuals by attempting to estimate directly the influence of changes in world mineral prices in dollars on dollar dividend payments, yield even higher estimates. The $18 million estimate may, therefore, be accepted as reasonable, with the caution that it is at best only a rough approximation, and that measurement of the indirect effects would probably reduce it a little.
The main item in the “other services” account is the personal remittances of Mexican contract labor in the United States. During World War II, when there was a severe labor shortage in the United States, receipts on this account were high. Receipts in 1947–50, though not nearly so high as during the war, were fairly stable, amounting to $22.7 million in 1947, $21.9 million in 1948, $18.1 million in 1949, and $20.0 million in 1950.6 There was no reason to expect that the devaluations would have much effect on these receipts and, while there was some decline in the residual items in the account, this decline was probably not due to changes in the exchange rate.
Mexican exports are predominantly raw materials, and their major market is the United States. The proportion of Mexican exports sold in the United States, which had been 51 per cent in 1937, has consistently exceeded 75 per cent since the end of the war, and in 1950 was 86 per cent. The most important exports are of mineral extraction, though in recent years agricultural commodities (primarily cotton) and fish have become increasingly important.
The 1947 reductions in export duties and the devaluations in 1948 and 1949 were expected to increase the volume of exports. Also, since the dollar prices of most Mexican exports are set in world markets, their prices in pesos were likely to be increased. The price incentive effects of the tariff reductions and the devaluations were offset considerably by the 15 per cent ad valorem tax imposed on exports in August 1948. This tax was imposed so that the Government might share in the windfall profits accruing to exporters as a result of the devaluation, but it was expected that a sufficient price incentive for expansion of production and exports would still be left.
The experience of the war suggests, however, that production of most Mexican export commodities is not particularly responsive to price movements. During that period, the prices of raw materials entering the United States were controlled, and the prices received by Mexican producers for most mineral products were remarkably steady. But despite this price stability, the volume of metal production varied widely. For example, the output of zinc, whose price was practically unchanged throughout the war period, increased by more than 80 per cent between 1940 and 1945, and then in 1946 fell by one third. Copper, lead, gold, silver, and antimony, i.e., Mexico’s major metal exports, have similar histories. These variations may, perhaps, be explained as responses to incentive payments for marginal output—a method of increasing output which was widely used in the United States during the war. In the postwar period, while mineral prices have risen sharply, output has varied very little. The output of zinc in 1948 and 1949 was well below the level maintained in 1942–45.
Production of petroleum and petroleum products, which is controlled by a government monopoly, PEMEX, has been restricted more by capacity limitations than by the inadequacy of price incentives. If production were to be expanded significantly, additional investment in new wells and equipment would be a more important factor than higher prices.
Cotton production, also, had been restricted more by physical than by economic limitations, being affected by both the vagaries of weather and the amount of suitable land available; but as the Government’s irrigation program made more land available, cotton production increased. Also, the Government placed an embargo on certain cotton exports in 1950 in order to maintain domestic supplies at relatively stable prices.7
Thus, factors other than price incentives appear to have been responsible for limiting production (and therefore exports) of many of Mexico’s major export commodities, and the effect that devaluation could have on the volume of exports as a whole was rather limited. The mining industry had complained of unprofitableness, so that increased peso prices for its output might have been expected to increase production, or to check any tendency toward decline. But higher prices were more likely to produce higher profits than higher production. Measures other than devaluation appear to have been called for if output was to be expanded.
There were, of course, some lines that were likely to benefit by devaluation. Cotton textiles, for example, had been plagued by high costs and prices which were no longer competitive in world markets. Devaluation, however, was more likely to check any tendency for exports to fall rather than to increase their volume.
Thus, all in all, it appears that there was no good reason for expecting much increase in the volume of Mexican exports as a result of the export tariff reductions and the devaluations.
What happened to Mexico’s exports to the United States after the devaluations is indicated by the data in Table 4. The fact that the data for 1947 and 1948 are not strictly comparable makes it difficult to estimate the effects of the November 1947 reductions of export taxes. (The 1947 figures are available only for general imports, while those for 1948–50 cover imports for consumption, which usually are about 6 per cent less than general imports.) However, since the reductions were small, and in view of the inelasticity of production of export commodities, it seems reasonable to assume that their effects on exports were negligible.
|1st half||2nd half||1st half||2nd half||1st half||2nd half||1st half||2nd half|
|Animals and animal products, edible||6,842||9,584||12,004||15,292||15,884||18,796||20,494||22,998|
|Animals and animal products, inedible||3,816||3,124||3,894||3,290||4,668||2,644||3,714||3,130|
|Vegetable food products and beverages||83,710||52,378||96,276||41,398||87,522||48,628||82,614||62,364|
|Vegetable products, inedible, except fibers and wood||28,030||6,644||6,958||17,804||11,690||7,152||4,106||18,516|
|Textile fibers and manufactures||28,620||37,906||27,150||22,332||15,646||19,838||24,334||34,918|
|Wood and pulp||12,600||16,942||14,380||12,316||7,898||10,454||18,110||21,902|
|Metals and manufactures||69,598||68,288||59,040||95,168||97,672||64,184||86,160||100,108|
|Machinery and vehicles||160||210||506||184||146||118||268||142|
|Chemicals and related products||1,642||1,348||1,186||632||686||760||1,274||3,548|
Three of the main export groups, “edible animals and animal products”, “nonmetallic minerals”, and “metals and manufactures”, which together accounted for about 40 per cent of U.S. imports from Mexico in 1947, show increases after the devaluations, which might prima facie be interpreted as responses to the devaluations.8 For two of these groups—“nonmetallic minerals” and “edible animals and animal products”—however, the changes may be discounted, since they probably were mainly the result of other causes.
The increase in exports of nonmetallic minerals was accounted for almost entirely by petroleum and petroleum products. PEMEX, the Government’s petroleum monopoly, has pursued a policy of expansion in recent years; while most of the increase in production went to meet the expanding needs of the domestic market, where prices were kept relatively stable and a stimulus was thus given to increased consumption, some of the increase was allocated to foreign markets. The decision to encourage domestic consumption might well have conflicted with the desire to encourage exports, so that any actual increase in exports was likely to be a residual and should not be attributed to the devaluations.
Among edible animal and animal products, fresh and refrigerated fish accounted for the increased exports. There has, since the end of the war, been a rapid expansion of a commercial fishing industry, which sells in San Diego and San Pedro, California, and this might have been expected to continue independently of devaluation.
Thus, the increase in exports of metals and manufactures appears to be the most important export movement that can be reasonably attributed to the devaluations. The value of exports of this group increased by $39 million (annual rate) between the first half of 1948 and the first half of 1949, but declined by $12 million between the first half of 1949 and the first half of 1950. It is, however, necessary to distinguish here between the effects of the devaluations and the effects of the rather large world price movements that occurred at the same time. It was primarily these price changes, rather than any change in volume, that accounted for the marked decline in the value of metals exports in the second half of 1949. Copper, lead, and zinc comprise more than 90 per cent of Mexico’s exports of metals and manufactures to the United States. If the value of this group of exports is deflated by a price index9 of the U.S. prices of these metals, the volume may be estimated to have increased by 55 per cent after the first devaluation, and by 22 per cent after the second devaluation, or by $31 million and $13 million (at 1948 prices), respectively.
This change in Mexico’s metals exports may be regarded as a rough approximation to the change in Mexico’s total exports to the United States that is attributable to the devaluations. The U.S. share of Mexico’s exports was increasing during this period, and it is possible that the increase was, in part, due to a diversion of exports that would otherwise have gone elsewhere, especially when the low elasticity of supply of these metals is remembered. Thus, even though exports to the United States apparently changed as a result of the devaluations, Mexico’s exports to the world as a whole may not have been similarly affected, or not to the same extent.
The growth in imports immediately after the war was the primary reason for the imbalance in Mexico’s balance of payments and the decline in the Bank of Mexico’s holdings of gold and foreign exchange. During the period 1945–47, imports, growing more rapidly than any other item in the goods and services account, doubled in value; exports increased by only 64 per cent. It was against the effects of the growth in imports that the various corrective measures were primarily directed, and it was expected that the greatest results would be seen in imports.
Import prohibitions and import tariff increases
The import surplus and the ensuing decline of gold and foreign exchange reached such proportions in the first half of 1947 that the Government was forced to take drastic measures to correct the situation. Between the end of December 1946 and the end of June 1947, the foreign assets of the Bank of Mexico fell by slightly more than $50 million. On July 11, 1947, the Mexican Government published a list of items whose importation was temporarily prohibited. While throughout the period July 1947 to July 1950 new items were added to the list almost continuously, and from time to time some items were removed, it was the initial list of prohibitions that was the most important and that had the greatest impact on imports. Later additions and deletions were comparatively unimportant. The initial list of July 11 included articles in 128 tariff classifications which were deemed to be nonessential.10 Imports of these articles had amounted to $80 million in 1946 and to $110 million (annual rate) in the first five months of 1947, or 17.5 per cent of total imports in 1946 and 20.4 per cent of those in 1947.11
While the import prohibitions were intended as a temporary measure, pending improvement in the balance of payments, their main purpose was exchange conservation, although they also provided protection for certain sections of Mexican industry.12 On the other hand, the general tariff revision, promulgated November 13, 1947, had as its primary purpose the protection of Mexican industry. The revision was intended to restore the protective and fiscal levels of 1942 and to give protection—sometimes at prohibitive tariff rates—above the 1942 levels to industries newly developed in the intervening five years.13 The increased tariff rates were likely also to have some balance of payments effects, but this was not their primary purpose.
In assessing the effectiveness of the import prohibitions, it must be kept in mind that their full effectiveness was necessarily delayed by the way in which they were applied. The effective shut-off date for many of the prohibited imports was in October 1947, and even after that date imports of some prohibited articles were allowed under certain conditions. By the first half of 1948, however, the entry of prohibited articles, except as contraband or because of poor enforcement of the embargo, had practically ceased. Since this was also the period in which the tariff increases went into effect, it is extremely difficult to separate statistically the effects on imports of the two measures without a more painstaking analysis of detailed records than is feasible in this study. Therefore, no attempt has been made to distinguish between the two sets of effects, which for convenience are discussed together simply as the effects of the import prohibitions. This probably is not misleading, for the average increase in tariff rates appears to have been moderate, and the impact on imports also was probably moderate.
The effects of the import prohibitions. The importance of the United States as a supplier of Mexican imports is no less than its importance as a market for Mexican exports. Throughout the period 1947–50, the United States maintained a relatively constant share of the Mexican market for imports, accounting for 88 per cent of the total in 1947 and 1950, and for 87 per cent in 1948 and 1949. In relation to imports, it is therefore again permissible to accept the results of a study of the effects of prohibitions on Mexican imports from the United States as a fair indication of the general effects of the prohibitions.
In the period after the prohibitions became effective, i.e., the first half of 1948, the annual rate of the value of Mexican imports from the United States, as measured by U.S. exports to Mexico, was $533 million, or $117 million less than the annual rate in the first half of 1947. Since U.S. export prices were rising during that period, the decline in the volume of Mexico’s imports was more striking than the decrease in value.
Since the import prohibitions of July 1947 applied not to aggregate imports but to specific articles, it might have been expected that imports of nonrestricted articles would be directly substituted for restricted articles, or that demand might be shifted from restricted imports to domestic goods, thus leading to an expansion in domestic activity and subsequently to increased imports. In either case, the curtailment of imports attributable to the prohibitions would be expected to be less than the value of prohibited imports in the period preceding the embargo. In fact, however, Mexican imports were strikingly reduced in the period after the prohibitions. The questions, then, arise: How much of this decline was due to the prohibitions and how much to other causes? What were the other causes?
A comparison between (1) the declines from the first half of 1947 to the first half of 1948 in the major commodity groups of Mexican imports and (2) the 1947 value of the prohibited articles included in each of these groups is given in Table 5. These data show little correlation between the declines and the 1947 value of imports of prohibited articles in the same groups, and this suggests that the embargo was not the only factor curtailing imports in 1948.
|Commodity Group||Imports from United States||Imports in 1947 from All Countries of Prohibited Articles1|
|Jan–Jun 1947||Jan–Jun 1948||Col 1 minus col 2|
|Animals and animal products, edible||19,236||13,784||5,452||290|
|Animals and animal products, inedible||6,960||4,640||2,320||1,310|
|Vegetable food products and beverages||40,590||37,764||2,826||6,180|
|Vegetable products, inedible, except fibers and wood||16,416||11,906||4,510||—|
|Textile fibers and manufactures||33,280||21,726||11,554||5,790|
|Wood and paper||20,994||18,996||1,998||3,230|
|Metals and manufactures||76,822||55,878||20,944||6,190|
|Machinery and vehicles||283,844||235,566||48,278||87,470|
|Chemicals and related products||47,850||44,794||3,056||530|
The largest absolute decline was in the machinery and vehicles group. This decline, however, was considerably smaller than the 1947 imports of prohibited articles in this group, of which the most important were automobiles, parts, and accessories. The prohibitions restricted the entry of assembled automobiles, but import quotas were granted to domestic assembly plants, and the continued import of parts for assembly in Mexico was permitted. Between 1946 and 1948, the number of vehicles assembled in Mexico rose from 10,460 to 21,597. This is an illustration of how the prohibitions were prevented from being more effective. The recorded decline of $48 million (annual rate) can, however, probably be safely interpreted as a measure of the effects of the prohibitions upon this import group.
Not much of the decline in the imports of other groups seems attributable to the prohibitions. Imports of textile fibers and manufactures, metals and manufactures, and miscellaneous commodities, taken together, fell by $46 million (annual rate) between the first half of 1947 and the first half of 1948. The fact that the 1947 imports of prohibited articles in these three groups totaled only $16 million suggests that most of the decline must have been the result of factors other than the prohibitions. When some allowance is made for such effects as local assembly, or other substitutions for prohibited articles, an annual rate of $10 million seems a reasonable approximation of the probable effects of the prohibitions on imports in these three groups, the rest of the decline probably being due to a decline in demand. Industrial production slumped in the last half of 1947, and reached a low in the first quarter of 1948, when the production index was about 10 per cent lower than in the first half of 1947. This decline in activity undoubtedly weakened the industrial demand for raw and semifinished materials. Also, the high levels of imports in 1946 and 1947 and a slump in retail sales in 1947 left merchants complaining about the slowness of trade and their large and high-cost inventories. The postwar sellers’ market had apparently disappeared by 1948, except for a few luxury imports—mainly automobiles.
Most of the remaining decline in imports was in the agricultural and forestry products groups, which decreased by about $17 million (annual rate) between the first half of 1947 and the first half of 1948. More than half of this decline, however, was in groups including few prohibited imports, i.e., edible animal products and inedible vegetable products. In the latter group it was primarily industrial raw materials14 that declined, while in the former animal oils and fats accounted for most of the decrease. The decrease in industrial activity and a high level of domestic agricultural production probably accounted for most of the decline in these groups; a decline of about $5 million (annual rate) may perhaps be attributed to the prohibitions.
To sum up the effects of the prohibitions on imports, they appear to have accounted for roughly $65 million (annual rate) of the decline of nearly $120 million (annual rate) between the first half of 1947 and the first half of 1948. A slump in industrial activity and in retail trade, combined with high levels of agricultural production, probably account for most of the rest of the decline.
Effects of the two devaluations
After the first devaluation, the annual rate of Mexican imports from the United States declined from $533 million in the first half of 1948 to $515 million in the first half of 1949, i.e., by $18 million; and after the second devaluation the rate dropped to $458 million in the first half of 1950, i.e., by $57 million more (Table 6). Thus, while the peso was devalued by 44 per cent, the value of imports declined by only 14 per cent. Also, since in this period, 1948–50, the United States maintained a relatively constant share of the Mexican market, the fact that the decline in imports from the United States was so small cannot be explained by any shift of demand for imports from other countries to the United States in 1949 and 1950.
|1st half||2nd half||1st half||2nd half||1st half||2nd half|
|Animals and animal products, edible||13,784||10,424||10,194||9,078||8,506||9,980|
|Animals and animal products, inedible||4,640||5,232||4,708||2,580||2,380||3,042|
|Vegetable food products and beverages||37,764||34,228||24,618||25,254||32,552||43,766|
|Vegetable products, inedible, except fibers and wood||11,906||10,294||12,678||9,686||10,598||12,722|
|Textile fibers and manufactures||21,726||23,978||19,388||15,158||12,548||18,232|
|Wood and paper||18,996||13,846||12,722||10,940||11,410||16,540|
|Metals and manufactures||55,878||56,242||64,496||47,982||53,110||56,596|
|Machinery and vehicles||235,566||222,792||233,872||165,264||178,014||256,152|
|Chemicals and related products||44,794||44,276||51,272||46,092||54,366||62,932|
The effect of the devaluations on imports, however, cannot be estimated simply by measuring the change in imports after the devaluations, for the value of imports was also affected by other factors, such as the change in real income in Mexico and Mexican fiscal and monetary policies.
Real income and imports. Real income in Mexico is reported to have increased by 16 per cent between 1948 and 1950.15 An increase of such magnitude would be expected to have a considerable influence on the level of imports. The devaluations do not appear, moreover, to have been a significant factor in expanding income. Unfortunately, existing data do not permit a statistical estimate of the income elasticity of imports in Mexico. It may, however, be reasonably assumed that in Mexico a one per cent change in real income would induce at least a one per cent change in real imports, i.e., that the income elasticity of imports is at least unity; on this assumption, the effect of the change in income on imports can be separated.
National income data for Mexico are available only on an annual basis so that no direct comparison between the first halves of 1948, 1949, and 1950 is possible. Real income, however, is sufficiently closely correlated with the index of industrial production (which is available on a quarterly basis) to justify the use of this index to interpolate values for real income. The interpolated values show an increase in real income of 7 per cent between the first half of 1948 and the first half of 1949, and of 8 per cent between the first half of 1949 and the first half of 1950.16 On the assumption of unit income elasticity of imports, the value of imports corresponding to these real incomes would, if there had been no devaluations and the income effect alone had operated, have been $571 million in the first half of 1949 and $614 million in the first half of 1950—at first half of 1948 prices. Actual imports, at first half of 1948 prices, were $548 million in the first half of 1949 and $526 million in the first half of 1950.
The difference between the two sets of figures—that is, between estimated imports based on increased income and no devaluations, and actual imports—may be taken as an approximate measure of the effects of the devaluations. The effect on imports of the first devaluation would then be estimated at $23 million, and of the two devaluations together at $88 million. Thus, the 44 per cent devaluation of the peso appears to have induced a 17 per cent decline in the volume of imports. It should be remembered, however, that these estimates are dependent on the assumption of unit income elasticity for imports and on the estimate of changes in real income in Mexico. However, any variations in these assumptions that might be regarded as reasonable would probably not substantially affect the results of the calculations. The estimate of the change in income is probably a little high, while the assumed unit income elasticity for imports may be rather low. Thus, any variations that could be regarded as reasonable would tend to offset each other, with little net effect upon the estimates of the effects of the devaluation on imports.
The acceptance of these differences (or residuals) as estimates of the effects of the devaluations on imports means that the effects of any other factors that might have restricted imports are ignored. One such possible factor was the increase in agricultural production in 1948–50. This increase, however, in absolute terms was many times the negligible decline in agricultural imports and appears to have had little effect on imports. Two other more significant possible factors were Mexican fiscal and monetary policy in 1948–50.
Mexican fiscal and monetary policy, 1948–50. As stated earlier, President Aleman’s 23-point program to combat the effects of the devaluation of July 1948 called for the balancing of the federal budget and the provision of credit only to meet the legitimate needs of business. The Federal Government had a cash deficit of 204 million pesos in 1947 and of 305 million pesos in 1948. However, there was a cash surplus of 96 million pesos in 1949 and 121 million pesos in 1950 (Table 7). Since both money expenditures and expenditures in real terms continued to rise throughout the period 1948–50, the development of the cash surplus was not due to any reduction in expenditures. It was, rather, due to a sharp increase in receipts after 1948. While larger tax collections resulting from rising incomes and increased business activity were important causes of the increase in receipts, the most striking factor was the increase in collections from export taxes (Table 8). The 15 per cent ad valorem duty placed on exports in August 1948 raised receipts from export taxes from 5 per cent of total receipts in 1948 to 16 per cent in both 1949 and 1950.
|Millions of current pesos|
|Surplus (+) or deficit (–)||–204||–305||+96||+121|
|Millions of 1948 pesos2|
|Surplus (+) or Deficit (–)||–217||–305||+91||+108|
|Sales and consumption taxes3||295||281||384||482|
The anti-inflationary influence of the federal cash surplus was offset by an expansion in bank credit, which increased the money supply. Though the Bank of Mexico followed a restrictive policy in reducing the real volume of its loans to other banks and its holdings of short-term paper and rediscounts, its holdings of government securities and of “securities of businesses and individuals” (inversiones de empresas y particulares) increased, in real terms, by 70 per cent between the first half of 1948 and the first half of 1950 (Table 9). These increases in fact exceeded the curtailment of short-term credit by the Bank. During the same period, and despite frequent complaints in the banking press of the tightness of the money market, other banks found themselves able to increase their loans to businesses and individuals by 27 per cent in real terms. Real cash balances increased too, rising by 16 per cent between the first half of 1948 and the first half of 1950. A comparison of the increase in average monthly real cash balances and the changes in the federal cash budget between the two years 1948 and 1950 shows that the former was more than double the latter, i.e., real cash balances increased by 865 million pesos against a 413 million peso change in the cash budget. The increase in domestic loans to business and individuals by banks other than the Bank of Mexico, 397 million pesos, was slightly less than the change in the cash budget. (All figures are in 1948 pesos.)
(In millions of 1948 pesos2)
|Real cash balances3||3,805||3,967||4,085||4,219||4,391||4,832|
|Domestic assets of the Bank of Mexico|
|Securities of the Federal Government||946||1,066||1,338||1,553||1,613||1,542|
|Securities of businesses and individuals||155||161||217||244||254||240|
|Loans to banks||127||127||101||76||26||21|
|Short-term paper and rediscounts||695||713||688||422||189||173|
|Assets of other banks|
|Domestic loans to Government||197||191||176||181||192||233|
|Domestic loans to business and individuals||1,351||1,444||1,584||1,640||1,713||1,841|
In addition to offsetting the anti-inflationary influence of the federal cash surplus, the expansion in bank credit and in real cash balances reduced the “absolute price effects” of the devaluations that would otherwise have resulted. “Absolute price effects” may be described as those which depend on the general price level or changes in that level, and which would occur even if all prices were to change proportionally. These price effects usually depend on irrational behavior, on imperfections in the operation of the economic system, or on the inflexibility of the money supply. Since the money supply in Mexico was permitted to expand, the impact of the devaluations on the domestic economy and on imports was reduced. If a devaluation is to reduce a trade deficit, the relation of consumption and investment to income must be altered, for, except temporarily by drawing on stocks, it is impossible simultaneously to consume more domestic goods and produce less domestic goods for home consumption. A change in relative prices cannot bring about a change in the relation of consumption and investment to income if all individuals respond only to relative prices. Only if some individuals respond to the general level of prices can the necessary change in the relation be affected. If there are no absolute price effects accompanying a devaluation, the devaluation will have no effect on the trade deficit.17
Since Mexican monetary and fiscal policies together appear to have exercised little significant influence upon imports, the differences between actual and estimated imports, indicated above, may be taken as estimates of the effects of the devaluations on imports. If the credit expansion materially reduced the absolute price effects attending the devaluations, these estimates indeed probably understate these effects. The available data, however, throw little light on the possible magnitude of these influences, so that the estimates are probably as reasonable as can be obtained.
The effects upon the main components—i.e., foreign travel, imports, etc.—of the Mexican goods and services account with the United States of the measures taken by the Mexican Government to attain balance in its international exchange accounts are summarized in Table 10. In order that the effects of the devaluations on the various components may be summed, the figures in the table are given in current values and not in real terms. They represent, therefore, the values of the changes in the volume of exports and the volume of imports. When the values of exports of metals and manufactures are revalued by the change in metal prices, the changes in exports—discussed above—of $31 million and $44 million (in 1948 values) after the 1948 and 1949 devaluations become $34 million and $37 million, as shown in Table 10. Similarly, the changes of $23 million and $88 million in imports (in 1948 values) revalued by the change in U.S. export prices become $22 million and $77 million.
1948 and 1949
The estimates in Table 10 show that the corrective effects of the devaluations were greater than those of the prohibitions. This was to be expected, since the devaluations affected all components of the goods and services account while the effects of the restrictions were practically limited to the import component. However, both prohibitions and devaluations could reasonably have been expected to have been more effective. The influence of the prohibitions was lessened by poor enforcement and the import of parts for the local assembly in Mexico of manufactured goods. The devaluations could probably have curtailed imports even more than they did if the expansion in bank credit had been limited and the absolute price effects accompanying a devaluation had been allowed greater influence.
Mr. Sweeney, economist in the Special Studies Division, was educated at the University of Chicago and was formerly with the U.S. Bureau of the Budget.
Banco Nacional de Mexico, S.A., Review of the Economic Situation of Mexico, September–October 1948, pp. 5 ff.
Ibid., pp. 9 and 12.
Mexican travel data are consistently higher than U.S. data; the explanation of this difference is not known. The difference between the two was unusually great in 1948, when Mexican data showed large increases (in dollars) for both travel payments and travel receipts, whereas according to the U.S. data there were no such increases. It is reported that “since the depreciation of the peso in 1948, an increase in the number of travelers to Mexico has occurred, although total expenditures in Mexico had not been materially affected by the end of 1948” (U.S. Department of Commerce, The Balance of International Payments of the United States, 1946–48, p. 78), and it is thought that the Mexican figures may be based upon the 1948 number of tourists, multiplied by the estimated average expenditure per tourist in the previous year, rather than by the average expenditure of 1948, which, according to the U.S. Department of Commerce, was lower.
A bad season in Florida in 1948, the result of unfavorable weather conditions, affected travel to Cuba, which is normally closely correlated with Florida travel, and variations in cruise facilities make inappropriate any comparison between the trends of U.S. travel expenditures in Mexico and in the Caribbean area (including Bermuda).
The term “dividend payments” is used here and in the rest of this section to cover all income transfers, such as dividends, branch profits, etc., resulting rom equity investment.
U.S. Department of Commerce figures.
Banco Nacional de Comercio Exterior, S.A., Comercio Exterior de Mexico, p. 180.
The changes in exports, like the changes in the foreign travel account, have been measured by comparing the figures for the first half of 1949 and the first half of 1950 with that for the first half of 1948.
The arithmetic average of the price relatives.
The classes of commodities affected by the Government’s suspension order were canned meat; certain fresh, dried, and canned fruit; wood and metal furniture; tanned furs and fur-wearing apparel; bags, wallets, and purses of leather; cut diamonds; glass and crystal ware; jewelry; refrigerators; certain wearing apparel and hosiery; coated cotton cloth; velvets; carpets; certain cosmetics; wines; kraft paper and cardboard; certain advertisements; catalogs and calendars; phonographs; pianos; watches; assembled automobiles; trucks and buses; cigarettes; automotive chassis; and automobile wheels with tires. For a complete list of restricted items see U.S. Department of Commerce, Foreign Commerce Weekly, July 19, 1947.
U.S. House of Representatives, Committee on Interstate and Foreign Commerce, Fuel Investigation: Mexican Petroleum (Washington, D. C, 1949), p. 108.
Ibid., p. 109.
Ibid., p. 106.
Rubber and allied gums and manufactures, and naval stores gums and resins.
Estimate from Economic Development of Mexico (Johns Hopkins Press, 1953).
The regression equation of income on industrial production for 1947–50 is Y = 104.9879 P + 117.2485, where Y = real national income in millions of 1939 pesos and P = the index of industrial production (1948 = 100). The income figures are from Economic Development of Mexico, op. cit.
This concept has been developed in an unpublished paper by Sidney S. Alexander. See also Sidney S. Alexander, “Effects of a Devaluation on a Trade Balance,” International Monetary Fund, Staff Papers, Vol. II, pp. 263–78 (April 1952).