Taxation of Capital Gains in Developing Countries

ECONOMISTS GENERALLY AGREE that gains from capital are a proper source of taxation in developing countries. This view was expressed in the Technical Assistance Conference on Comparative Fiscal Administration in Geneva in 1951 and more recently in the Santiago Conference on Fiscal Policy for Economic Growth in Latin America.1 A capital gains tax is on the appreciation of capital assets and is commonly imposed only when the increase in value is realized through sale or exchange. It should be distinguished from net wealth tax, death duties, and other capital taxes in that these are assessed on the total value of assets.


ECONOMISTS GENERALLY AGREE that gains from capital are a proper source of taxation in developing countries. This view was expressed in the Technical Assistance Conference on Comparative Fiscal Administration in Geneva in 1951 and more recently in the Santiago Conference on Fiscal Policy for Economic Growth in Latin America.1 A capital gains tax is on the appreciation of capital assets and is commonly imposed only when the increase in value is realized through sale or exchange. It should be distinguished from net wealth tax, death duties, and other capital taxes in that these are assessed on the total value of assets.

ECONOMISTS GENERALLY AGREE that gains from capital are a proper source of taxation in developing countries. This view was expressed in the Technical Assistance Conference on Comparative Fiscal Administration in Geneva in 1951 and more recently in the Santiago Conference on Fiscal Policy for Economic Growth in Latin America.1 A capital gains tax is on the appreciation of capital assets and is commonly imposed only when the increase in value is realized through sale or exchange. It should be distinguished from net wealth tax, death duties, and other capital taxes in that these are assessed on the total value of assets.

Capital gains in developing countries differ from those in developed countries. In the former, capital gains are mainly from the sale or exchange of real estate, and in the latter, chiefly from the sale of securities. Three reasons account for the preponderance of capital gains from real estate in developing countries: the concentration of wealth held in real estate; the dominance in the corporate sector of foreign corporations whose shares are owned by nonresidents who are taxed abroad; and the widespread use of bearer shares, which limits the effectiveness of taxation of capital gains from shares.

Because capital gains in developing countries result largely from investments in land, the taxation of these gains is justifiable in that such investments are not socially productive and are highly speculative. Therefore, a capital gains tax discourages investments that are not in line with the social and economic objectives of developing economies. A capital gains tax is on an increment which generally accrues to the high-income group and thus provides an element of progressivity in the tax system. This function is particularly important in developing countries with a high concentration of wealth and/or with an otherwise regressive tax system. A capital gains tax combined with a well-structured progressive income tax and levies on capital would be useful in promoting tax equity.

The lock-in effect of a capital gains tax in developing countries may have important repercussions on the mobility and composition of investments; on the other hand, the adverse effect on investments is less serious than claimed, judging from the general low rate structure and the restricted nature of the capital gains tax in many developing countries.

Administration and the revenue potential of the tax are to be considered in the introduction of a capital gains tax. Such a tax is complex and necessitates high administrative cost to make it effective. Experience in many countries shows that the revenue yield from the tax is low. However, the high cost of administering a capital gains tax and the low potential yield may be partly compensated for by the fact that tax administrators may gain access to records that aid in the administration of the income tax and other taxes. A developing country considering the adoption of a capital gains tax should weigh carefully its economic, social, administrative, and revenue consequences.

This paper is divided into five sections: Section I discusses the nature of capital gains in developing countries; Section II evaluates the factors to be considered in the taxation of capital gains; Section III describes the various forms under which realized capital gains are taxed in selected countries; Section IV assesses the size of the contribution made by capital gains tax to total revenue in these countries; and Section V discusses some of the problems in administering a capital gains tax in less developed countries.2

I. Nature of Capital Gains in Developing Economies

The composition of capital gains in developing countries is different from that in advanced countries. In advanced countries, a considerable proportion of capital gains comes from the sale of securities. In the United States, for example, 42 per cent of the reported gross capital gains in 1962 came from security sales and exchanges.3 In the United Kingdom it is estimated that the proportion of capital gains from the sale of securities would approximate that of the United States. In contrast,4 in developing countries gains resulting from the sale or disposal of physical assets, particularly in the form of real estate, predominate. Several reasons account for this phenomenon: (1) the concentration of wealth in developing countries held in real estate; (2) the prevalence of foreign-owned or foreign-dominated corporations whose shares are traded in the home country or in other foreign capital markets; (3) the widespread use of bearer shares, which makes it difficult to enforce a tax on capital gains arising from securities. These reasons are examined in the following section.

Concentration of wealth held in real estate

In advanced countries wealth is principally in the form of securities. In the United States, for example, it is estimated that of total investment by the household sector investment in real estate is 28 per cent, against 29 per cent in corporate securities.5 The pattern of investments of top wealth holders in the United States shows that their investment in corporate shares is almost double that in real estate, the former representing 40 per cent and the latter only 22 per cent of the total. Smaller percentages are held in bonds, mortgages and notes, life insurance reserves, etc. In contrast, wealth holding in developing countries is generally concentrated in real estate and other physical assets, while only a small proportion is in corporate securities.6 Table 1 gives a classification of investments for four Asian countries. Investment in corporate stocks is on average 6.1 per cent of total investments in Malaysia, 23.3 per cent in the Philippines, 6.5 per cent in India, and a negligible 0.1 per cent in Ceylon. Investments in physical assets (land, residential construction, and consumer durables) constitute a major proportion of the total investment in these countries, ranging from 39 per cent to 67 per cent. In Ceylon, 60 per cent of total investment was in real estate. In the Philippines, residential construction constitutes 29 per cent of the total investment; physical investment represents 39 per cent of the total household sector investment.

Table 1.

Selected Asian Countries: Pattern of Assets in Household Sector

(In per cent of total)

article image
Sources: Central Bank of Ceylon, Report on the Sample Survey of Consumer Finances, 1963 (Colombo, 1964), pp. 122–25; Reserve Bank of India Bulletin, March 1965, p. 324; Richard W. Hooley, Saving in the Philippines, 1951–1960 (University of the Philippines, Manila, 1963), pp. 22–32; Indian Institute of Asian Studies, Savings in Asian Economy, ed. by M. R. Sinha (Bombay, 1965), pp. 198–214.

In the Philippines, corporate securities represent net investment in new security issues of corporations and amounts paid as capital subscriptions in partnerships and proprietorships. For India, this column represents investments in new and old issues of corporations and cooperatives.

Includes such contractual assets as insurance premiums, contributions to provident funds, investments in mutual funds, and net claims on the government.

Includes consumer durables.

Includes government bonds and savings certificates.

Includes land.

Excludes consumer durables.

A survey of the pattern of urban savings in the city of Dacca, Pakistan showed a similar pattern of investments—39.4 per cent of total investments is in real estate while a negligible 0.04 per cent is invested in corporate shares.7

Investment in real estate in developing countries is explained by several factors, including the lack of developed capital markets, the holding of land as a hedge against inflation, and the desire to gain long-range stability, political power, and social prestige.

A continuing pressure on land in many developing countries—arising from a combination of factors, such as population growth, urbanization, general economic development, and the preference for real estate investment—has the effect of inflating real estate values and results in the accrual of capital gains for the real estate owners.

In many African countries, however, land is largely communal property. The tribe or the community has the right to the land, and the individual members have only the right of use. Even if individual ownership is possible, uncertainty of title often exists. These factors inhibit the sale of land.8

Foreign corporations

The fact that many, if not the majority, of the big businesses in developing countries are foreign-owned or foreign-dominated means that their securities are generally bought by nonresident investors who normally are not subject to the capital gains tax in the country in which the business is carried on. A survey of the largest enterprises in Argentina, Brazil, Chile (excluding copper and nitrate producers), Colombia, and Mexico shows a pattern of ownership in which foreign private capital and domestic private capital share about equal percentages of ownership in 3 of the 5 countries (Table 2). Similar statistics are not available for Africa, but a National Planning Association study indicates that foreign companies dominate the modern African private sector, particularly in finance, manufacturing, and the extractive industries.9

Table 2.

Selected Latin American Countries: Ownership of Large Enterprises1

(In per cent of total)

article image
Source: Based on a survey of the 30–32 largest enterprises as reported by Frank Brandenburg in The Development of Latin American Private Enterprise(National Planning Association, Washington, May 1964), pp. 50–66.

Size of the firm is determined by capital and reserves.

Excluding petroleum exploration and development companies.

Excluding copper and nitrate exporters.

Excluding petroleum companies and two state enterprises owing to unavailability of reliable data. Percentages do not total 100 because of the omission of the Ecuadoran Government’s ownership in a shipping firm.

Mexican law stipulates that at least 51 per cent of ownership must be Mexican.

In Malawi, the importance of foreign capital may be gleaned from the income tax assessment in 1965. Foreign companies (incorporated outside Malawi) accounted for 39 per cent of the total taxable income reported; inclusion of foreign-owned companies incorporated in Malawi would raise this ratio.10 Foreign capital also has made an important contribution to the industrial development of Jamaica and of Trinidad and Tobago. At the end of 1964, foreign capital represented 58.4 per cent of total capital invested in approved enterprises in Jamaica. In Trinidad and Tobago, a survey in 1959 showed that foreign capital was responsible for 83.4 per cent of the total capitalization of 52 firms.11 Foreign equity investment in Pakistan over the period April 1959 to December 1963 accounted for 44 per cent of the total investment, the remainder being owned domestically.12

Foreign enterprises in developing countries operate as either branches or subsidiaries. With foreign direct investment in the form of branches and subsidiaries in contrast to portfolio investment or joint ventures, control remains with the parent companies, and there is less incentive to tap the domestic supply of capital or to promote domestic participation. For example, the value of German investment in developing countries between 1952 and 1961 consisted of 54 per cent in the form of subsidiaries and branches and 46 per cent in minor participations.13 U.S. direct investments in Africa, Asia, and Latin America are mainly those in which control rests with the parent companies. This is shown both in value of investments and in number of firms established between 1951 and 1957. U.S. investments in those firms in developing countries in which U.S. ownership comprises at least 95 per cent accounted for 87 per cent of the total value of U.S. direct investments; 77 per cent of the total number of these establishments was made up of firms with 95 per cent or more of U.S. ownership.14 Capital requirements that are met by parent companies and by retained earnings deter domestic participation in the equity holding of foreign enterprises.

Use of bearer shares

The use of bearer shares, especially popular in Latin America, also inhibits the taxation of capital gains from shares in developing countries. Bearer shares are transferable without any formal registration on the books of the corporation, and thus permit anonymity of ownership.15 Outside Latin America, bearer shares are used in most European countries and in Ghana, India, and Israel. Twelve of 19 countries included in this study have a capital gains tax on sale of securities; but only 6 of the 12 countries which permit the issue of bearer shares have a tax on capital gains arising from sale or exchange of securities (Table 3).

Table 3.

Selected Countries: Bearer Shares and Tax on Capital Gains on Shares

article image

Except those issued by closed corporations and held for more than one year.

Gains reinvested in approved enterprises within 15 months of sale are exempt.

Except those traded on the stock exchange.

II. Considerations Involved in Taxing Capital Gains

The taxation of capital gains in developing countries may be justified on equity and economic grounds. On equity grounds, capital gains enhance a person’s taxpaying capacity. Although realized capital gains are a small fraction of total income, they represent a higher proportion of the income of the high-income recipients than that of those in the low-income brackets. Therefore, taxation of such gains is a progressive element in the tax system and a force for reducing the high concentrations of wealth existing in many developing countries.

An important economic function of the capital gains tax is to curb speculation—especially in real estate, the values of which have been rising owing to population growth, urbanization, and development programs—and to encourage investments which are economically productive. Adverse economic effects of a capital gains tax on the supply of savings and on the mobility of capital need to be weighed carefully against possible undesirable effects on the allocation of investment which may occur when ordinary income is taxed and capital gains are tax free.

Few developing countries are adequately prepared to administer a capital gains tax because of its complex and sophisticated nature. Unless such a tax is administered effectively, evasion and avoidance are encouraged and inequities result.

Equity considerations

Modern income taxation is based on the relative economic position of persons. Since capital gains increase a person’s taxable capacity, their taxation can be supported on equity grounds.16

Concentration of gains

The equity reason for taxing capital gains in developing countries is emphasized by the high concentration of wealth in the hands of the few. Capital gains accrue only to those who own property, and failure to tax these gains would discriminate in favor of property owners and would encourage reinvestment of these gains in assets which would perpetuate severe inequalities in income and wealth.17

Another argument for the inclusion of capital gains in the tax net is the relatively low taxation of real estate in most developing countries. There are two reasons for this: (1) the very low assessment values of real property subject to tax and (2) the low rates of property taxation.18 The forms of property taxation that have been suggested to supplement the real property tax are a net wealth tax, a land increment tax, a capital gains tax, and other capital taxes, such as estate and gift taxes. Ownership of property, particularly financial assets, may be encouraged if a capital gains tax is limited to real estate, but a net wealth or estate tax would be levied on these assets.

The distribution of capital gains by income classes shows that, as income increases, capital gains are an increasing fraction of the total income. For example, in India, the ratios of capital gains to total assessed income of individuals for the fiscal year 1962/63 range from 0.02 per cent to 6.52 per cent for the income brackets between Rs 3,001 and Rs 5,000, and over Rs 500,000, respectively. The distribution of capital gains among resident taxpayers in Ceylon also shows that capital gains as a percentage of assessed income are higher among the high-income groups. A more detailed analysis of the distribution of capital gains by income brackets in Ceylon and India is shown in Tables 4 and 5. Similar distribution of capital gains is revealed in the U.S. data.19

Table 4.

Ceylon: Distribution of Capital Gains as Percentage of Total Assessed Income of Resident Individuals, 1961/62–1964/651

article image
Sources: Administration Report of the Commissioner of Inland Revenue(Colombo), 1962, 1963, 1964, and 1965.

The ratios are derived by dividing assessed capital gains by assessed total income in each income bracket.

Failure to tax capital gains accruing to those who receive high incomes puts a greater relative burden of the income tax on those who do not enjoy such gains. If capital gains are not taxed or if they are given preferential treatment, there is an incentive to convert ordinary income to capital gains.

Unearned increment

Economic development involves the building of roads, public highways, and harbors, the installation of power plants, and other public improvements. Growth of the economy results in rising land values as increases in population and incomes press on the limited resources, and property values increase without the deliberate effort of the owners. Proponents of the capital gains tax argue that gains arising from projects created by the community are unearned increments, and that a tax on them would impose a small sacrifice on the taxpayer. The introduction of the capital gains tax in India was based on the unearned increment argument. In presenting the budget for 1947/48, the Minister of Finance said, “there is stronger justification for taxing these profits than there is for taxing ordinary income since they represent what is properly described as unearned increment.” 20 In justifying a capital gains tax in Malaysia in 1965, the Minister of Finance said “these gains have accrued not as a result of the taxpayer’s own exertions, but purely fortuitously, often as a result of the growing prosperity of the society in which he is fortunate enough to live and to be allowed to operate.” 21

Table 5.

India: Distribution of Capital Gains of Individuals as Percentage of Total Assessed Income, 1960/61–1962/631

article image
Sources: Central Board of Revenue, All-India Income-tax Revenue Statistics, 1960/61 and 1961/62; for 1962/63, the publication was issued by the Central Board of Direct Taxes.

The ratios are derived by dividing assessed capital gains by assessed total income in each income bracket.

The unusually high ratio of 2.64 in 1962/63 is explained by the number of assessees: 461 in 1962/63; 23 in 1960/61; and 22 in 1961/62.

The unearned increment theory of taxation has been advanced to justify the tax on land values. This theory holds that increases in land value are not the result of the effort or skill of owners but of social factors and hence may be fairly subjected to special taxation to finance community needs. Such taxation, moreover, has no adverse effect on economic incentives.22 The capital gains tax in developing countries usually applies mainly to real estate gains, and it may be regarded as a good substitute for a special tax on increments in land values.

Effects of speculation in land and of inflation

Speculation in real estate arises from inflation, population growth, and urbanization, which have characterized most of the developing countries. A notable feature of land prices during an inflationary period is that they increase much faster than the general price index. In Chile, for example, in 1963 the average price index (1955 = 100) of registered real estate sales rose to 865, while the wholesale price index registered 671.23 In Colombia, where inflation was less pronounced than in Chile, in 1965 the average price index (1955 = 100) of registered real estate sales was 615, and the wholesale price index was 317.24

The historical development of land prices in Israel has also shown that population growth, urbanization, and inflation have pushed land prices upward. The demand for land in the Tel Aviv region and other urban centers has resulted in the prices of land increasing by as much as 150 times. For example, in the northern quarter of the Tel Aviv Yafo, land which was priced at I£75 per square meter in 1953 was sold for I£1,000 per square meter in 1963. In the coastal part of Bat Yam, the price of land has increased from I£l per square meter in 1953 to I£150 in 1963.25

The following quotation describes land speculation in Israel:

… unlike these two types of demand for land earmarked for building, there has been the demand for land as a value-accumulating capital. This demand, which is by nature speculative, is governed by three main motives: First, since there have always existed in Israel certain elements of under-development in its economy, and consequently investment possibilities were limited, the capital was often geared towards the purchase of land. This was most conspicuous in periods when the accumulation of capital in the country was speeded up. Secondly, owing to long periods of inflation, investment in land was, in many cases, the best way to secure the value of currency. One more fact can be added to the above factors: As years went by, it became obvious that the prices of land were constantly rising, generally at a faster rate than other prices. Consequently, more people started buying land, expecting a price increase and high profits. This trend was the more visible in view of the fact that the capital market in Israel was undeveloped and the expectancy for a land price increase was of the same nature as the expectancy, in industrial countries, for an increase of stock value.26

The case for a capital gains tax is made on the grounds that investments of this nature are speculative and not socially productive and that a capital gains tax is a suitable instrument for tapping speculative gains. The property tax is not well adapted for this purpose, owing to the typically wide gap between cadastral and current market values.

In Israel, the strengthening of the capital gains tax on land (Land Betterment Tax) in 1963 is said to have brought land speculation to a standstill.27 Many developing countries that have recently proposed or enacted capital gains taxes have done so in order to curb speculation in land values, e.g., Colombia, Ghana, and Korea.

It is argued that the appreciation of capital assets during inflation is merely a paper gain and therefore illusory. Because inflation raises the nominal value of the assets without a corresponding increase in real values, some economists oppose a tax on capital gains under inflationary conditions. In rebuttal, it is argued that a rising price level affects people in different ways. Holders of such assets as corporate shares and urban real estate may make real gains because the value of their property increases more rapidly than the general price level, while holders of bonds and of other fixed money claims suffer real losses. A capital gains tax is defended as a means of reducing such inequities.

Others contend that an adjustment for general price changes should be made by applying a suitable price index (e.g., the consumer price index) and that gains or losses should be recognized only to the extent of changes in deflated values. The correction factor is determined by dividing the price index for the year of sale of the asset by the corresponding index for the year of the purchase.28 The original cost of the asset is multiplied by the correction factor to obtain the adjusted cost. The difference between the sales price and the adjusted cost would give the gain to which the capital gains tax rate is applied to determine the tax liability. Not only would administrative complications be involved but, with any adjustment in price, the stabilizing power of the tax in periods of inflation would be lost.

In opposing the application of deflationary factors, the members of the minority of the British Royal Commission in their Memorandum of Dissent stated

Equity cannot be secured by ignoring relative changes in the taxable capacities of different property owners; and, if it were held to be desirable (and possible) to exempt that part of capital appreciation which was commensurate with the general price rise, it would follow that any lesser degree of capital appreciation should be regarded as a loss. It is not our view that the tax code should be so devised as to insure taxpayers against the risk of inflation. Indeed we should consider any such intention singularly inappropriate, for taxation must be regarded as one of the principal weapons in the armoury of the central government for combatting inflation.29

However, others believe that on equity grounds the capital gains arising from rapid inflation should remain untaxed. Mutén supports the idea of freeing inflationary gains from taxation:

Generally, one might say that the worse the inflation is in any country, the better are the reasons for making the tax system such as to diminish its hazardous effects, either through introducing a valoristic system or through abstaining from capital gains taxes.30

A compromise solution would be revaluing the assets subject to the capital gains tax and taxing only those gains on assets held for a certain number of years; after that, no tax would be charged on capital gains. For example, in Sweden, gains from shares and from movable property held over five years and from real property held ten years or more are tax free.31

Effects on economic development

An appraisal of a capital gains tax includes particularly its effect on investment. Investment is one of the most important variables in economic development, and capital gains taxation is directly related to the supply of savings, the composition of investments, and the mobility of capital.

In the following analysis, it is important to bear in mind the preferential treatment of capital gains given in the majority of the countries studied.

Effects on the supply of savings

In order to analyze the effect of a capital gains tax on savings, it is important to know who are the savers and what effect the tax has on incentives to save. In developing as well as developed economies, savings ratios are high among the high-income recipients and low or negative for those in the low-income brackets.32 In many preindustrial and primary producing countries, landowners frequently are the richest members of the community and a principal source of savings. Savings by other members of the community, particularly the urban commercial class, are also important.

It has been held that the capital gains tax reduces savings more than an income tax of the same yield because the capital gains tax, being a nonrecurrent tax, is more likely to be paid out of capital. Also, it is argued that the income from capital gains cannot be depended upon to meet consumption expenditures and that those who realize capital gains are not likely to increase consumption substantially nor likely to reduce their consumption because of a capital gains tax.33 It is suggested that savings are reduced by “something not far from the amount of revenue produced.” 34

Concessionary rates, combined with such provisions as exemptions and nontaxation of gains transferred at the time of death, mitigate the effect of the tax on savings. It is argued, moreover, that in developing countries, the capital gains tax may influence to a large extent savings that would be directed toward investment in real estate and other assets of low social productivity. The role that the capital gains tax, together with other taxes, plays in penalizing undesirable investment is underscored by Tripathy:

… taxes should penalize the diversion of savings into land, buildings, inventories, and similar investments held for speculative gain or prestige purposes rather than productive uses. In order to achieve these ends such tax instruments as land value increment taxes, taxes on idle land, progressive taxes on either net worth or real estate holdings and tax on capital gains should be imposed. It is necessary to tax capital gains in the developing countries because they provide an incentive for speculative investments which draw savings from the developmental channels.35

Effects on the composition of investments

Preferential taxation of capital gains influences the composition of investment in that a capital gains tax at a low rate induces the shifting of investments to those which yield capital gains from those which yield income taxed at a higher rate, e.g., dividends, interest, and rent. Also, lower tax rates on capital gains than on ordinary income favor the retention of profits in corporations rather than their distribution, reducing the availability of funds for investment in new enterprises and other outlets.36 Furthermore, it is contended that a preferential capital gains tax that induces retention of earnings decreases the liquidity and turnover in the capital market, and lessens the efficiency of the market mechanism in allocating resources. In developing countries a need to develop capital markets and to encourage the diversification of investments is very important.37

In many of the countries studied the preferential rates on capital gains would encourage the reallocation of investment resources. These low rates on capital gains may induce some shifting of investments toward assets producing capital gains. Some of these investments generally represent venture capital, which contributes in an important way to economic development,38 but others are speculative investments in real estate, which contribute nothing to industrialization and productive activities.

Many developing countries have attempted to influence the allocation of investment along desired lines by exempting from tax the gains from the sale of certain securities and of agricultural land. To stimulate investment in shares, Argentina, El Salvador, Israel, and Mexico have excluded gains from the sale of certain securities.

Effects on the mobility of capital

Considerable emphasis has been placed on the lock-in effect of a capital gains tax in inhibiting the sale of capital assets which have appreciated in value.39 The result is an increase in the reservation price of the asset and a reduction in the flow of investment. In developing countries, the lock-in argument is of special importance because of the need for greater investment mobility. It is alleged that the capital gains tax discourages a shift in capital from the unproductive investments in real estate to more venturesome undertakings in commerce and industry; consequently, funds are withheld and unavailable for the development of capital markets.

Thus, in the Philippines, for example, in 1963 a bill to suspend the capital gains tax for five years passed in Congress, but was vetoed by the President.40 The explanatory note to the House Bill No. 2794 stated that its purpose was “to unlock immobile capital … funds locked up in land holding [that] could be mobilized and made instruments of production if some inducements can be given to the owners to dispose of their holdings for purposes of investments.” In explaining the repeal of an earlier capital gains tax in India in 1949, the Minister of Finance stated that the tax was having a bad psychological effect on investment and hindered the movement of capital.41 Late in 1965 Israel exempted from capital gains tax gains on shares traded on the stock exchange because of its alleged dampening effect on the stock market.

The lock-in effect of the capital gains tax has been questioned. In the United States, where capital gains have been taxed since 1913, the validity of the lock-in argument has been attacked. The empirical evidence discounts the heavy weight usually charged to the lock-in effect of the tax. A recent survey in the United States of taxpayers with incomes of $10,000 and over revealed that only 19 per cent of those interviewed stated that they did not sell their appreciated assets because of taxes, while 61 per cent did not sell because of the lack of better investment opportunities.42

The influence of the capital gains tax on the locking-in of investment depends upon the level of rates and the structure of the tax itself. A very high rate will encourage investors to hold on to their assets in order to save on taxes. The holding-period provision also influences the timing of the sale of capital assets. In many countries short-term gains are charged at the ordinary income tax rate, while long-term gains are taxed at preferential rates. Investors may postpone the sale of assets in order to take advantage of reduced tax rates on long-term gains. Another factor concerns the definition of realization as applied to capital gains. If realization is construed as only the sale or transfer for a consideration and does not include transfer at death, the tendency, especially for older people, is to retain appreciated assets to avoid the tax.

Administrative difficulties

One of the most serious objections to a capital gains tax is the unusual administrative burden which such a tax entails. To administer the capital gains tax properly requires records of property held, the length of time held, original costs and related costs, sales price, etc. Developing countries are ill-equipped to cope with the task of recording, unearthing old records, revaluating assets, etc. Inadequate real estate records, including the lack of up-to-date property assessment rolls and the underreporting of sales prices, contrive to make the proper administration of the capital gains tax difficult. Honest taxpayers pay their liabilities while others avoid or evade the tax by resorting to under-the-table deals or other avoidance schemes.

The widespread use of bearer shares makes it possible to transfer shares with complete secrecy and to leave gains untaxed.43 As a result, the capital gains tax tends to rest on real estate transactions, leaving untaxed a large volume of transactions in shares and other claims to property.

III. Capital Gains Tax Structure of Selected Developing Countries

The tax treatment of capital gains varies widely in developing countries. At one extreme are countries which exempt capital gains from taxation: Nigeria, Kenya, and Uganda (which have followed the British tradition44 of excluding capital gains from the concept of income), the Dominican Republic, and Nicaragua. At the other extreme, the Malagasy Republic taxes capital gains at rates higher than those applicable to ordinary income, and the Republic of China, Guatemala, Honduras, Iraq, and Venezuela tax at the full income tax rate. In between are a number of countries which tax capital gains at preferential rates.

Definition of capital assets

Capital gains are generally defined as gains arising from the sale or exchange of capital assets. Capital assets subject to a capital gains tax are defined by statutes and generally represent only a portion of the net wealth of a person or a business.45 Gains or losses from the sale of assets not comprehended in the capital asset concept are taxed as ordinary income at income tax rates rather than at the capital gains tax rate, which is generally lower. Two different approaches are followed in identifying “capital assets.” One includes all assets except those specifically excluded; the other approach specifies those assets chargeable to tax. When the latter concept is used, the assets specified may be quite limited. Colombia, Panama, and Peru, for example, cover only real estate, and Bolivia and the Malagasy Republic cover only urban real estate.

Capital assets subject to the capital gains tax usually include neither stock in trade nor those assets of the business that are not for sale to customers. Such assets include machinery, equipment, buildings, and land, and gains or losses realized from their sale are generally taxable at ordinary income tax rates. Exceptions to this rule are found in Ghana and India, where depreciable business assets and real estate are considered capital assets and are treated in the same manner as other capital assets.46 A going business, including good will, is considered a capital asset, the gains or losses on its sale being subject to a capital gains tax.

The concept of capital assets may also be narrowed by other exclusions. In four of the countries studied (Argentina, El Salvador, Israel, and Mexico), shares are not classified as capital assets. In 1965, Mexico excluded from capital assets shares disposed of by individuals; Israel excludes only those shares which are traded on the stock exchanges. The announced purpose of this exclusion is to encourage investment through the purchase of shares and to develop the capital markets. In 1963, Israel’s announcement of a tax on the gains of share transactions aroused criticism of its possible adverse effects on investment.47 Argentina excludes gains on all securities while El Salvador exempts only gains on proceeds that are reinvested in shares of other companies and not sold within a period of four years.

Personal effects are specifically excluded from capital assets in Argentina, Burma, India, and Pakistan. The disposal of personal effects is most difficult to trace and to detect and to do so would involve excessive administrative difficulties. Even countries which do not statutorily exclude personal property tend, in practice, to ignore this type of property.

Agricultural land is exempt from the tax base in some countries, including Burma, Chile, Ghana, India, and the Malagasy Republic. Such exemption is generally justified on the grounds that there is less speculation in agricultural land than in urban land and that prices for agricultural land do not rise as much as prices for urban land. Moreover, capital gains taxes in many countries are aimed specifically at recouping the speculative values arising from urban development. Exemption of gains from the sale of agricultural land sometimes is believed necessary to unlock investments in large agricultural tracts held by relatively few owners.

Another type of asset, owner-occupied residential property, is commonly excluded from capital assets. The exclusion of the gains realized from sale of this property is frequently limited by a provision that the proceeds must be used to construct or to replace the property within a certain period of time (usually one year). Chile provides an 18-month replacement period from the date of sale; Mexico and India allow a one-year interval. In India, a further condition is made that gains from the sale of residential property are exempt only if the sales price does not exceed Rs 25,000 and the aggregate fair market value of all real property owned does not exceed Rs 50,000; in Chile, the sale of low-cost housing is excluded from tax. Two reasons are given to justify this exclusion: (1) a sales price that reflects replacement value leaves the position of the taxpayer no better than before the sale; and (2) home ownership is encouraged.

The concept of realization and the accrual method

Since capital gains taxes are generally based on the realization principle, it is important to determine when realization occurs. In general, the disposal of an asset for a consideration is regarded as realization; the simplest form is the sale of an asset, but it may include as well liquidation of a business, expropriation of private property, and, infrequently, gratuitous transfers. For example, Israel treats as a realization the liquidation of a corporation and as capital gains the distribution of capital assets in excess of their cost, the assets being valued at their market price. Peru and the Malagasy Republic regard the expropriation of real estate by the government as a realization. Mexico gives capital gains treatment to securities redeemed or amortized at values above their cost of acquisition. Ceylon and Panama appear to be the only countries in this study which consider transfers by donation and by inheritance as constructive realization.48

A comprehensive concept of realization closes possible loopholes for avoidance of the tax. Tax avoidance by means of free transfers has often been cited as a problem in countries where these transfers remain outside the capital gains tax regime. Nonrealization of transfers at death is considered one of the most serious loopholes in U.S. capital gains taxation.49

Taxing gains only when realized engenders a problem of bunching income in the year of realization. Under a progressive income tax system the bunching of capital gains pushes the taxpayer into a high-income bracket, thereby subjecting him to a higher marginal rate, unless the gains are spread by means of averaging (or unless an accrual method is followed).

Holding-period provisions and various forms of preferential treatment of capital gains are employed to mitigate the bunched income effect. The accrual method of taxing the gains has found much support in principle, but poses rather difficult administrative problems. The accrual method includes in the tax base the net accrued gain or loss on capital assets that occurs during a period whether or not realization actually takes place.50

The accrual method is held to be more equitable than the realization method in that

the owner of the appreciated [but unrealized] asset pays no tax on the appreciation. Yet, in the broad sense of the term, his taxable capacity is as great as that of the taxpayer realizing his gains. The former could borrow against the rising value of his assets without selling them and in effect “live off capital” without ever paying any tax on his capital gains.51

Another argument for the accrual method is that full recognition of the unrealized gain would eliminate the lock-in effect of the tax. The tax would encourage the asset holders to dispose of appreciated assets which do not earn enough income to pay the tax currently due. As Seltzer puts it,

when prices [of assets] were rising, investors with large unrealized capital gains would no longer be “frozen” in their assets by their desire to avoid capital gains taxes, and when prices were falling, the decline would not be accentuated by the desire of those with unrealized losses to obtain a tax deduction by selling.52

The difficulties of valuation, especially of assets which are infrequently traded and of unquoted shares, appear to make this approach administratively impracticable. In developing countries, the task of valuation is even more formidable because of undeveloped capital markets and of the type of assets widely held (e.g., equity in family corporations and real estate).

Moreover, owners of wealth may not be recipients of current income, and therefore the taxation of unrealized gains would force many taxpayers to sell their assets in order to pay their capital gains tax liability. Although this may not be objectionable on grounds of wealth redistribution, it may have an unsettling effect on the economy.

The holding period and the rate structure

The capital asset holding period distinguishes short-term capital gains from long-term gains, the former generally being taxed at ordinary income tax rates, and the latter at reduced rates. The holding period is intended to differentiate between transactions entered into for a profit and those made for investment purposes. Short-term changes in value are said to originate in speculative transactions, while investments are made primarily for annual yield and growth in value.

Ten countries included in the study do not provide for a holding period: Argentina, Bolivia, Burma, Ceylon, the Republic of China, Ecuador, the Malagasy Republic, Panama, Peru, and Venezuela. Pakistan has adopted a six-month period as a minimum basis for determining long-term gains, and Chile, India, and the Philippines a one-year period. Colombia, Ghana, Israel, and Mexico use a system which reduces the tax liability according to the length of the period for which the assets are held. Ghana’s procedure operated by means of a reduction in the tax rate, as does Israel’s. Mexico and Colombia reduce the percentage of gains taxable as the holding period increases, and no gains are taxed on assets held over ten years.

El Salvador, which has recently introduced a capital gains tax, is experimenting with the method of averaging the capital gain over the holding period. The average gain is added to the total ordinary income to determine the total taxable income. A progressive rate schedule is then applied to the taxable income. The resulting tax, in turn, is divided by the total taxable income and half of this effective rate is applied to the capital gains.

Various methods are thus employed to give preferential treatment to capital gains: (1) reduced tax rate; (2) lower tax base at ordinary income tax rate; (3) combination of reduced tax rate and lower tax base; and (4) a system of averaging.

The extent of preference or nonpreference given to capital gains may be measured by the ratio of the capital gains tax rate to the ordinary income tax rate. To what extent capital gains are given preferential treatment under a schedular system of income taxation is, however, difficult to determine. Under the schedular method, the rate of taxation varies with different sources of income, so that the degree of differentiation between the capital gains tax rate and other rates is not clear cut as in a global system where a progressive rate schedule is applied to aggregate income.

For purposes of analysis, three capital gains tax rate structures of 19 developing countries are used: (1) flat rate, (2) progressive rate, and (3) percentage inclusion with progressive rates.

Flat rate structure

The flat rate structure is common to countries with a schedular system. For example, Argentina charges 10 per cent, Bolivia 4 per cent on gains from urban real property and 10 per cent on certain rural real estate, and Chile 8 per cent for assets acquired before February 14, 1964 (otherwise, 20 per cent). All countries that use a flat rate charge lower tax rates on capital gains than they charge on ordinary income.53

The first advantage of a flat rate structure is its simplicity. Second, a flat rate tax has the advantage in that a taxpayer is less concerned about spreading the realization of the capital gains over a period, because the tax rate is the same whether or not all the gains are realized in one year. A progressive tax structure (including the percentage inclusion at ordinary income tax rates) may increase tax liability if realized net gains are higher in one year than in other years, and the taxpayer may withhold the sale of assets in order to minimize tax. Under a flat rate system, the taxpayer is not concerned about gains realized in one year against those in other years, as there is no tax inducement to postpone the realization of net gains (assuming that no changes are impending in the capital gains tax structure). An inducement to defer realization under a flat rate system would, however, arise because the taxpayer would realize savings from the interest on the amount of the postponed tax payment. A third advantage of the flat rate structure is closely associated with the second: that is, the mobility of capital is less impeded by a flat rate tax.

On the other hand, a flat rate tax may discriminate against low-income groups and may benefit taxpayers in the higher income brackets, because the differential between a flat rate capital gains tax and an ordinary income tax is greater in the high-income brackets. The ratios of the capital gains tax rate to the ordinary income rates in countries with a flat rate (e.g., Chile) decline as income increases (Table 6). The inequitable result of the flat rate system, therefore, is a point against it.

Table 6.

Selected Countries: Effective Average Capital Gains Tax Rates as Percentage of Average Ordinary Individual Income Tax Rates, by Statutory Income Brackets1

article image

Chile charges a flat rate on capital gains, and the other countries apply progressive rates.

Progressive rate structure

Countries which tax capital gains at progressive rates include Burma, Ceylon, the Republic of China, Ecuador, Ghana, Israel, the Malagasy Republic, Peru, the Philippines, and Venezuela. With the exception of Burma, Ghana, the Malagasy Republic, and Peru, these countries apply the progressive ordinary income tax rates to capital gains. Ecuador makes an exception for gains on occasional sales of securities, which are taxed at a flat rate of 8 per cent. Ceylon applies the ordinary income tax rates, with a maximum flat rate of 25 per cent; Israel also applies ordinary income tax rates, with a ceiling of 25 per cent which is reduced by 5 per cent a year and eliminated after 18 years.54 Peru and Burma have preferential progressive rates depending upon the size of the taxable gains. In Peru the rates applicable to individuals and corporations range from 7 per cent on gains of S/. 10,000 to 15 per cent for those over S/. 100,000. The rates for Burma are 20 per cent on gains of K 5,000 to K 25,000, 30 per cent between K 25,000 and K 100,000, and 40 per cent on gains over K 100,000.

The Malagasy Republic has a progressive capital gains tax of 10 per cent, 20 per cent, and 30 per cent. The effective capital gains tax rates are higher than the ordinary income tax rates for incomes of FMG 80,000 to FMG 1,000,000. On incomes of over FMG 1,000,000 the capital gains rate is a flat rate of 30 per cent, against the ordinary maximum income tax rate of 40.5 per cent.

A progressive tax structure is held to be more equitable than a flat rate tax, since capital gains increase a taxpayer’s taxable capacity. The equity justification for a progressive rate structure is strengthened by the fact that capital gains are distributed in favor of the propertied class and the high-income taxpayers. A low flat rate favors investors and speculators in property over other income earners.

Arguments against the progressive structure of capital gains taxation arise from two considerations: equity and capital mobility. As capital gains realized in one year may have been accrued over a number of years, graduated rates applied to gains in a single year may unduly burden the taxpayer by pushing him into a higher income tax bracket. Such higher tax on bunched income has been mitigated in some countries by the use of preferential progressive rates. Another method to reduce the burden of progressive rates is to spread the realized gains over the number of years the asset is held, and to tax the average annual increment at the taxpayer’s marginal tax rate for that year. This may, however, be too complicated to administer, especially in developing countries.55

The second argument against a progressive rate structure is that graduated rates encourage postponing the sale of the assets so as to minimize the tax liability, and thereby impede the mobility of capital. The sale of appreciated assets may be withheld to avoid bunching of gains, sometimes until capital losses are realized on other assets in order to offset any gains realized during the year.

Percentage inclusion at progressive rates

The percentage-inclusion plan apparently originated in Sweden and continues to be used there. In the United States, the Revenue Act of 1934 provided for an inclusion of a percentage of capital gains in the taxpayer’s ordinary income corresponding to the number of years the asset was held. This method of taxing capital gains was simplified in 1938 by a single holding period with the inclusion of one half of the long-term capital gains.56

Colombia, Mexico, and the Philippines employ the percentage-inclusion plan—charging progressive ordinary income tax rates on a portion of capital gains realized during the year. The percentage of gains included in the taxpayer’s income is based on the length of the holding period. The following data show the proportions included by Mexico:

article image

In Colombia the amount of gains taxable is reduced by 10 per cent for each year the property is held, including the years before the introduction of the capital gains tax in 1960.

A simpler version of the percentage-inclusion method is to take a portion of capital gains, regardless of the holding period above a certain minimum, and to charge this portion either preferential rates or ordinary income tax rates. Pakistan charges normal income tax rates on net capital gains minus PRs 10,000 or two thirds of the net capital gains, whichever is greater, on assets held six months to five years. Normal income tax rates are charged on gains minus PRs 20,000 or five sixths of the net capital gains, whichever is greater, on assets held more than five years. Companies are taxed at 20 per cent of capital gains on assets held six months to five years and 10 per cent on assets held over five years. The Philippines follows the U.S. system of including one half of long-term capital gains in taxable income, but with no maximum alternative rate (25 per cent in the United States).

The percentage-inclusion plan, combined with progressive rates, takes into account the increased taxable capacity with capital gains, and at the same time reduces the tax burden on a lumped income by including only a portion of the gains. The step-scale inclusion method used in Mexico and Colombia has the added advantage of differentiating taxpayers with net gains that have accrued over different periods. Compared with full taxation of capital gains at ordinary rates (with or without income averaging), however, the percentage-inclusion method is especially advantageous to high-income groups.

The step-scale inclusion plan is criticized on the grounds that it induces taxpayers to concentrate transactions at certain time periods. The critics of the step-scale method in the United States pointed out that the preferential treatment of gains based on different holding periods gives taxpayers an incentive to take losses at the earlier high rates and to postpone taking gains.57 For example, under its system in effect between 1934 and 1938, 80 per cent of the gains were included in the ordinary income if the assets were held between one and two years. Data showed that taxpayers would tend to realize losses within the two-year period but would postpone realizing gains after two years, when only 60 per cent or less of the gains would be taxable.

Besides impeding transactions in appreciated assets, the step-scale inclusion is administratively complex. The percentage inclusion with a single holding period is an attempt to simplify the integration of the capital gains with the income tax concept and the administration of the tax but still does not fully satisfy the objective of equity among all taxpayers, i.e., between those who have gains and those who do not.


In addition to excluding certain transactions from the capital gains tax, some countries provide for basic exemptions of certain amounts from the gains. Table 7 indicates the exemption limits in absolute terms and the ratio of the exemptions from capital gains to ordinary income for individuals.

Table 7.

Selected Countries: Capital Gains and Ordinary Income Exempt from Taxes

article image
Sources: Price Waterhouse & Co., U.S.A., Information Guide for Doing Business in Argentina, February 1966; Board of Inland Revenue, Income Taxes Outside the United Kingdom, 1966, Vols. 1, 2, 4, and 6 (London, 1967); Notice sur les impô ts droits et taxes(Tananarive, 1964).

If capital gains accrue to persons who have no other taxable income for that year and the preceding two years, or the total assessable income for three years did not exceed the total family allowances.

From the sale of movable property except shares and debentures.

Or five sixths of capital gains, whichever is greater, for assets held over five years.

Or two thirds of capital gains, whichever is greater, for assets held six months to five years.

The ratio of capital gains exemptions to income tax exemptions for these countries ranges from 66.7 per cent to 333.3 per cent; except for Ceylon (movable assets) and the Malagasy Republic, capital gains tax exemptions range from 1.2 times higher to 3.3 times higher.

Unlike ordinary income tax exemptions, however, capital gains tax exemptions do not allow additional exemptions for dependents. It is not clear what is the economic or equitable basis for granting capital gains tax exemptions, as 5 out of 6 countries which provide for exemptions have preferential rates for capital gains. The position in the Malagasy Republic is different because rates on capital gains generally are higher than ordinary income tax rates, and the use of exemptions is a way of moderating the gains tax burden.

The exemption of a certain amount of capital gains may be desirable from an administrative point of view. Because of the nonrecurrent nature of capital gains and the relative ease of concealing capital transactions—especially those involving small amounts—the provision of an exemption limit would greatly reduce the number of cases requiring enforcement and processing.

Treatment of capital losses

The treatment of capital losses has as important a bearing upon equity and investment incentive as the preferential taxation of capital gains; yet many developing countries neglect proper tax treatment of capital losses. Several of the countries in the study apparently have no provisions concerning the treatment of capital losses: Bolivia, Ghana, the Malagasy Republic, Panama, and Peru. Except for Colombia, where capital losses are allowable deductions from gross income with no provisions for carry-over, the rest of the 19 countries limit the offset of capital losses to capital gains, but with provisions for their offset against capital gains of future years.

Ceylon provides an indefinite carry-forward of net capital losses. Another interesting feature of the capital loss treatment in Ceylon is the carry-back of capital losses at death against capital gains; if gains are not sufficient to cover losses, they may be offset against ordinary income for three years preceding the year of death.

In India, short-term and long-term losses are not distinguished for the purpose of their deductibility from capital gains during the year of assessment. Provisions for carry-forward allow only short-term capital losses to be set off against future short-term gains for a maximum period of eight years, while net long-term losses can be carried over for only four years. A net capital loss of less than Rs 5,000 cannot be carried forward. India’s treatment of capital losses, aside from being asymmetrical in its treatment of short-term and long-term losses, favors those in the high-income group against the middle-income and low-income earners. Those in the latter group are often at a disadvantage because they have less chance of timing the sale of assets in order to realize capital losses to offset capital gains, and because they are more affected by the Rs 5,000 limitation.

In the Philippines, long-term capital losses can be deducted only from long-term capital gains in the year of assessment. Any remaining net capital loss, however, may be carried over to succeeding years as a short-term capital loss and deducted from future short-term gains. Since short-term gains are taxed at ordinary income tax rates, the carryover of a net long-term capital loss provides a tax benefit to those who have net short-term capital gains.

Israel previously provided for an offset of capital losses against gains of the previous two years and the succeeding five years, but now provides for a seven-year carry-forward.

Among the Latin American countries, Argentina has one of the most liberal provisions for the carry-over of capital loss. Net capital losses are deductible from subsequent capital gains for ten years. Chile has a two-year carry-over, while in El Salvador and Mexico net capital losses may be carried forward for five years.

Colombia’s treatment of capital losses is unusual in that they are deductible from other income, but the loss is reduced by 10 per cent a year from the date of the acquisition of the asset to the year of sale. In order to prevent fictitious sales prices, provision is made to disallow a loss based on a selling price lower than the current assessed value on the date of sale unless the taxpayer can prove that the selling price corresponds to the market value. The 10 per cent per annum reduction of losses applies to assets acquired before the law became effective in 1961. Thus, a loss on an asset which was acquired in 1958 and sold in 1967 would be fully absorbed by the taxpayer.58

A fundamental problem with the offset of capital losses against ordinary income is the opportunity given to the taxpayer to time the sale of his assets so as to realize a loss. A system where ordinary income rates are higher and more progressive than capital gains rates would give an inducement to realize losses and to postpone the realization of gains. Large investors are in a position to take advantage of such manipulations because of their diversified investments.

Whatever the form of the allowance for loss—whether it be a loss offset against capital gains during the year of assessment, carried over against future gains, or against ordinary income—it is important to consider the effect of such provisions on liquidity and risk taking. Loss allowances not only ease the investors’ financial strain but also strengthen investment incentives by permitting losses to be shared by the government.

IV. Revenue Yield

The capital gains tax yields little revenue in developing countries. Data for 8 of the countries in Table 8 show that in a recent year revenue ranged from less than 0.05 per cent of total income tax collections in Burma and Ghana to 3 per cent in Argentina. In the Philippines the capital gains tax on individuals amounted to 6 per cent of individuals’ income tax collections.

Table 8.

Selected Countries: Capital Gains Tax and Income Tax Collections

(In thousands of national currency)

article image
Sources: Argentina: Ministerio de Hacienda, Memoria, 1965. Burma: Ministry of Finance and Revenue, Budget Estimates, 1962–63. Chile: Norman D. Nowak, “The Chilean Tax System,” Bulletin for International Fiscal Documentation, Vol. XX (1966), p. 453. India: Central Board of Direct Taxes, All-India Incometax Revenue Statistics, 1962–63, and Comptroller and Auditor General, Central Government Finance Accounts, 1962–63. Peru: Contraloria General de la República, Cuenta General, 1963. Philippines: Internal Revenue Commission, Annual Report, 1964. Venezuela: Ministerio de Hacienda, Memoria, 1964.

Less than 0.05 per cent.

Amount collected for the Land Betterment Tax only; does not include the capital gains tax collected from other private property.

Collection from individuals only.

The relatively low yield of the capital gains tax can be explained in part by evasion of the tax, including failure to disclose transactions and the underreporting of the sales price. A report submitted to the 1960 Congress of the International Fiscal Association in Basle stated that in Israel there was widespread evasion of the capital gains tax on private property, and that administration of the tax was practically ineffective.59 Argentina lowered the capital gains tax rate in 1960 principally because the high rate was believed to be one of the major reasons of “hidden assets” and false declarations. In Ghana, where the tax on gains was in effect for only a short time, real estate transactions were rarely reported for capital gains tax purposes, even though they were registered with the Lands Department and published in the Gazette.

Table 9 shows the trend of the capital gains tax yield in 4 countries: Argentina, Burma, India, and Peru. In absolute amount, the capital gains tax yield shows, generally, an increasing trend. An exception to this is Burma where the yield had been declining, but a sizable increase in revenue was registered in 1960/61. Capital gains revenue as a percentage of income tax collections has been declining in Argentina and Peru, but rising in India. The declining relative importance of revenue from the capital gains tax collection in Argentina and Peru has been due principally to the increases in income tax revenue collected. Argentina’s revenue from the capital gains tax increased by about 68 per cent from 1955/56 to 1961/62, but income tax revenue rose five times in the same period. The decline in the ratio of capital gains tax revenue to income tax revenue in Argentina is also attributable to changes made in 1958 and 1959 which narrowed the taxable base for capital gains. In 1958, a provision for postponing the tax on gains from housing (rollover) was introduced, and in 1959 securities were exempted and the revaluation of assets was provided in order to mitigate the impact of inflation.60 Peru’s capital gains tax revenue increased even more slowly—in 1962/63 revenue was about 1.7 times the 1955/56 yield, against a fivefold increase in income tax collections. In India, the increasing ratio of capital gains tax revenue to income tax revenue can be explained by the increase in the number of assessees and assessments since the reintroduction of the tax in 1957/58.

Table 9.

Selected Countries: Capital Gains Tax, 1955/56–1962/63

(Columns 1, 3, 5, and 7 in thousands of national currency)

article image
Sources: Argentina: Ministerio de Hacienda, Memoria, 1960/61 to 1965; Burma: Ministry of Finance and Revenue, Budget Estimates, 1957–58 to 1962–63; India: Central Board of Revenue, All-India Income-tax Revenue Statistics, 1957–58 to 1961–62 (for 1962–63, the publication was issued by the Central Board of Direct Taxes), and Comptroller and Auditor General, Central Government Finance Accounts, 1959–60 to 1962–63; Peru: Contralora General de la República, Balance y Cuenta General de la República, 1957 to 1963.

Less than 0.05 per cent.

Assessments completed during the year.

Calendar year; year 1956 included under Fiscal Year 1955/56, etc.

Data on capital gains tax collections for other countries are unavailable because revenues are combined with income tax collections.

V. Administration of the Tax

Effective administration of tax laws rests largely on their simplicity and public acceptance. Although it appears simple in principle, the taxation of capital gains introduces many complexities in the tax laws that may not be apparent on first view. The requirements for administration of the capital gains tax have been cited as one reason for the United Kingdom’s reluctance prior to 1965 to adopt a tax on long-term gains.61 Malaysia repealed its capital gains tax in 1966, a year after its adoption, largely for administrative reasons. Ghana also repealed its capital gains tax in 1967, largely because of poor performance.

The administrative problems of capital gains taxation are involved with (1) identification of taxpayers, (2) the problem of valuation, and (3) its international aspects.

Identification of taxpayers

The problem of capital gains tax administration is accentuated by the fact that capital gain or loss transactions are mostly nonrecurrent. Since the roll of taxpayers with capital gain or loss differs from year to year, a register of taxpayers is not so useful an administrative device as in the case of income-tax payers. Information on capital transactions must be obtained, and vigilance of indirect sources must be maintained.

The search for capital gains tax liability may be facilitated by access to the following sources: (1) Income tax returns, especially those showing high incomes or income derived from real estate and other investments. (2) Declarations of assets for the purpose of a net wealth tax, and statements of assets and liabilities sometimes required of all who file income tax returns, as in Argentina, Brazil, and the Philippines. (3) Record of property transfers kept by the valuation office or by the registry of public transfers. (4) Records of real estate sales kept by official notaries. In Venezuela, for example, official notaries prepare “red cards” listing the parties to the transactions, and other information. (5) Municipal records showing property taxes paid, building permits granted, and lists of rental information, etc. (6) Advertisements in the national and local press and in trade journals.

In many countries, gains derived from sale of personal property, especially stocks and bonds, are more difficult to discover than real estate gains. Sale and transfer of intangibles can be easily concealed, while the sale of real estate is relatively easy to trace. The prevalence of bearer shares, especially in Latin America, creates a special problem, because there is no record of the transaction on company books. Another problem arises in the transfers of rights rather than of titles to property, not requiring registration. Unless the transfer of rights to property is specifically considered a realization event, such transfers would avoid a capital gains tax. This was the practice in Israel, which led to modification of the Land Betterment Tax in 1963.

Aside from the problem of discovering transactions that result in realized taxable gains, there is the problem of avoidance through free (gratuitous) transfers. Kaldor’s proposal for capital gains tax reform in India in 1956/57 was to treat free transfers as capital gains transactions. Kaldor maintained that

If … transfers by way of gifts, bequests etc., are exempted, not only will a considerable part of capital gains escape from taxation, but the taxpayer will be given an unhealthy incentive so to manipulate his asset holdings as to make net unrealised gains as large as possible, since all unrealised gains are wiped out for tax purposes when the property passes hands on decease. These exemptions may easily cut the potential long term yield of the tax by two-thirds or three-quarters and they do not seem to be justified on any principle of equity.62

For administrative reasons, however, the Indian Government continued the exemption of gratuitous transfers.

It has been argued that capital gains which escape taxation through gratuitous transfers are subject to gift taxes and death duties, since these taxes are levied upon the total value of wealth, including increases in capital value. However, in countries where gift taxes and death duties are very low they would not be an adequate substitute for the capital gains tax.

Two approaches are considered to deal with the tax treatment of gratuitous transfers: (1) treatment of property transmitted at death or by gift as constituting realization (presumptive realization) and (2) carrying over the cost to the decedent or donee for the purpose of computing the capital gains tax on transferred property. The administrative difficulties of both approaches would entail revaluation of assets and considerable reporting of estates.

The valuation problem

A tax on a change of asset value over a period of time presents a serious valuation problem because of the need to establish both the initial value and the sale or disposal value of the asset sold.63 The introduction of a capital gains tax raises the question of what initial asset values should be recognized. This is necessary to establish the tax basis of the assets with respect to which the capital gain is measured. There are two possibilities: (1) acquisition cost and (2) the market value at an appointed day, usually the date of enactment of the tax. (Sometimes gains are determined by applying to sales value a ratio that reflects average changes in prices during the period.)

The initial value of the asset may be stated as the acquisition or original cost to the taxpayer. Acquisition cost is defined as the amount or value of the consideration which the person paid for the asset, together with the incidental costs of acquiring the asset plus capital expenditures on the asset after acquisition. For an asset which is not bought but created (e.g., a copyright), the expenditure which the taxpayer incurred wholly and exclusively in creating such asset is deemed its acquisition cost. In practice, however, the acquisition cost of the asset to the taxpayer is taken as the basis of initial value only when it was acquired since enactment of the law. For equity reasons, most countries use the market value as of the enactment date of the tax if the asset was acquired prior to this date. By this rule, retroactivity of the tax is avoided with respect to unrealized gains before the adoption of the law.

In any case, when a taxable transaction takes place, problems of administration arise in tracing the basis for taxation of property. It may be necessary to go back many years to reconstruct what the value was at the date of enactment of the law, frequently on the basis of inadequate records. In some countries—including the Philippines and some Latin American countries—the cadastral value may be employed. But this is not dependable because of the wide disparity that often exists between cadastral values and market values, especially when there has not been a recent revaluation. For this reason, the enactment of a capital gains tax should be accompanied by a new cadastral survey that updates all real property values. Sometimes this is accomplished by the use of coefficients applied to an old cadastral value.

Different formulas have been adopted to update original cost or cadastral values. In the Malagasy Republic, for example, initial value is determined by a coefficient which varies with the acquisition date. According to this, original cost is multiplied by 50 if the asset was purchased before January 1, 1920; 20, if before January 1, 1939; 6, if before January 1, 1945; 2, if before January 1, 1949; and 1.5, if before January 1, 1957. This approach ignores variations in the rate of change in the value of individual properties and is quite arbitrary. But it has the virtue of allowing for rapid changes in real estate value. In Argentina, an optional revaluation of fixed assets was made in 1960 whereby fixed coefficients derived from official statistics for changes in domestic prices were applied to original cost. Chile also determines the initial value of an asset by applying to its original cost a coefficient based upon a consumer price index.

A problem of valuation also arises when the asset is sold. Because of the ease with which the price of property can be concealed, an independent check is frequently necessary. Bhargava has pointed out that in India realized gains were frequently much higher than reported to tax authorities. However, in discussing this problem Bhargava states

When the seller tries to understate the price realised by him, it should be noted that the buyer’s interest will be to overstate the price he paid in order to escape the C.G.T. [Capital Gains Tax] when he, in turn, effected a sale. This conflict between the interests of the buyer and the seller was the strongest safeguard of the interest of the state and no buyer was likely to agree, in his own interest, to declare a lower purchase price than what he actually paid.64

The difficulty of ascertaining the actual sales price of an asset has led to the use of a presumptive sales price which may be expressed as a factor of the acquisition cost, the cadastral value, or the market value at date of enactment of the law. The presumptive sales price would ordinarily reflect changes in property values during the period the asset was held. The presumptive approach is arbitrary and laborious, and is little used.

The capital gain itself may also be determined directly by the presumptive approach. The taxable gain would be fixed at a certain percentage of the sales proceeds regardless of the actual gain.65 Among the countries in this study, Ghana employed this technique as an alternative method of ascertaining the gain. If the historical cost could not be determined, the capital gain was said to be 25 per cent of the sales price less allowances for alterations, improvements, and the cost of disposal. This method avoids the calculation of the initial values, but still requires checking the correctness of the sales price. The major problem is what per cent of the sales price should represent the gain. This can be determined empirically on the basis of sample transactions. Once the coefficient is determined, it need only be revised periodically to reflect price changes.

International aspects of capital gains tax

The international problems of capital gains taxation involve the difficulty of taxing transactions of residents and of nonresidents. The question arises of how to treat the capital gains or losses of foreigners who sell shares of stock of a corporation deemed to be resident in developing countries. Unlike dividends and interest income accruing to foreign shareholders, capital gains realized by foreigners are difficult to reach by a tax. Tax on dividends and interest can be easily collected at the source, while withholding on capital gains is not practicable. Another matter of concern is in regard to residents or citizens realizing capital gains or losses in transactions abroad either on shares of domestic corporations or on shares of foreign corporations abroad. Depending of course on the jurisdictional rule followed and the administrative efficiency of the taxing country, there is a good chance that residents can escape altogether a tax on capital gains.

There are two general ways of determining tax jurisdiction: (1) by residence or nationality of the taxpayer and (2) by source of income. Taxation by residence or nationality of the taxpayer would tax the capital gains where the taxpayer is a bona fide or permanent resident or a national of a country regardless of the country of source of income. The residence or nationality rule is justified on the grounds that a country protects its residents and citizens, and that this protection should be matched by its right to collect taxes from them. The source rule would require taxpayers to pay tax on income arising within the border of the taxing country. Taxation at source is justifiable on the principle that benefits received by foreigners are associated with income earned in the country where the economic activity is performed.

The tax treatment of capital gains realized by residents and nonresidents varies among countries. There is, however, a trend among developing countries to tax their residents and nationals on their world wide income as far as personal income is concerned. This trend is less clear for corporations.66 In most instances, the tax rule governing the jurisdiction over income applies to the treatment of capital gains. In India, nonresidents are taxed on income deemed to accrue or to originate in India and on income received or deemed to be received in India. For residents, the residence rule applies in that all income is taxable including that obtained outside India. A special provision is, however, made for capital gains for both residents and nonresidents. According to this rule, capital gains are taxable in India if the capital asset sold, exchanged, or transferred is situated in India.67 The Philippines adopts the residence rule in taxing its residents and nationals. Nonresidents, however, are taxed on the basis of the source rule. Argentina, in general, applies the source rule for both resident and nonresident taxpayers. Resident individuals are taxed on foreign income received from “occasional” activity carried on abroad. According to the Colombian income tax legislation, resident nationals are taxed on all income, including that from abroad; foreigners and nonresident nationals are subject to tax on their income originating in Colombia. For the purpose of the capital gains tax, nonresidents pay the Colombian tax on capital gains resulting from the transfer of real property located in Colombia.

The allocation of taxing jurisdiction is often resolved by international tax treaties. For example, the tax treaty (March 18, 1958) between Germany and India provided that “capital gains arising from the sale, exchange or transfer of a capital asset, whether movable or immovable, may be taxed in the territory in which the capital asset is situated at the time of such sale, exchange or transfer. For this purpose, the situation of the shares of a company shall be deemed to be in the territory where the company is incorporated” (Article X).68 On the other hand, the tax treaty between the United States and India gives the jurisdiction on taxation of capital gains from immovable property to the country where the property is situated; but in regard to capital gains derived from the contribution of capital by a resident in one country toward the formation of a new company incorporated in the other country, the country of residence of the shareholder has the jurisdiction over the tax. Article IV of the Israel-U.S. tax treaty (September 30, 1960) provides that “gains, profits and income derived from the purchase and sale of movable property shall be treated as derived from the country in which such property is sold.” Gains from the sale of real property are deemed to be income in the country where the real property is situated. Similar provisions in the tax treatment of capital gains between the United States and Israel are found in the tax treaties between Pakistan and India and between Japan and Malaysia.

A model tax convention which was formulated by the Fiscal Committee of the Organization for Economic Cooperation and Development (OECD) would tax capital gains from immovable property in the place where the property is situated. Capital gains from movable property would be taxed in the country of the seller’s residence.69 It is noted that more and more tax treaties follow the OECD model in the allocation of tax jurisdiction on capital gains.

Imposition des gains en capital dans les pays en voie de développement


Cette étude porte sur 19 pays en voie de développement qui imposent les gains en capital et elle examine les principales questions liées à cet impôt, notamment la définition des avoirs en capital, la notion de réalisation, les méthodes d’imposition des plus-values non réalisées, la durée de possession, l’échelle des taux et les exonérations. Le rapport attire l’attention sur le fait que le traitement des pertes en capital, bien qu’aussi important que celui des gains, est négligé dans de nombreux pays.

Dans les pays en voie de développement, la composition des gains en capital n’est pas la même que dans les pays industrialisés; dans ces derniers, la plus grande partie des gains en capital provient de ventes de valeurs mobilières, et dans les premiers de ventes de biens meubles et immeubles et plus particulièrement de biens fonciers. Dans les pays en voie de développement, 1) la fortune est généralement concentrée en biens fonciers, 2) les sociétés à capitaux étrangers prédominent et le marché de leurs actions est international, 3) enfin l’usage d’actions au porteur est généralisé et il est par conséquent difficile de percevoir un impôt sur les gains en capital provenant de titres.

La forte concentration de la fortune et la légèreté relative des impôts fonciers ainsi que les “plus-values non gagnées” dont bénéficient fortuitement des particuliers par suite de la croissance économique de la société soulignent l’équité d’une imposition des gains en capital dans les pays en voie de développement. Cette imposition peut également contribuer à freiner la spéculation, plus particulièrement la spéculation foncière, et à décourager les investissements économiquement improductifs. Les répercussions nuisibles à l’économie que peut avoir cet impót sur l’offre d’épargne et sur la mobilité du capital doivent être soigneusement pesées, en tenant compte du fait que la nouvelle répartition des investissements qu’il provoquera peut être moins bonne que celle qui existait lorsque seul le revenu ordinaire était imposé.

Le rendement fiscal de l’imposition des gains en capital est faible dans les pays en voie de développement et l’administration de l’impôt est par ailleurs complexe et coûteuse. Un pays en voie de développement devra donc évaluer les mérites de cet impôt non seulement en termes de recettes fiscales, mais également en fonction de ses conséquences économiques, sociales et administratives.

La tributación de las ganancias de capital en los países en desarrollo


En este trabajo se examina la experiencia habida en 19 países en desarrollo en cuanto a la tributación de las ganancias de capital y se tratan los principales aspectos del impuesto, incluyendo la definición de activos de capital, el concepto de realización, métodos de devengo, la duración de la tenencia, la estructura impositiva, y las exenciones. Se advierte el hecho de que, aunque el tratamiento de las pérdidas de capital es tan importante como el de las ganancias de capital, en muchos países se prescinde de él.

La composición de las ganancias de capital en los países en desarrollo es diferente de la de los países adelantados; en los segundos, la mayor parte de las ganancias de capital se derivan de las ventas de títulos o valores, mientras que en los primeros se derivan principalmente de las ventas de activos físicos, especialmente de propiedad inmobiliaria. En los países en desarrollo: 1) la riqueza tiende a concentrarse en la propiedad inmobiliaria, 2) predominan las empresas de propiedad extranjera y las acciones por ellas emitidas se negocian internacional-mente, y 3) se halla muy difundido el uso de las acciones al portador, lo que dificulta la exacción de un impuesto sobre las ganancias de capital procedentes de títulos o valores.

La elevada concentración de la riqueza y la relativamente baja tributación de la propiedad inmobiliaria, junto con los “incrementos no ganados”, o sea los aumentos de valor que disfrutan los individuos como resultado puramente fortuito del crecimiento económico de la sociedad, subrayan lo justo que es en los países en desarrollo hacer tributar a las ganancias de capital. Un impuesto sobre las ganancias de capital puede coadyuvar también a coartar la especulación, especialmente en lo referente a la propiedad inmobiliaria, y a desalentar las inversiones que sean económicamente improductivas. Hay que sopesar con cuidado los efectos económicos adversos que un impuesto sobre las ganancias de capital pudiera tener sobre la oferta de ahorro y sobre la movilidad del capital. Para ello hay que tener en cuenta los inconvenientes que puedan producirse, en cuanto a la distribución de la inversión, cuando se grava el ingreso ordinario pero quedan libres de impuestos las ganancias de capital.

La renta obtenida en los países en desarrollo con el impuesto sobre las ganancias de capital es baja, siendo además compleja y costosa la administración del mismo. Un país en desarrollo tendrá, por tanto, que ponderar las virtudes de dicho impuesto no solamente en términos de la renta que produce sino pensando también en sus consecuencias económicas, sociales y administrativas.

In statistical matter (except in the résumés and resúmenes) throughout this issue,

Dots (…) indicate that data are not available;

A dash (—) indicates that the figure is zero or less than half the final digit shown, or that the item does not exist;

A single dot (.) indicates decimals;

A comma (,) separates thousands and millions;

“Billion” means a thousand million;

A short dash (–) is used between years or months (e.g., 1955–58 or January–October) to indicate a total of the years or months inclusive of the beginning and ending years or months;

A stroke (/) is used between years (e.g., 1962/63) to indicate a fiscal year or a crop year;

Components of tables may not add to totals shown because of rounding.

Publications of the International Monetary Fund

International Financial Statistics

This is a monthly publication which presents for most countries of the world current data useful for analyzing international payments and problems of inflation and deflation. Each issue contains about 360 pages. An annual supplement gives comparable figures for prewar and early postwar periods. International Financial Statistics is now available, at the option of the subscribers, in English only, or in English, French, and Spanish. Subscription (for 12 issues and an annual supplement), $10.00. Single copy, $1.50.

Direction of Trade

The great expansion in world trade in recent years has emphasized the importance of current information on the developments and shifts in trade patterns of the industrial and less developed countries. The Direction of Trade (DOT) has been designed to meet this need.

DOT assembles and makes rapidly available detailed trade-by-country information for almost all countries of the world, arranged in a uniform pattern of partner countries and expressed in U.S. dollars. Monthly issues provide the latest data for the current year and comparative data for the preceding year on each country’s trade with all other countries. Totals are shown for the major geographic and economic areas. An annual issue gives detailed country data for several years and includes world totals and a number of summary and analytical tables for major areas. DOT is published jointly by the International Monetary Fund and the International Bank for Reconstruction and Development.

Subscription for 12 monthly issues and the annual issue, $10.00. Single copies of monthly issues, $1.00 each and of the annual issue, $3.00 each.

Balance of Payments Yearbook

This annual series, now issued monthly in looseleaf sections, provides comprehensive information on international transactions for more than 80 countries. Subscription for each volume, $7.50.

University libraries, university faculty members, and university students may subscribe at the reduced rate of $3.00 a year to any one of the Fund’s four subscription publications—Staff Papers, International Financial Statistics, Direction of Trade, and Balance of Payments Yearbook—and to all four at the rate of $10.00 a year.

Subscriptions are quoted in U.S. dollars but are accepted in the currencies of most countries.

Address all inquiries to

The Secretary


19th and H Streets, N.W.

Washington, D.C. 20431, U.S.A.


Mrs. Amatong, a graduate of Silliman University in the Philippines and of Syracuse University, was an economist in the Fund’s Fiscal Affairs Department when this paper was prepared. She is now on the faculty of Andres Bonifacio College in the Philippines.


See United Nations, Technical Assistance Administration, Taxes and Fiscal Policy in Under-Developed Countries (New York, 1954); Joint Tax Program of the Organization of American States, Inter-American Development Bank, and Economic Commission for Latin America (hereinafter referred to as Joint Tax Program, OAS/IDB/ECLA), Fiscal Policy for Economic Growth in Latin America(Baltimore, 1965), pp. 164–65 and 187–96.


Principally Argentina, Bolivia, Burma, Ceylon, Chile, Colombia, Republic of China, Ecuador, El Salvador, Ghana, India, Israel, Malagasy Republic, Mexico, Pakistan, Panama, Peru, Philippines, and Venezuela. Although Ghana repealed its tax in 1967, reference is made to its provisions.


Securities here include corporate stock, U.S. Government obligations, state and local securities, bonds, notes, and debentures. The ratio rises to 70 per cent if all other nonphysical assets are included. U.S. Treasury Department, Internal Revenue Service, Statistics of Income, 1962, Supplemental Report(Washington), September 1966, pp. 21–27.


See, for example, Ifigenia M. de Navarrete, “El Impuesto a las Ganancias de Capital en la Teoría y en la Práctica Fiscal,” El Trimestre Económico (Mexico, D.F.), April-June 1963, p. 210; Ram Niranjan Tripathy, Public Finance in Under-Developed Countries (Calcutta, 1964), p. 142; J.J. Puthucheary, Ownership and Control in the Malayan Economy(Singapore, 1960), pp. 136–37.


Estimates were based on the estate multiplier method as used by Robert J. Lampman, The Share of Top Wealth-Holders in National Wealth, 1922–56(National Bureau of Economic Research, Princeton University Press, 1962), pp. 157 and 192–93.


A good socioeconomic study of savings and wealth in South Asia is found in United Nations Educational, Scientific and Cultural Organization, The Role of Savings and Wealth in Southern Asia and the West, ed. by Richard D. Lambert and Bert F. Hoselitz (Paris, 1963).


M. Habibullah, Pattern of Urban Savings: A Case Study of Dacca City(Bureau of Economic Research, Dacca University, 1964), pp. 25–48 and 140–41.


For a discussion of land tenure in Africa, see African Agrarian Systems, ed. by Daniel Biebuyck (Oxford University Press, 1963), pp. 52–64.


Theodore Geiger and Winifred Armstrong, The Development of African Private Enterprise(National Planning Association, Washington, March 1964), pp. 21–23 and 62–76.


Department of Taxes, Income Tax Statistics(Blantyre, Malawi, February 1967), Table 4.


George E. Lent, “Tax Incentives for Investment in Developing Countries,” Staff Papers, Vol. XIV (1967), p. 301.


Report of the Economic Adviser to the Government of Pakistan, Economic Survey of Pakistan, 1963–64(Ministry of Finance, Rawalpindi, 1964), pp. 44–45.


Developing countries of Africa (excluding South Africa), Asia (excluding Japan), Central America (including Mexico), and South America. See Helmut Giesecke, Industrial Investments in Developing Countries(Hamburg Archives of World Economy, Hamburg, 1963), p. 237.


Giesecke, ibid., p. 240. More recent data on long-term private investments in developing countries by industrial countries show an increasing trend, but classification as to ownership control is not available. See Marcus Diamond, “Trends in the Flow of International Private Capital, 1957–65,” Staff Papers, Vol. XIV (1967), pp. 1–42.


Alexandre Kafka ascribes the presence of bearer shares in developing countries to the lack of confidence in the nondiscriminatory treatment of individual property rights. “The alternative to bearer shares may often be capital flight…. the transferability of registered shares, which is often made difficult by the lack of institutions such as transfer agents or by the sheer problem of communications.” Kafka’s comments on the paper, “Corporate Income Taxation in Latin America,” Joint Tax Program, OAS/IDB/ECLA, Fiscal Policy for Economic Growth in Latin America(cited in footnote 1), p. 258.


For a discussion of the different kinds of capital gains and their effect on taxable capacity, see Nicholas Kaldor, An Expenditure Tax(London, 1955), pp. 41–46.


In the United Kingdom, the Memorandum of Dissent of the Royal Commission on Taxation of Profits and Income argued for the inclusion of capital gains in income taxation on the grounds that “… capital gains increase a person’s taxable capacity by increasing his power to spend or save; and since capital gains are not distributed among the different members of the taxpaying community in fair proportion to their taxable incomes but are concentrated in the hands of property owners (and particularly the owners of equity shares) their exclusion from the scope of taxation constitutes a serious discrimination in tax treatment in favour of a particular class of taxpayer.” Royal Commission on the Taxation of Profits and Income, Final Report, Cmd. 9474 (London, 1955), p. 365.


In Colombia, the OAS/IDB mission reported that the average effective rate of the real property tax for all municipalities would be about $4.00 per $1,000. Joint Tax Program, OAS/IDB, Fiscal Survey of Colombia (Baltimore, 1965), p. 145. The OAS/IDB mission to Panama also recommended an increase in the property tax from 0.75–1.5 per cent to 1–2 per cent, because of the “comparatively light burden borne by the owners of land and buildings.” Joint Tax Program, OAS/IDB, Fiscal Survey of Panama: Problems and Proposals for Reform(Baltimore, 1964), pp. 62 and 79.


In the United States, net capital gains as a percentage of the adjusted gross income (AGI) were 18 per cent of AGI below $10,000; 11 per cent between $10,000 and $50,000; 21 per cent between $50,000 and $100,000; and 29 per cent over $100,000. U.S. Treasury Department, Internal Revenue Service, op. cit., p. 70.


R. N. Bhargava, Public Finance, Its Theory and Working in India(Allahabad, India, 1954), p. 279.


Budget speech of the Minister of Finance, 1965 Budget(Kuala Lumpur, November 1964), p. 47. In 1966 the capital gains tax was abolished for administrative reasons.


For a discussion on the taxation of increments in land value, see George E. Lent, “The Taxation of Land Value,” Staff Papers, Vol. XIV (1967), pp. 101–10.


Dirección de Estadística y Censos, Boletín(Santiago), 1960–64.


Departamento Administrativo Nacional de Estadística, Boletín Mensual de Estadística (Bogotá), October 1966; Revista del Banco de la República(Bogotá), October 1966.


Paper submitted by H. Darin-Drabkin to the Seminar on the Supply, Development and Allocation of Land for Housing and Related Purposes (Economic Commission for Europe, Paris, 1965), p. 170.


Ibid., p. 166.


E. W. Klimowsky, “Capital Gains Taxation in Israel—Modifications,” Bulletin for International Fiscal Documentation, Vol. XVIII (1964), p. 455. See also Lent, ‘The Taxation of Land Value” (cited in footnote 22), p. 109.


Carl W. Cloe, “Capital Gains and the Changing Price Level,” National Tax Journal, Vol. V (1952), pp. 209–10. To use Cloe’s example, if the consumer price index in 1951 (the year of sale) is 185.6 and the corresponding index in 1945 (the year of purchase) is 128.6, the correction factor is 185.6128.6, or 1.44.


Royal Commission, op. cit., p. 366.


Leif Mutén, “Some General Problems Concerning the Capital Gains Tax,” British Tax Review, May-June 1966, p. 141.


The tax on gains from sale of shares in Sweden has recently been included in the tax base. Long-term capital gains on real property (held ten years or more) are recommended for taxation by the Land Value Committee. See Martin Norr and Nils G. Hornhammar, “Extension of Capital Gains Taxation in Sweden,” Canadian Tax Journal, Vol. XIV (1966), pp. 413–21.


There are reserved opinions on this statement pointing to the fact that in developing countries savings behavior, as indicated by family budget studies, is measured by average savings ratios rather than by the marginal propensities to save. It is held that the latter differs less between income classes. See Richard Goode, “Taxation of Saving and Consumption in Underdeveloped Countries,” National Tax Journal, Vol. XIV (1961), pp. 309–10.


See, for example, Lawrence H. Seltzer, The Nature and Tax Treatment of Capital Gains and Losses (National Bureau of Economic Research, Inc., New York, 1951), p. 87, and Henry C. Wallich, “Taxation of Capital Gains in the Light of Recent Economic Developments,” National Tax Journal, Vol. XVIII (1965), pp. 140–45.


Wallich, ibid., p. 144.


Tripathy, op. cit., p. 94.


Martin David, “Economic Effects of the Capital Gains Tax,” The American Economic Review, Vol. LIV (1964), pp. 296–97.


Ragnar Nurske (Problems of Capital Formation in Underdeveloped Countries, Oxford University Press, 1953, pp. 4–31) has expounded the theory that in developing countries diversification of investments is a key factor in increasing the size of the market. The latter is said to be the principal limiting factor in the inducement to invest.


W. W. Rostow (The Stages of Economic Growth, Cambridge University Press, 1960, pp. 46–57) believes that three factors are responsible for take-off in the historical economic development of countries: (1) a supply of loanable funds; (2) a source of entrepreneurship; and (3) the presence of leading sectors in the take-off stage. The last two factors are highly associated with risk taking. Rostow concludes that: “At any period of time it appears to be true even in a mature and growing economy that forward momentum is maintained as the result of rapid expansion in a limited number of primary sectors, whose expansion has significant external economy and other secondary effects.”


See, for example, Jonathan A. Brown, “The Locked-in Problem,” pp. 367–81, and Walter W. Heller, “Investors’ Decisions, Equity, and the Capital Gains Tax,” pp. 381–90, in Joint Committee on the Economic Report, Federal Tax Policy for Economic Growth and Stability(84th Congress, 1st Session, Washington, November 9, 1955).


The approved bill was vetoed by the President on the grounds that it would cause a loss of revenue of no less than P = 50 million annually, or a total of P = 300 million for the duration of the said exemption, without any reasonable assurance of attaining the objectives behind the proposed grant of exemption. (Reported in “President Macapagal’s Veto Message on the Capital Gains Bill,” Philippine Tax Journal, July 1963, pp. 373–74.)


See Nicholas Kaldor, Indian Tax Reform(Indian Ministry of Finance, New Delhi, 1956), p. 32.


Robin Barlow, Harvey E. Brazer, and James N. Morgan, Economic Behavior of the Affluent (The Brookings Institution, Washington, November 1966), pp. 118–20. A similar conclusion was reached by a group at Harvard. Of the 746 investors interviewed, only 5.7 per cent reported that the long-term capital gains tax influenced their investment decision. See J. Keith Butters, Lawrence E. Thompson, and Lynn L. Bollinger, Effects of Taxation: Investments by Individuals(Cambridge, Massachusetts, 1953), p. 339.


For a discussion of the administrative problem involved in the discovery of assets other than real estate, see Noboru Tanabe, “The Taxation of Net Wealth,” Staff Papers, Vol. XIV (1967), pp. 151–52.


The United Kingdom has recently introduced a capital gains tax on long-term gains. See Finance Act, 1965.


The problems and difficulties involved in giving content to the definition of capital assets in the United States are very well described by Stanley S. Surrey, “Definitional Problems in Capital Gains Taxation,” in Tax Revision Compendium, Vol. 2, pp. 1205–15 (U.S. House of Representatives, Committee on Ways and Means, Washington, November 16, 1959).


The calculation for capital gains and losses from the sale of depreciable assets is generally as follows. If the sales price exceeds the book value of the asset, the excess which represents accumulated depreciation is taxable as business income, and the remainder, if any, is taxable as capital gains.


The Economist, November 2, 1963, p. 506, commenting on the tax said: “… This could prove dangerous. The tax may also affect industry’s ability to raise capital through the stock exchange. The Israel stock exchange has only recently begun to play a significant part here; to impose a tax now may clip the bird’s wings before it has left the nest.” Israel introduced a capital gains tax on shares in April 1964 and abolished it in November 1965.


In the United States, charitable gifts (which are deductible as contributions at market value) remain exempt from the capital gains tax, as do transfers at death. The new long-term capital gains tax of the United Kingdom considers free transfers as constructive realizations.


Brown, op. cit., pp. 368–72.


See, for example, Seltzer, op. cit., pp. 289–95; A. J. Merrett, “Capital Gains Taxation: The Accrual Alternative,” Oxford Economic Papers, Vol. XVI (1964), pp. 262–74; Wilbur A. Steger, “The Taxation of Unrealized Capital Gains and Losses: A Statistical Study,” National Tax Journal, Vol. X (1957), pp. 266–81; and Bernard Okun, “The Taxation of Decedents’ Unrealized Capital Gains,” National Tax Journal, Vol. XX (1967), pp. 368–85.


A. R. Ilersic, The Taxation of Capital Gains(London, 1962), p. 46.


Seltzer, op. cit., p. 290.


In Venezuela taxable gains from real estate were formerly reduced by 6 per cent of the original cost of the asset for each year held, to allow for imputed interest on the investment. This practice was criticized by the Shoup mission in 1958 because the 6 per cent assumed income is not taxed under other schedules. In the reform in June 1961 (Decree No. 580) the 6 per cent assumed income added to the cost of the asset was still continued, but the law was recently amended to tax capital gains at income tax rates. See Carl S. Shoup and others, The Fiscal System of Venezuela: A Report(Baltimore, 1959), pp. 163–64.


Israeli rates here are apart from the rates of the Land Betterment Act of 1949. Under the Land Betterment Tax, the progressive rates of 20 per cent to 40 per cent is dependent upon the ratio of the gains to original investment: 20 per cent on appreciation not exceeding 200 per cent of value; 30 per cent on appreciation between 200 per cent and 400 per cent of investment; and 40 per cent if the gain/investment ratio exceeds 400 per cent. In Ceylon the maximum rate is 45 per cent if there is a change of ownership of property occurring on the death of the owner or cessation of residence prior to the year of assessment 1964/65, and for assessments prior to 1965/66.


Costa Rica’s draft of the proposed income tax law is an attempt at using the averaging device to tax long-term capital gains. The proposed tax rate would be the result of dividing the marginal income tax rate by the number of years the asset is held.


Anita Wells, “Legislative History of Treatment of Capital Gains Under the Federal Income Tax, 1913–1948,” National Tax Journal, Vol. II (1949), pp. 12–32.


The percentage-inclusion plan which was in effect from 1934 to 1938 had the following steps:

article image

See U.S. Treasury Department, Federal Income Tax Treatment of Capital Gains and Losses(Washington, 1951), pp. 26–29.


Joint Tax Program, OAS/IDB, Fiscal Survey of Colombia(cited in footnote 18), pp. 27–33.


Joseph Stern (Rapporteur for Israel), in International Fiscal Association, Cahier de Droit Fiscal International(Basle), Vol. XLII, September 1960, pp. 94–95.


Enrique Jorge Reig, “Considérations sur l’imposition des gains de capital,”Revue de Science Financière(Paris, 1962), p. 253.


The Royal Commission, in rejecting the introduction of the capital gains tax in its Final Report, stated: “… no measure for bringing capital gains under taxation could be a simple one. Just because they fit so awkwardly into a scheme of income tax that is steeply progressive, we must envisage a measure that is complicated by a number of exemptions and qualifications; that has to provide different treatment for long-term and short-term gains; and that has to establish machinery for dealing with other problems … such as the relation of the tax to property acquired by gift or inheritance, the valuation of improvements made upon improvable property, the valuation of all realisable property as at the opening date of the tax (unless it is to bring into tax increases of value accrued before the date of its imposition). No doubt every tax seems complicated in advance, before the officials and the public have become habituated to its existence. But for all that it would be unrealistic to suppose that it would not require a substantial addition of staff to administer it… Royal Commission, op. cit., pp. 33–34.


Nicholas Kaldor. Indian Tax Reform(cited in footnote 41), p. 34.


Nicholas Kaldor, a proponent of capital gains taxation in developing countries, recognizes that valuation is a real problem in the administration of the capital gains tax. In explaining the difference between the valuation problem of a capital gains tax and that of a net wealth tax, he says: “The uncertainties and inaccuracies involved in the valuation of capital assets present a serious problem also in the case of a tax assessed on the value of assets. But they weigh far more heavily in the case of a tax assessed on the change in values over a period. In one case errors of valuation have a more or less corresponding effect in the calculation of the tax liability; in the second case they tend to have an altogether disproportionate effect.” An Expenditure Tax(cited in footnote 16), p. 38.


Bhargava, op. cit., p. 281.


Sweden in 1966 used a “standard method” of calculating the long-term capital gains tax on shares. Taxpayers with long-term gains (on assets held more than five years) include in the taxable income 10 per cent of the sales proceeds. Capital gains tax is not paid if it can be shown that the gain is not higher than 5 per cent of the sales price. A standard annual deduction of SKr 500 is also allowed. Thus, with a sales price of SKr 15,000 the tax is computed on the following basis:

article image

Net long-term capital losses are not deductible because they are already taken into account in fixing the taxable portion of a long-term gain at 10 per cent under the standard method. Norr and Hornhammar, op. cit., pp. 417–18.


For a summary list of the tax treatment of residents and nonresidents in developing countries, see Oliver Oldman, “Tax Policies of Less Developed Countries with Respect to Foreign Income and Income of Foreigners,” Taxation of Foreign Income By United States and Other Countries(Tax Institute of America, Princeton, 1966), pp. 80–83.


Harvard Law School, International Program in Taxation, World Tax Series: Taxation in India(Boston, 1960), pp. 290, 296, and 300–301.


The same provision was made for the taxation of capital gains in the treaty between India and Denmark that was signed on September 16, 1959.


Organization for Economic Cooperation and Development, Fiscal Committee, Draft Double Taxation Convention on Income and Capital(Paris, 1963).

IMF Staff papers: Volume 15 No. 2
Author: International Monetary Fund. Research Dept.