Journal Issue
Finance & Development, March 1980

Limits to taxation: The growing ratio of tax revenue to real output in many countries emphasizes the need to recognize tax limits

International Monetary Fund. External Relations Dept.
Published Date:
March 1980
  • ShareShare
Show Summary Details

Richard Goode

Tax limits are economic in the sense that exceeding them will gravely harm production and distribution. They are also political in that governments recognize the bad economic consequences of excessive taxation and the resistance to it and try to keep taxes within the limits. An absolute limit would prevail if no additional revenue could be obtained by raising tax rates or imposing new taxes. An absolute limit, however, is rarely encountered. Usually a tax limit is not fixed at any precise level but reflects a judgment that the harm done by raising taxes beyond some point would outweigh the advantages of obtaining greater revenue to finance public expenditures or to reduce the rate of domestic inflation.

Tax limits may operate because excessive taxation impairs productive capacity, weakens economic incentives, arouses resistance and evasion, or imposes insupportable administrative burdens.

Taxation can reduce productive capacity by destroying capital or by reducing the consumption needed to maintain the energy and health of workers. Whether these are serious threats depends in good part on how the tax revenue is used. If the government spends for well-chosen investments, necessary maintenance of useful facilities, or for health services and education, the risk that productive capacity will be harmed is much less than if the revenue is used to finance ill-chosen prestige projects or to support an idle bureaucracy.

By reducing economic rewards, high taxes can make people less willing to work, assume managerial responsibility, control costs, or invest. If the tax revenue is then used for welfare purposes that may be applauded for social reasons, the effect can be to accentuate the adverse effects on incentives. If, for example, the government subsidizes or provides free goods and services that people would otherwise buy for themselves, such as food, housing, and health care, workers will feel less need to earn income to pay for these. While the community as a whole may benefit from social expenditures, the amount of benefit received by an individual or family is not related to the amount of tax paid by that person or family. Most people feel that taxation is a burden regardless of how much they may approve the purposes for which the proceeds are used. Few citizens are as farsighted as the great American judge, Oliver Wendell Holmes, who said: “I like to pay taxes. With them I buy civilization.”

Taxation, in fact, is limited by the resistance that may be aroused when taxes rise rapidly and are regarded as excessive. Resistance may take the form of increased evasion—particularly of direct taxes on income, profits, and wealth, which depend to a great extent on the cooperation and voluntary compliance of taxpayers. Evaders include not only those who conceal profits by involved commercial and financial transactions or the falsification of accounts, but also workers in the “black,” or parallel, economy.

Tax evasion, however, is a complex phenomenon that may be linked more closely with traditions and attitudes toward laws and government than with tax rates. Equally or more important may be tax avoidance that does not break the law but diverts activities and transactions from commercial or organized markets to channels that are less exposed to taxation. In less developed countries, farmers may consume more of their output or use it for barter in the villages and offer less for sale in the cities or for export. In industrialized countries, high tax rates may induce householders to divert time and energy from earning taxable income to maintaining and improving their houses and gardens and other household activities. There is an economic loss, since the advantages of specialization are sacrificed and economic efficiency lessened, but this loss is smaller than the loss of the monetary value of income and consumption that can be taxed.

Another form of resistance to taxation shows itself in inflation. Although the failure of governments to impose sufficient taxation in relation to expenditures is a much more common cause of inflation, high taxes themselves can aggravate inflation. That occurs when workers obtain wage increases to compensate them for higher taxes. It was first evident with respect to indirect taxes, which entered into market prices and provoked demands for offsetting wage increases. Sometimes wages were formally linked with the cost-of-living index, and the adjustments were more or less automatic. Recently, demands for compensation for higher income taxes have been recorded. Governments have had only limited success in combating wage retaliation against taxation. To be sure, wage demands could be discouraged and their inflationary effects largely neutralized by firm monetary policy. But governments are reluctant to face the strife, disruption, and unemployment that are likely when strong and insistent wage demands are met by unyielding monetary restraint.

It is important to take into account the time factor when considering tax limits. The tax level could temporarily exceed the safe long-term limit because the harmful effects of a tax on the capital stock and on readiness to work and productivity occur only over a period of years. People may also be willing to accept on a temporary basis unusually high taxes during a war or other emergency. On the other hand, economic incentives and voluntary compliance may be less damaged by a slow upward creep of taxation than by a sudden jump. Time is required to strengthen administrative capacity. These considerations suggest that the rate of change of taxation, as well as the level, determines the limit; it appears that tax limits tend to increase over time in most countries, though at varying rates.

Factors affecting tax limits

A common opinion is that the greater the per capita national income, the higher the tolerable limit of the ratio of taxation to income. It is true that the high-income countries as a group have higher tax ratios than the low-income countries as a group. But the relationship between per capita income and the tax ratio is by no means close and uniform. Among industrial countries, per capita income accounts for only a minor part of the variability of tax ratios. For less developed countries, repeated studies carried out in the Fiscal Affairs Department of the International Monetary Fund over the past 12 years have found that per capita income is of doubtful significance in statistical analyses intended to “explain” tax ratios.

Certain kinds of activities and natural resources are relatively easy to tax because they are visible, are concentrated in particular locations, or are immobile. They offer what have been aptly called “tax handles.” Foreign trade is an important example. It is more easily taxed than domestic trade because it must cross frontiers that are subject to control and because of the long-established practice of imposing special taxes on imports and exports. Mineral resources are another example. They are visible, concentrated in location, and immobile; and in many countries, they historically belonged to the sovereign. Furthermore, in less developed countries, minerals are usually exploited by foreign-owned corporations. Agriculture, on the other hand, is hard to tax because of the importance of home consumption of output, the feasibility of barter, and, in many areas, the small scale of operations. Statistical studies confirm that tax ratios in developing countries tend to be positively related to the shares of foreign trade and mineral production in gross national product (GNP) and negatively related to the share of agriculture.

Tax rates in other countries influence the tax limit by affecting opinions about what is bearable. The possibility that domestic capital will flow abroad to countries with lower taxes and that foreign capital will be discouraged from entering a high-tax country has restrained the taxation of profits, investment income, and financial wealth.


It has been argued that the tax limit is lower for progressive taxes, which take an increasing fraction of income as it rises, than for regressive taxes, which rise less rapidly than income. High marginal rates on increments to income, which are characteristic of steeply progressive taxation, are particularly harmful to economic incentives. Progressive taxation reduces the extra economic rewards for hard work, risk-taking, and innovation. Also, progressive taxes fall more heavily than other taxes on persons who save an above-average fraction of their income. Hence, progressivity may restrict capital accumulation. This latter point was stressed in early discussions of tax limits. Recently it has excited less concern, perhaps because other aspects of fiscal policy, particularly the size of the budget deficit, are now seen to influence total national saving much more than the degree of tax progressivity.

If progressivity holds down the tax limit, a conflict may arise between revenue requirements and tax reform. This is so because reform proposals often call for greater reliance on direct taxes and a more progressive distribution of the tax burden.

The facts, however, do not support the belief that progressivity is incompatible with a high tax ratio. To verify this point, it is advisable to examine separately revenue statistics for industrial countries and less developed countries, as the many differences between the two groups preclude a valid comparison between them. The leading industrial countries are members of the Organization for Economic Cooperation and Development (OECD). Among these countries a large part of the difference between the tax ratios of the most highly-taxed countries and the other countries is due to the greater use of generally progressive income and profits taxes by the former countries. In the high-tax countries, revenue from income and profits taxes represents a greater percentage of gross domestic product, and it also accounts for a greater percentage of total tax revenue. This analysis is based on a classification of 23 OECD countries, according to their overall tax ratios, as follows: 5 countries with the highest ratios, 13 countries with intermediate ratios, and 5 countries with the lowest ratios (see Table 1).

In less developed countries, tax ratios are generally lower than in the OECD countries, but, as is true in the OECD, the countries with relatively high tax ratios probably have more progressive—or less regressive—systems than those with lower ratios. The countries with relatively high ratios rely much more heavily on income and profits taxes, as can be seen from Table 2, which covers a sample of 62 less developed countries. Indirect taxes on goods and services, which are the major revenue source in the less developed countries, may be less regressive on the whole in these countries than in the OECD countries because many of the poorest people in the less developed countries either are engaged in subsistence agriculture or spend a large part of their income on food and other articles that are not taxed.

A partial explanation of the association of high tax ratios with the use of progressive taxes is that the political attractions of progressivity have outweighed its economic disadvantages. Especially in democracies, the use of direct taxes and steeply graduated rates on the affluent few may be necessary to win acceptance of heavy taxes on the masses. The top rates may have far more significance for revenue and the overall tax ratio than their direct yield indicates.

Although the facts disclosed in Tables 1 and 2 are encouraging to tax reformers, they do not justify disregarding the dangers of excessive tax progressivity. Exorbitant marginal rates are likely to cause economic distortions and stagnation and to stimulate tax avoidance and evasion. Some critics contend that a number of industrial countries are already experiencing these ills.

Table 1OECD countries: tax ratios and composition of tax revenue, 1972-76(Unweighted arithmetic means, in per cent)
Overall tax ratio1
Highest 5Middle 13Lowest 5
Ratio of tax revenue to GDP
All taxes44.4633.5722.15
Income and profits taxes19.9913.106.04
Social security contributions9.297.256.12
Property taxes1.602.311.56
Taxes on goods and services12.9510.307.79
Other taxes0.630.620.64
Composition of tax revenue
All taxes100.00100.00100.00
Income and profits taxes44.9339.6027.57
Social security contributions20.9521.0227.96
Property taxes3.677.167.05
Taxes on goods and services29.0630.4834.45
Other taxes1.381.742.97
Source: Organization for Economic Cooperation and Development, Revenue Statistics of OECD Member Countries, 1965-76 (Pans, 1978).
Table 2Less developed countries: tax ratios and composition of tax revenue, 1972-76(Unweighted arithmetic means, in per cent)
Overall tax ratio1
Highest 15Middle 32Lowest 15
Ratio of tax revenue to GNP
All taxes25.7015.398.79
Income and profits taxes9.623.761.95
Social security contributions20.870.500.14
Property taxes0.330.240.22
Taxes on goods and services13.729.905.76
Other taxes1.150.990.73
Composition of tax revenue3
All taxes100.00100.00100.00
Income and profits taxes37.4324.4322.18
Social security contributions23.393.251.59
Property taxes1.281.562.50
Taxes on goods and services53.3964.3365.53
Other taxes4.476.438.30
Source: Alan A. Tart, Wilfrid L. M. Grátz, and Barry J. Eichengreen, “International Comparisons of Taxation for Selected Developing Countries, 1972-76,” International Monetary Fund, Staff Papers, Vol. 26 (March 1979), pp. 155-56.

Innovations and administration

Readiness to accept tax innovations, if not carried to extremes, can help push up the tax limit. Several important innovations have helped to increase the tax limit since the end of World War II. Among these were the development of broad-based sales taxes, particularly value-added taxes in Europe and value-added taxes and manufacturers sales taxes in less developed countries; the reduction of the threshold of income tax coverage made possible by withholding tax at the source and by modern data processing systems; the shift of payment of nonwithheld income and profits taxes to a more current basis (especially important in inflationary periods); and the conversion of many excises and customs duties from specific to ad valorem bases. None of these innovations represented a wholly new idea or practice; in taxation, as in industry, there is often a long lag between discoveries and their general use.

The adaptation of the tax system to a country’s administrative capabilities and to the willingness of taxpayers to comply can increase the revenue potential by minimizing tax evasion and the harm done by unequal application of taxes. The recognition of these constraints may inhibit innovation and may appear also to conflict with some of the objectives of tax reform. It is highly important, nevertheless, to scrutinize critically tax reform proposals that call for the introduction of measures that may go beyond the realistically evaluated administrative and compliance capacity. In-complete and unequal enforcement of sophisticated tax measures can cause them to lack in practice the advantages that are attributed to them in textbooks and treatises.

In conclusion, the tax limit and factors affecting it are highly important for policy making. A country that has exceeded the safe limit of taxation must cut government expenditures, unpleasant as that may be. A country that has not reached its tax limit has more options. The government can continue to hold taxation below the limit, or it can draw on the unused taxable capacity to raise more revenue and thus reduce a budget deficit or finance larger public expenditures. F&D

Other Resources Citing This Publication