Measurements of taxable income must reflect changes in the value of assets and liabilities. If deductions are taken too quickly or if the recognition of income is unduly postponed, the present value of tax liability is reduced. Tax shelters are based on the creation of artificial accounting losses that result from acceleration of deductions and the deferral of recognition of income; such losses arise from partnership investments and are used to offset income from other sources. Depreciation is the most obvious and most important timing issue, but it is not unique. Industries in which timing issues (and therefore the possibility of tax shelters) are especially important include oil and gas, timber, orchards and vineyards, and real estate. The timing rules that are required for preventing the mismeasurement of income can add considerable complexity to the tax system.

The tax systems of most countries are based on the implicit assumption that prices are stable. If, instead, there is inflation, real (inflation-adjusted) income is mismeasured, and distortions and inequities occur. For example, tax is paid on (or deductions are allowed for) the full amount of interest earned (or paid), even though inflation is eroding the principal. (Part of interest can be seen as merely offsetting this erosion; it is neither income nor an expense.) Tax is also paid on capital gains, with no allowance for inflation; thus, fictitious gains are taxed, and a tax may even be levied when no real gain has occurred. Finally, business is not allowed to recover tax-free its investment in depreciable (and similar) assets and inventories.

Although many less-developed countries that have experienced high rates of inflation provide for inflation adjustment in the measurement of income, no industrialized country does so. As long as inflation is expected to be low, the benefits of inflation adjustment are generally thought not to be great enough to justify the increased complexity that would be involved.

It has been argued that one way to avoid the complexities of both timing issues and inflation adjustment is to switch from a tax system based on income to one based on consumption. Under such a system all business purchases would be deducted immediately, or “expensed.” Borrowing in excess of investment would be added to income, and lending would be subtracted; the resulting tax base would be consumption. Through the tax saving resulting from expensing, the government, in effect, becomes a partner in all investments; the revenues it subsequently receives are best seen as the return on its investment. A consumption-based tax imposes no burden on income from marginal investments, because the private investor keeps all of the income relating to his share of the investment. As a result, such a tax does not favour present consumption over saving for future consumption, as the income tax does. Advocates of consumption-based taxation believe that simplicity—the lack of timing issues and the fact that inflation would have no chance to distort the measurement of consumption—may be even more important than the economic advantages they envision from such a tax.

Some economists view the flat tax as an alternative that is even simpler than consumption-based taxation but would achieve similar economic effects. It works by exempting most capital income from taxation at the individual level; that is, only labour income is taxed. This proposal, like consumption-based taxation, suffers from the loss of progressivity that results when the tax on most capital income is eliminated. No country uses either of these consumption-based direct taxes, but Croatia has employed a system that has similar effects.

Major corporations operate across state and national boundaries. Because most jurisdictions tax income that is earned within their boundaries, it is necessary to determine the source of income of a multijurisdictional entity. The states of the United States follow a practice that is quite distinct from that in the international sphere. National governments commonly resort to the convention of “arms-length” prices—the prices that would prevail in trade between unrelated entities—to determine the split of income resulting from transactions between related parties. The states, by comparison, employ formulas to divide the income of a multistate corporation or a group of related corporations engaged in a “unitary business” between in-state and out-of-state income. Neither of these approaches is totally satisfactory.

Some countries (including the United States) exercise the right to tax the whole income of their nationals, even if it is earned abroad. Almost all countries consider it their right to tax income arising within their borders, whether or not the income is earned by individuals or corporations having their residence or exercising their management and control in the country. Increasing attention has therefore been given to the prevention of double taxation between countries, especially in response to the continuing rise in the number of corporations operating in more than one country and the number of stockholders of a corporation residing outside the country in which it operates.

To illustrate how double taxation may come about, consider a corporation A that has its headquarters in country X and a manufacturing plant in country Y. Country X may tax the profits earned in Y and so may Y. Further complications may arise if some of the shareholders of A live in country Z and are subject to income tax there on dividends received from A, which may also be subject to a withholding tax in X. Relief from double taxation can be provided unilaterally or by treaty. Country X may allow corporation A a foreign tax credit for income tax paid in Y; this is done by, for example, the United States, the United Kingdom, Canada, and Germany. Alternatively, country X might unilaterally give up its right to tax certain profits earned abroad; this approach is followed by, for example, France and the Netherlands. Countries X and Z might enter into a tax treaty relieving dividends paid by corporations in X to shareholders residing in Z from withholding tax and providing some compensating advantages for X. A network of tax treaties exists among the industrialized countries, but they apply only sketchily to the less-developed countries. There are doubts as to whether the standard provisions found in agreements between rich countries are suitable for agreements between industrialized countries and those at earlier stages of economic development.

The varying national tax policies can also be used to avoid paying taxes. Many developed countries do not actually tax the majority of investment income (especially interest) that originates within their borders and flows to foreigners. They may thus attract capital from less-developed countries that either do not or cannot tax such income when it is received by their residents, but this worsens problems of capital shortages. Investment and the related income sometimes are channeled through “tax haven” countries in order to take advantage of tax treaties. To illustrate how this approach can be used to avoid taxes, consider the case of a resident of country R who wishes to invest in country I, with which country R has no tax treaty. If the funds flow through country T, which has a treaty with I, and if income is not reported to R, tax due to I, as well as tax due to R, can be avoided. (It might more properly be said that this involves illegal evasion rather than legal avoidance.)

The rise of e-commerce (the electronic sale of goods and services over the Internet) has posed new questions of tax policy and administration. E-commerce makes it easier for business to be conducted in a country without creating a “permanent establishment,” which would subject the seller to income taxes. It blurs the distinctions between the sale of goods, the provision of services, and the licensing of intangible assets, each of which is subject to taxation. Equally problematic is a reliance on arms-length methods of income measurement. Tax codes continue to be revised as governments determine reasonable approaches to the taxation of electronic transactions.

Henry C. Simons, Personal Income Taxation (1938, reprinted 1980), is a classic discussion of the concepts of personal income as a basis for taxation; William Vickrey, Agenda for Progressive Taxation (1947, reprinted 1972), provides an excellent analysis of many of the problems of income taxation; B.E.V. Sabine, A History of Income Tax (1966), surveys the policies in Britain and the United Kingdom from 1799 to the middle of the 20th century; and Edwin R.A. Seligman, The Income Tax: A Study of the History, Theory, and Practice of Income Taxation at Home and Abroad, 2nd ed., rev. and enlarged (1914, reprinted 1970), includes in its coverage the United States and its pre-World War I policies. Other useful studies include Richard Goode, The Individual Income Tax, rev. ed. (1976); Joseph A. Pechman (ed.), Comprehensive Income Taxation (1977); David F. Bradford, Untangling the Income Tax (1986); Stanley S. Surrey, Pathways to Tax Reform: The Concept of Tax Expenditures (1973); Harry Kalvin, Jr., and Walter J. Blum, The Uneasy Case for Progressive Taxation (1952, reprinted 1978); and Stanley S. Surrey and Paul R. McDaniel, Tax Expenditures (1985).

Useful studies include Richard Goode, The Corporation Income Tax (1951); Robert M. Willan, Income Taxes: Concise History and Primer (1994); Mervyn A. King, Public Policy and the Corporation (1977); J. Gregory Ballentine, Equity, Efficiency, and the U.S. Corporation Income Tax (1980); Charles E. McLure, Jr., Must Corporate Income Be Taxed Twice? (1979); and Charles Adams, For Good and Evil: The Impact of Taxes on the Course of Civilization (1999).