A major effort has been made in the past to collect taxes at the source, through withholding. In addition to withholding on salaries and NII, which is common in other countries, withholding is also imposed on payments for goods and services supplied by one business to another, as well as on dividends and other distributions.

All businesses are required to file separate monthly returns, with payment, for VAT, for withholding of income tax on salaries and other payments, for withholding of NII, and for estimated income tax. All businesses also file yearly profit and loss statements and balance sheets, whose complexity depends on the level of bookkeeping required. Individuals with significant non-wage income are required to file individual income tax returns. Many individuals, especially those with business income, are required to file extremely detailed statements of personal and family net worth every few years, as a check on the veracity of their yearly returns.

3. The Present System - Some Problems

   Inflation

   Since the Israeli economy seems to be stuck, for the foreseeable future, in an inflationary framework, the problem of tax collection in such an environment must be dealt with simply. The system currently in use (the "Steinberg Law" additions to the company tax) involves a consistent, systematic (and complicated) adjustment for inflation to the profit/loss statement and to the balance sheet, with sometimes surprising results. For example, a company which is completely inactive in a given year but had a negative net worth at the end of the previous year, will find that the inflationary adjustment of the negative net worth generates an item of income and a tax obligation in the current year. 2

   Although understandable from the accounting point of view, this kind of result destroys the direct connection between income and income tax, as far as the average businessman is concerned. The system we propose is simple and transparent in its working.

   Incentives to Industry

   The present system of incentives to industry consists of ruinous tax rates together with benefits and subsidies to those who fulfill certain requirements set by the government. Some of these requirements are general, such as location in certain areas, and some are specifically set up on a project by project basis by the bureaucracy that approves these projects. Those who receive benefits enjoy a great advantage over those who do not, an advantage that can lead to monopoly status. There is no reason to suppose that government functionaries are better equipped to judge the viability of industrial projects than those whose money is on the line. Incentives should be built into the system, rather than given out project by project.

   Evasion

   Although difficult to measure, the rate of tax evasion in Israel is generally acknowledged to be relatively high. Parenthetically, it is worth noting that the Internal Revenue Service in the United States estimated several years ago that 40 percent of self-employed income is unreported. 3 In Israel the United States is regarded as a nation of honest taxpayers. At issue here is evasion of taxes on income from legitimate business activities because of high marginal rates, the crucial factor in the psychology of tax evasion. Yet tax evasion in Israel has other more serious effects than mere loss of government revenues. We should remember that given the easy travel and communications of today, transfer of income and assets abroad is a simple matter.

   Anyone even marginally active in export and import can routinely transfer assets abroad. And even that unreported income which is not transferred abroad is generally held in foreign currency, due to the understandable reluctance of Israelis to hold cash in shekels. Tax evasion resulting from high marginal rates thus leads not only to loss of tax revenue, but also, and at least as importantly, to loss of foreign exchange. Furthermore, assets based on unreported income, whether kept here or abroad, are unproductive. In India families keep their savings in the form of gold and silver jewelry, which is devastating for the economy. Similarly, dollars kept in mattresses or under floor tiles do not recirculate in the economy and are unavailable for growth and development. Tax evasion due to high marginal rates thus leads to loss of tax revenue, loss of foreign exchange, and the stifling of development.

   Regressivity

   A tax is regressive if it consumes a higher percentage of lower incomes than of higher incomes. A head, or per capita tax (sometimes called a poll tax) is a good example, since the flat, fixed amount constitutes a higher percentage of low incomes than of high incomes. This kind of tax is usually introduced not in order to punish the poor, but because it is easy to collect, requiring only identification of the taxpayer and no income information. Regressive taxes are considered undesirable and even immoral in our time. But even taxes that are not obviously regressive may, on closer examination, turn out to be so, if, for example, they are charged only up to a fixed ceiling, or if they are charged on types of income that are more prevalent among low income groups, such as earned income as opposed to unearned income.

   It is a surprising fact that in socialist Israel there are significant regressive taxes. These ought to be eliminated. A trivial example is that of the television fee, a flat fee of about $100 per year levied on every owner of a television set. Assuming that television broadcasting should be a government function, then its cost of about $100 million should be budgeted and paid for out of tax revenues. There is no reason to assume that those with low incomes derive so much more enjoyment from TV that they should be forced to pay a higher percentage of their income for it than the wealthy.

   A much more serious case is that of NII. The notion that NII is some kind of self-insurance scheme, whereby people pay during their working years to insure their own income after retirement was probably never true, and is certainly not true today. The reality is that those who are working pay for the pensions and other benefits of those who are retired, disabled, or for whatever reason not working. The imposition of means tests clarifies the essential character of NII as a tax-funded way of redistributing income. If so, then it should be budgeted as any other government expense and paid for from general tax receipts. The present system of collecting NII tax, with its ceiling and its application, in most cases, to earned income and salaries alone (excluding fringe benefits, often the major part of larger salaries), is regressive and inequitable.

   Progressivity

   Progressivity, meaning that high income groups pay a higher percentage of their income in taxes than low income groups, is generally commended, since it contributes to the socially desirable aim of redistributing income. But the question of progressivity in tax rates is subsumed under the larger issue of whether social ends can be achieved through the tax structure, as opposed to achieving them through the expenditure side of the budget.

   Since one of the purposes of the national budget is to achieve social ends (health, education, general redistribution of income, etc.), there is nothing intrinsically wrong with trying to attain them, at least in part, through the tax structure. The only question is whether this method works. There is sufficient evidence to show that it is not very effective. Those in higher income brackets, who should be paying high marginal rates, often pay those rates on only a small fraction of their income. The average rates they pay on their total income are often lower than the average rates paid by middle income salaried taxpayers. At the other end of the scale, the low tax rates paid by low wage earners are deceptive, since we usually forget to include the 15 percent NII, which is assessed only up to a fixed ceiling.

   There is now evidence, very well documented by Rabushka and Hanke, that lowering the extreme marginal tax rates leads, surprisingly, to a higher percentage of the tax burden being borne by the wealthy. 4 As noted in their report, the best way to make the tax system more progressive is to reduce the marginal rates, since then the wealthy climb out of their tax shelters and join the taxpaying public. Section V contains data from the United States showing changes in the amount of taxes paid by various income groups as a result of the tax reforms there in the 1980s. Social ends such as income redistribution are more effectively achieved explicitly through the expenditure side rather than through progressive tax rates that exist only on paper.

   Costs of the system

   We have no detailed estimates regarding the material costs of compliance and enforcement in Israel. But for the United States, Hall and Rabushka estimate the total costs at about 10 percent of the taxes collected. Furthermore, they estimate the amount of tax evasion at about 35 percent of taxes actually collected. 5 They estimate the loss to the economy, in the form of lost output, resulting from the pursuit of tax-advantaged investments, at 30 percent of tax receipts. We have no reason to think that the situation is much better here in Israel.

   In addition to the material costs, there are human costs that although not quantifiable are real enough and, in our opinion, at least as significant as the material costs. The integration of computerized data banks into the enforcement apparatus has provided the tax authorities with a level of knowledge of individuals' activities that rivals that of the security services. The tax authorities employ many of the same tools, such as snooping, informers, entrapment, detention before trial and even before indictment, and so on. But whereas civil rights groups actively oppose the invasions of privacy and the general "Big Brother" aspect of the security services, there is no similar awareness of the same aspect of the tax authorities' operations.

   This is curious, since, after all, security is more important than tax collection, which is only a matter of money. The detailed declaration of net worth which is required of taxpayers who have significant non-salary income (a goodly fraction of the population) is essentially a statement by the government that their annual tax returns are a priori suspect and need corroboration. One commonly encounters people in Israel who simply avoid productive economic activity because of the contact, both direct and indirect (in the form of surveillance), with the tax authorities that this would entail. Not the least of the advantages of the tax system we propose is that it would enable a drastic reduction in the policing activities of the tax authorities and a consequent improvement in the atmosphere.

Part III