Who pays the highest income tax?

Focusing on “headline” rates of tax can easily generate misleading conclusions about how much marginal income tax people pay and about the effect taxes can have on earner behaviour. In the second of their two-part series on income taxes, Flid de Kam and Chiara Bronchi explain why.

It is not always high earners who pay the highest marginal rates of taxes on income. This assertion may appear to contradict what one would expect of progressive tax systems. Yet, in most OECD countries many individuals in low- to middle-income brackets find themselves exposed to higher marginal rates -- that is the rate applied to the last additional dollar, yen or franc earned -- than even the very rich. The question is why?

Part of the answer lies in "bubbles", which are humps in the structure of taxes on income. Bubbles can develop in cases where income is subject to both personal income tax and social security contributions. The tax base of those contributions may be identical or similar to that used for personal income tax. But unlike for income tax, a ceiling or cap often applies; earnings above that ceiling are not subject to social security contributions. A bubble appears if the combined marginal rate of income tax and "capped" social security contributions exceeds the marginal income tax rate applicable to income earned above that contributions ceiling.

For example, take a country that imposes social security contributions at a flat rate of 15% on the first 50,000 units of income. Also, suppose the first 25,000 units earned are subject to 10% personal income tax, the second 25,000 units is taxed at 20%, and all income over 50,000 is taxed at the top rate of 30%. To judge by the headline rate alone, the latter rate of 30% would seem like the highest of the lot. But in practice it is those with taxable income in the middle bracket who pay the highest marginal rate, since the marginal income tax and social security add up to 35% of their additional earnings. But taxpayers in the highest bracket are not required to pay the 15% social security contribution and so only pay 30% on their highest earnings.

But bubbles do not just show up in "all-in" rates of the combined taxes on income. Occasionally, they appear in standard personal income tax schedules as well. In the second half of the 1980s the US federal income tax had such a rate structure. At the time, income in the first bracket was taxed at 15% and income in the top bracket at 28%. It follows that tax relief for high-income earners, which is determined by the marginal tax rate, was almost twice the tax relief for low-income earners. To recoup the higher tax relief for well-off taxpayers, lawmakers introduced a new 33% bracket which they sandwiched between the low and high brackets. The new middle rate worked like this. Suppose for the sake of illustration that the personal exemption on income tax was $4,000 across the board.

Under the old structure before the 33% band was created the tax bill of low-income earners would have been reduced by $600, since with the exemption they would not have had to pay the 15% tax on that $4,000. The tax bill for those in the highest taxed bracket would have been slashed by $1,120, because they would have been exempt from paying 28% on their highest $4,000 of income. The difference of $520 in favour of higher earners was clawed back by inserting a middle bracket of $10,400 taxed at 33%, that means 5% more tax to pay than before, or $520. So, although the tax relief for the highest earners remained at 28%, or $1,120, by taxing middle earnings more, the new bracket effectively balanced the tax relief for those in the low and the top brackets at $600. The rates of the federal income tax in Switzerland show a similar "bubble" today.

A job can make you poorer

Another rather curious situation which does not show up when studying headline rates is that low earners can find themselves confronted with very high marginal tax rates, in some rare cases exceeding 100%. The reason for this is that lower earners not only pay more tax when their income goes up, but in many cases they lose part of their means-tested tax relief, subsidies and benefits as well. The loss of this income acts as an "implicit" tax at the margin. The rational response of workers who find themselves in this situation is to reduce the number of hours they work. Their gross wage would of course be lower if they did, but in return they would pay less tax and receive more means-tested subsidies and benefits. As a result, their net disposable income would increase despite putting in fewer hours.

This type of situation occurs to varying degrees in different OECD countries, depending on the peculiarities of various social protection programmes. Take the example of an unemployed couple with two young children. Suppose that after five years' unemployment, one of them takes up a lowly paid job. In Finland or Sweden net income in and out of work would be the same in that case, since each unit of income earned is cancelled out by a unit of benefits foregone once employment is taken up. In other words, there is an implicit tax rate of 100%. In the case of Denmark and the Czech Republic, the implicit rate in a similar case would be almost 100%, and in Germany and the United Kingdom it would be around 80%.

In France and the United States the implicit rate would be about 50%, since half the increase in earnings is wiped out by a loss of benefits. In Japan, the implicit tax actually exceeds 140%, meaning our one-earner couple would be worse off with the new job than without it. What's more, they may have to be wary when it comes to staying in the job itself, since small wage increases can expose low-wage earners to high implicit tax rates as their means-tested benefits get cut further.

As the next chart shows, the "all-in" top rates of the combined taxes on personal income may also vary by type of income. Labour income is more heavily taxed if it is subject to contributions earmarked to finance employee social insurance. Rates of personal income tax proper may also differ, depending on the type of earnings. For example, capital income, which is often an important revenue component of the well-off, is often not subject to most social security contributions. In addition, over the past fifteen years a number of OECD countries have introduced low, flat rates for certain types of capital income, notably interest and dividend earnings. This is the case for Belgium, the Czech Republic, Greece, Hungary, Italy, Poland and the Nordic countries. Flat rates on capital income can reduce the overall progressiveness of the income tax and undermine its redistributive effect. Moreover, any examination of the highest statutory income tax rates that does not also take these low, flat rates on capital income into account tends to overstate the tax burden of high earners.

For larger graph click here

Why were these flat rates introduced? One explanation is that they were a response to growing pressures from cross-border tax competition. Financial capital, being highly mobile, tends to flow to those jurisdictions where it is taxed at the lowest rates. To address capital flight, tax policy-makers may decide to reduce the domestic tax burden on capital income. The moves were also part of a more general strategy designed to lower the efficiency costs of taxation by reducing rates, while at the same time spreading taxes more widely along the capital income tax base.

Planning privilege

Another point to remember when looking at "statutory" tax rates, which are the tax rates as set by law, is that many high-income earners actually escape paying them. There is of course no automatic reason to assume that high-income groups are more inclined to evade taxes than are low to middle income groups simply because they are better off. Nevertheless, it is true that the self-employed, who make up a significant proportion of high-income earners, are generally in a better position than other groups of taxpayers, particularly those who are taxed at source, to limit their tax obligations. This they can do legally, by using tax breaks for business, and illegally, by under-reporting their income (see Spotlight on Taxation in Observer 215, January 1999).

Tax planning is often used to take the bite out of high rates. If properly done, high-income earners may see their taxable income shrink, their tax bills fall and their disposable income rise. For example, in some countries the corporation income tax rate is substantially lower than the top income tax rate, providing the self-employed with a strong incentive to cloak their business in a corporate veil while paying themselves only a small director's salary. In the Netherlands the gap between the corporation and top personal income tax rates is 25 points -- 35% and 60% respectively. Another practice is to transform taxable capital income into tax-exempt capital gains when the latter are exempted from personal income tax.

The fact that tax planning exists shows that there is much more to taxation than the "headline" rates suggest. Building a full and accurate picture of income taxes, particularly where marginal rates are concerned, requires information on fiscal and social security programmes, as we have shown in this mini-series of articles. In fact, it is only by looking at such areas as all-in rates, income-related relief and benefits, and other taxes, such as low rates on capital income, that we can begin to appreciate how truly wonderful the world of taxation really is.


The data in this article are drawn from the OECD Tax Data Base.


©OECD Observer 217-218, Summer 1999

Economic data

GDP growth: -1.8% Q1 2020/Q4 2019
Consumer price inflation: 0.9% Apr 2020 annual
Trade (G20): -4.3% exp, -3.9% imp, Q1 2020/Q4 2019
Unemployment: 8.4% Apr 2020
Last update: 9 July 2020

OECD Observer Newsletter

Stay up-to-date with the latest news from the OECD by signing up for our e-newsletter :

Twitter feed

Digital Editions

Don't miss

Most Popular Articles

NOTE: All signed articles in the OECD Observer express the opinions of the authors
and do not necessarily represent the official views of OECD member countries.

All rights reserved. OECD 2020