TAXES ON THE RICH IN CONTEXT
The historical evidence in the previous two chapters tells us that transitions to democracy tended not to result in large increases in top rates of income and inheritance taxation. But, such changes did take place during periods of mass mobilization for war. These conclusions seem clear, but three important questions remain unanswered.
Were there additional taxes on the rich that did not come in the form of standard inheritance or income taxation? Ignoring these other taxes might lead us to biased conclusions. For example, perhaps when governments expanded the suffrage they did not raise top rates of income and inheritance taxation, but they taxed the rich in other ways. A look at other ways in which governments have taxed the rich reinforces our general conclusion: compensatory arguments have had the greatest impact on taxation of the rich. We can see this in particular with the imposition of war profits taxes and capital levies.
Were increases in top rates a sign that tax systems were becoming more progressive? We have already established that when top income tax rates went up, the income tax system as a whole became more progressive. The same holds true for the inheritance tax. But how much did other taxes borne principally by the poor and middle classes increase at the same time? There is no doubt that when governments financed the two world wars they raised indirect taxes, in addition to direct taxes. The incidence of many of these indirect taxes fell on groups other than the rich. It needs to be established how our overall conclusions about taxation of the rich are altered once we take this into consideration. Looking at the overall tax burden on different income groups can help address this issue, taking into account both direct and indirect taxes. As part of this, we also briefly consider the allocation of government spending between different groups, as well as the role of government debt. The evidence shows that our overall conclusions still hold even after accounting for the incidence of indirect taxation, spending, and debt.
Was an increased tax burden on the rich during wartime dictated by simple necessity? Perhaps governments needed money, and they took it where it was easiest to find it. This would be a bit like the reason why Willie Sutton said that he robbed banks—because that’s where the money was. In this chapter we show that this argument also fails to convince.
CAPITAL LEVIES AND WAR TAXES
In addition to taxing the rich through income and inheritance taxes, some governments during the twentieth century implemented one-off capital levies. A number of countries also established war profits taxes or excess profits taxes. Though these latter two terms were sometimes used interchangeably, the strict interpretation of the first is that it was a tax on all company profits earned during the war, whereas the second was a tax on the excess of company profits earned above a peacetime benchmark. To make matters even more complicated, some postwar capital levies were one-off impositions on all capital, whereas others were levies only on what was judged to be capital accumulated during the war. Readers interested in investigating the details further should consult the thorough 1941 study by Hicks, Hicks, and Rostas. In this section we confine ourselves to reporting instances where countries made use of one or more of these taxes. The arguments made in favor of capital levies, excess profits taxes, and war profits taxes inevitably took a compensatory form—given that many were sacrificing or had sacrificed, those who had done well out of the war should help pay for the war.
Capital levies are actually a very old idea, and they can have significant consequences for top fortunes. The question for us is when these levies have occurred and how it influences our broad conclusions about democracy, war, and wealth. Capital levies existed during the classical period in Athens, as well as in Rome. Some medieval rulers also imposed them. As an example, in AD 1188, wealth holders in England were obliged to pay a tax of one-tenth of the value of rent and movables in order to help finance the Second Crusade to expel Saladin from Jerusalem.1 It would hardly be surprising if we concluded that most early capital levies were associated with war, because this is basically all that governments did with their money at this time.
Barry Eichengreen has charted the twentieth-century history of capital levies showing when they occurred and what the outcome was.2 Based on this evidence, it seems clear that levies on capital have continued to occur for the same reason that they always did—because of war. However, in keeping with the new context of mass warfare, the rates imposed in twentieth-century levies were substantially higher than had been seen in prior centuries. In the wake of World War I the imposition of a capital levy was a very hotly debated topic in many countries across the European continent. In Czechoslovakia, Austria, and Hungary, governments implemented levies with marginal rates reaching up to 30 percent.3 Postwar levies in Germany and Italy also had very high top marginal rates, although with a provision that they could be paid over a term of thirty or twenty years, respectively.4
In the United Kingdom the idea of a general levy on capital first found support both from those on the left of the political spectrum as well as from some economists, such as Arthur Pigou. He made an explicit compensatory argument in favor of a capital levy:
From the statistics of estates passing at death it can be deduced that practically all the material capital of the country is held by persons over twenty years of age; that persons over forty-five, who constitute about one-third of these persons, own about three-fourths of the whole; so that the representative man over forty-five holds about six times as much material capital as the representative man between twenty and forty-five. But young men, who excel older men in physical strength, have been forced to give their physical strength in the war, while older men have been exempted. The fact that old men excel young men so greatly in financial strength suggests that the balance might be partly adjusted, and something less unlike equality of sacrifice secured, by a special levy whose incidence would in the main fall upon persons exempted from military service.5
In the end the United Kingdom did not adopt a capital levy, but one of the main reasons for this was that it was recognized that very high top rates of income and inheritance taxation were already serving essentially the same purpose.
In addition to implementing one-off capital levies, during the twentieth century wartime governments also levied special taxes on excess profits or war profits. The story of war profits and of public dislike of them is an old one, extending in the United States back to the Civil War and even the American war of independence.6 In 1937, the Gallup Poll asked Americans whether they thought the government should regulate profits during wartime. Fully 70 percent of those surveyed responded affirmatively.7 In 1938, the British Institute of Public opinion asked whether profits of armaments manufacturers should be limited. Eighty-one percent of respondents supported restrictions.8 Given these poll results, it is no surprise to see the following statement in the 1941 survey by Hicks, Hicks, and Rostas: “The sense of unfairness is particularly aroused when the high incomes are earned, not by those who are in the centre of the war effort, but by those who are on the edge of it.” They then added:
It is undoubtedly because of this feeling of unfairness that most of the schemes we are going to study have been imposed; even if it is recognized that the economic incentive to efficiency may be damaged by them, they are still considered to be justified as means of fostering national unity and maintaining morale.9
In the case of the two world wars, the unprecedented level of mobilization brought with it unprecedented opportunities for profit and unprecedented debate over what to do about the problem. Belligerent governments felt compelled to respond in some way with legislation designed to tax the windfall profits that many firms earned as a result of increased demand for their goods. The story of these war taxes tells us something about the extent to which the rich were taxed. It also tells us something about the terms of the debate.
Participants in World War I did not expect at first to be fighting a long war, and so there was a delay before debates began about the issue of war profits. The measures eventually adopted by governments were extensive but also sometimes only partially effective. This meant that political debates about taxing war profits would continue after war’s end. The United Kingdom adopted an Excess Profits Duty in 1915 that remained in place through 1921.10 The tax was a massive source of funds, bringing in one quarter of total tax revenue during the period. It was levied at variable rates that averaged out at 63 percent, but because of exemptions the average effective rate during this period was more like 34 percent. The United Kingdom was certainly not alone in imposing this type of tax. In the United States, in 1917 Congress passed legislation establishing an excess profits tax. The French government waited until two years after the war’s outbreak, but it too eventually established an excess profits tax.11 Meanwhile, Italy imposed a number of extra war-related duties on profits.
All the countries referred to in the previous section were democracies during the First World War. The autocratic countries that participated in the war also passed legislation to tax war profits. However, this legislation was less restrictive than in the case of the more democratic belligerents. Equally importantly, the autocratic countries were less effective at implementing this legislation. Germany still had a decentralized fiscal system with little revenue generated at the federal level. In the case of Austria-Hungary and Russia, fiscal institutions were simply less developed. These institutional differences make it difficult for us to know whether autocracies during World War I taxed war profits less because they were autocracies or because they had weaker institutions.
Debates about war profits took place not only in countries that were direct participants in the conflict. In neutral countries demands for war profits taxation also arose because of the windfalls that some industries enjoyed from increased demand for their products by belligerents. This was particularly the case in the Scandinavian countries, although marginal rates on these taxes were substantially lower than in belligerent countries.12 This result is logical given the type of compensatory claims that could be made. Some citizens in Scandinavian countries were benefiting from a windfall, but this was not a windfall generated by other citizens sacrificing on the battlefield.
Now consider the case of war profits during World War II. On the Allied side the authorities generally adopted war profits taxes that were even more extensive than those adopted during the First World War. In the United Kingdom the excess profits tax provided in principle for a 100 percent marginal rate of taxation on those profits in excess of a prewar standard. In the United States the Roosevelt administration also adopted a very restrictive regime for taxing war profits, even if taxes in the United States did not reach the same marginal rates as in Great Britain. In France the government took measures to restrict war profits as early as 1939, but occupation soon rendered these impositions meaningless. As a result, debates about war profits would reemerge after the liberation of the country. We return to this question in chapter 8. Finally, war profits taxes were not unique to the Allied side. Germany passed a war profits tax in 1939, albeit with substantially lower rates than existed in the United States or UK. Japan also imposed an excess profits tax during World War II.
During the Second World War belligerent governments also addressed the issue of potential war profits in other ways. In some cases they intervened directly to limit incomes and redistribute wealth. In the United States, the government made a brief attempt to limit the pretax incomes of chief executives. In 1942, the Roosevelt administration placed a cap on after-tax salaries equivalent to $25,000 a year. However, Congress soon repealed this measure. A more long-lasting regulation involved a limitation on salary increases. The salary increase limit was left in place until 1946, and it had some effect on executive pay trends during the period.13 Of more importance to the rich in some countries were the nationalizations of industries that took place immediately after 1945. In France a number of industries were nationalized, with their stockholders receiving either below market values for their shares or in some cases no compensation at all. The latter occurred in the case of Renault, which was judged to have collaborated with the enemy during the war. We return to this issue in chapter 8.
Overall, when we consider the broader context for taxation or levies imposed on the rich, our core conclusions are only reinforced. During the two world wars and in their immediate wake, the rich were taxed to an extent that previously would have seemed unimaginable. The story of war taxes and capital levies only reinforces our interpretation that compensatory arguments provide the most powerful political support for taxation of the rich. The question to which we now turn is whether other forms of taxation imposed equally heavy burdens on other segments of the population during wartime. To examine this we need to inquire especially about indirect taxation.
THE INCIDENCE OF INDIRECT TAXATION
In the previous section we concluded that, as with income and inheritance taxation, other taxes targeted at the rich have been closely associated with mass warfare. But so far we have said little about wartime levies that may have hit the poor and middle classes most heavily. This would have been principally a result of indirect taxes on common consumption goods. These are usually thought to be taxes with a regressive incidence because the poor and middle classes spend a larger fraction of their income on such items. There is no question that governments increased indirect taxes during the two world wars. We also noted in chapter 1 that prior to the nineteenth century, indirect taxes had been the primary means by which most European governments had financed their wars. The real question for us is whether the effect of indirect taxes during the nineteenth and twentieth centuries was large enough to outweigh the impact of increased income and inheritance taxes on the rich. In this section we present evidence to show that this was not the case. Even after the impact of indirect taxation is taken into account, wartime governments increased taxes on the rich more than the rest.
Calculating overall tax burdens is not a simple task. Ideally, we would like to know the percentage of an individual’s income that winds up being paid in taxes of any form, be they direct or indirect. One option would be to assume that the burden of indirect taxation is borne mainly by the poor, that direct taxes impact mainly the rich, and so the relative share of indirect to direct taxes is a measure of the overall progressivity of the tax system. The problem with this is that some indirect taxes, such as those on luxury goods, are borne by the rich. Likewise, in the modern era direct taxes like the income tax have fallen not only on top incomes but also on the middle class. The issue is further complicated by the fact that with indirect taxation, an increase in the tax on a good may either be borne by the producer in the form of lower profits or passed on to the consumer in the form of higher prices. Finally, conclusions about the burden of indirect taxation depend upon assumptions about the goods each group consumes and how much it consumes.14 Early tax burden studies of the sort that we consider in this section typically assumed for simplicity that either the entirety or a large part of indirect taxation was passed on to the consumer. The risk of this assumption is that the studies might overestimate the extent to which indirect taxation results in a decrease in progressivity. If we can show that war mobilization was associated with increased progressivity even under this assumption, our conclusions will be robust.
By far the best historical evidence on the overall tax burden comes from the United Kingdom. In 1919, Herbert Samuel published a study entitled “The Taxation of the Various Classes of the People” in which he attempted to calculate the overall tax burden on British households earning different incomes prior to and immediately following the First World War. One of Samuel’s main conclusions was that tax changes adopted during the course of the war had made the British tax system substantially more progressive. This is important evidence that reinforces our main conclusions.15
Building on the study by Samuel, in 1943 G. Findlay Shirras and L. Rostas published a more extensive analysis of the burden of British taxation. They provide figures on how much tax a family of five would pay for specific fiscal years between 1903 and 1941. The calculation includes all income and surtaxes, in addition to most indirect taxes. The results are shown in table 5.1.
A first feature of the Shirras and Rostas results is what they have to say about taxation prior to World War I. In fiscal 1903 the overall tax schedule resembled that of a flat tax levied at around 5 percent of total income. This means that the burden of the income tax and of indirect taxes essentially canceled each other out. It was as if the compensatory arguments made in the nineteenth century about the need for a progressive income tax to offset regressive indirect taxes had resulted in the reestablishment of treatment as equals.
Now consider the figure for fiscal 1913/14. This figure would reflect the changes introduced during the People’s Budget of 1909/10, but it would not yet reflect changes introduced during the First World War. The People’s Budget is conventionally described as having made the British tax system substantially more progressive. To a certain point this was true, but the extent of the change needs to be put into perspective. Whereas the tax burden for the poor and middle classes remained largely unchanged, the tax burden for the rich now crept up from about 5 percent to around 8 percent of total income. This may have been an unprecedented change, but it was very small compared to what would soon follow.
If the People’s Budget increased progressivity somewhat, the effect of World War I was on an entirely different level. By the end of the war, the tax burden on the lowest income groups had doubled, but the burden on the richest had increased more than fivefold. This is an extraordinary jump, and it shows that wartime increases in indirect taxation did not offset the massive increases in income and other taxes on the highest earners. It is also interesting to consider what happened after the end of World War I and during World War II. After the 1918/19 fiscal year, the overall tax burden on low-and middle-income groups remained essentially unchanged through the beginning of World War II. For higher income groups, the overall tax burden fell somewhat in the mid-1920s (between 6 and 13 percentage points), but was still dramatically higher than before the war. Between 1925 and 1937, rates on higher incomes returned to their end of World War I levels. This increase had two sources. First, Britain during the 1920s and 1930s was in a state of near permanent fiscal crisis. Second, by the mid-1930s, the international environment had become more dangerous, and the UK engaged in some rearmament (though not as much as Churchill and others advocated). An additional and more dramatic increase in progressivity occurred during World War II, as we can see based on the data for the 1941/42 fiscal year.
It is evident from the Shirras and Rostas data that mass warfare was associated with a drastic increase in the overall progressivity of the tax system in the United Kingdom. Ideally, we would have access to similarly detailed studies tracking tax burdens over time in the other countries. Unfortunately this is not the case. There are nonetheless some scattered studies for a few countries that provide useful information.
The most detailed study on French tax burdens during the First World War was conducted by Robert Murray Haig. Though his assessment was far less detailed than that of Shirras and Rostas, Haig did provide a comparison of the relative tax burden in 1913 and in 1919. To do so he took total revenues generated from different taxes and then made a judgment whether the incidence of the tax in question fell mostly on the poor or mostly on the rich. As Haig emphasized, this was preferable to a simple distinction between indirect and direct taxes, because many of the indirect tax increases that France had adopted during the war were on luxury items purchased primarily by the rich. Haig calculated that in 1913, 46.5 percent of France’s taxes were weighted against the rich, 38.4 percent against the poor, and 15.2 percent had no definite weight. It is possible that this meant that the overall schedule in France at this time resembled a flat tax, just as was the case in the United Kingdom prior to the People’s Budget. But of course this evidence is too approximate for us to make this judgment. Now turn to Haig’s conclusions about the relative burden of taxation in 1919: 61.2 percent weighted against the rich, 31.2 weighted against the poor, and 7.4 percent with no definite weight. In other words, the burden of French taxation became more substantially progressive as a result of the First World War.
Unfortunately for our purposes, detailed studies of the overall tax burden in the United States did not begin until the 1930s. Given this, the best assessment of the effect of World War I on the tax burden was provided by none other than Edwin Seligman. In a study of U.S. war revenue acts, Seligman concluded that among the portion of war funds raised through taxation, fully 73 percent was raised through taxes on wealth, 13 percent through taxes on luxuries or “harmful” consumption (i.e., liquor), and the remaining 13 percent from other taxes that were more likely to impact groups other than the rich.16 On this evidence alone it seems hard to deny that the First World War resulted in the overall U.S. tax system becoming substantially more progressive.
In 1937, Mabel Newcomer produced the first study that gives us a true picture of the overall burden of taxation in the United States. She based her study on tax figures from the Revenue Act of 1936. By this time the Roosevelt administration, as well as the Hoover administration in its last days, had raised taxes on the rich. This reversed a trend in which prior Republican administrations had lowered taxes on the rich relative to their World War I peak. In her calculations Newcomer included estimates of the effect of all direct taxes, including estate and gift taxes, as well as all indirect taxes. She then considered the tax burden for ten hypothetical families with annual income ranging from $500 to $1 million. The calculations also took account of the likely sources of income for people at these different levels (in particular capital versus labor income). Newcomer found that the very wealthiest households (those earning $1 million a year) would pay roughly 80 percent of their income in taxes, whereas at the next level down (those earning $100,000 a year) this figure dropped to roughly 40 percent. At the next level down (those earning $20,000 a year) the combined tax rate dropped to roughly 30 percent. The trend continued thereafter, with the lowest income groups paying between 10 and 15 percent of their income in taxes, depending on the assumptions. We can conclude from Newcomer’s study that there was a high degree of progressivity in the U.S. tax system in the late 1930s.
Evidence for the World War II era suggests that the overall tax burden in the United States became even more progressive as a result of this conflict. It is known that during World War II the Roosevelt administration lowered the exemption limit for the income tax to make it a mass tax. We need to take this change into account along with any other tax increases to see the effect of World War II on overall progressivity. The best available evidence on this subject was compiled by John Adler in 1951, based on his estimates for the postwar tax burden and prewar estimates by Gerard Colm and Helen Tarasov. Adler’s results suggest that the U.S. fiscal system became more progressive during World War II.17
In the end, it would be helpful to have more evidence, but the few studies that we have cited in this section all point in the same direction. Participation in mass warfare was associated with increased progressivity of the overall tax burden. In the next section we see if a look at spending and debt might alter our conclusions.
THE EFFECT OF SPENDING AND DEBT
In considering taxes on the rich, we are not directly interested in government spending, not with government borrowing, but since spending and debt have clear distributional implications, we still need to be concerned with these issues.
If taxes vary in their incidence on different social groups, the same is of course true with spending. One possibility is that the expansion of the suffrage did not lead to heavier taxation of the rich, but it did lead to greater welfare spending on the rest of the population. In other words, maybe governments didn’t shift the tax burden toward the rich, but they did shift the benefit of spending more toward the poor and middle classes. The short answer to this question is yes, there was an increase in redistributive spending following suffrage expansions, but this increase was on such a low base that the end effect was minimal. It was not until World War II that governments created anything resembling modern welfare states, and this was well after the expansion of the suffrage.
One of the most frequently cited studies on this subject is Peter Lindert’s book, Growing Public. Lindert compiled data on public spending in a set of industrial countries from 1890 to 1930. He looked in particular at public spending on welfare, pensions, health, and housing, the main categories of redistributive transfers at the time. Based on these data, Lindert argued that the suffrage played a significant role in the expansion of public spending on these categories. However, he also took pains to place this conclusion in context. By the 1930s, even in those countries where redistributive transfers were highest, they amounted to only about 3 percent of GDP. This is of course an extremely small figure by post-1945 standards.
We used Lindert’s data to perform the same sort of differences in differences test that we have performed elsewhere in this book. This analysis suggested there was in fact no effect of extension of the suffrage on redistributive spending. However, when we substituted our competitive elections measure previously defined in chapter 3, we did see such an effect. Countries in which at least 50 percent of adult males could vote and where chief executives were elected in multiparty competition had overall redistributive spending that was one-half of one percentage point of GDP higher than those that didn’t. This is a statistically significant effect, but it is small in terms of magnitude. In a related study using different data, Toke Aidt and Peter Jensen looked at the effect of suffrage extension on total public spending. Using a similar statistical procedure, they found that suffrage extensions were associated with an expansion of the public sector by about 1.5 percentage points of GDP.18 This again is a statistically significant effect, but it is a small one.
Now, let’s consider the effect of mass warfare on redistributive spending. The end of the Second World War is most often associated with an increase in welfare spending due either to programs explicitly targeted at veterans, such as the GI Bill in the United States, or programs targeted at the whole population, such as Great Britain’s National Health Service.19 Some observers have claimed such an effect for World War I as well.20 However, if we return to the Lindert data we see that by 1930 overall redistributive spending was no higher in countries that had participated in World War I when compared with those who had not. Most of the distributional changes that occurred as a result of the First World War were due to changes in the tax system. This would not be the case with World War II, and we consider that issue further in chapter 8.
In addition to spending, we also need to consider the effect of government borrowing. During the two world wars, governments were forced to finance the bulk of their expenditures by borrowing. The alternative of financing the war only through current tax increases would have quickly asphyxiated their economies. At the time, and particularly during World War I, it was often argued that resorting to borrowing favored the rich. Allowing the rich to buy government bonds would allow their funds to be remunerated, which of course is not the case with taxes.
To what extent does a consideration of government debt alter our core conclusions about mass warfare and taxing the rich? We can first think about this problem in the abstract. Say that a society is divided into those who can supply labor and those who own capital. If capital owners invest in government bonds they earn a return on this asset, whereas if labor is conscripted then it runs great risks and is paid a below-market wage. The unfairness seems evident. One response to this is to say that a capital owner will be simply earning a return on government debt instead of on another asset.21 While a true statement, in the early twentieth century this argument was often insufficient to quell the critics. It also ignores the possibility that if capital is in short supply, then government borrowing may push up interest rates, leading to a higher return for capital owners than they would enjoy in peacetime.
The first question we need to address then is whether investors in government debt during the two world wars earned higher returns than they would have had the wars not taken place. In an ex post sense the answer is no, this was certainly not the case. We know that investors suffered heavily from the inflation, and in other cases outright default, following these wars. But what about in an ex ante sense; did investors earn higher expected returns on government debt than they would have otherwise? The short answer to this is that even if investors during World War I did earn higher nominal yields, it’s not clear that they earned higher expected returns than would otherwise have been the case. For one, inflation crept up in most countries during the war, cutting into returns. Second, there must have certainly been a perceived increase in default risk for some countries. This too would have reduced expected returns.
The second question we need to address is whether governments did anything to address the perception that bondholders were unfairly profiting from the war effort. In fact, governments intervened quite heavily in capital markets during both wars to lower their effective cost of borrowing. There was an obvious financial incentive to do this, but this move also addressed perceptions about the unfairness of borrowing. In any case, these interventions certainly did not help investors. As an example, in the United States during World War II the Federal Reserve and Treasury pursued a policy of keeping the interest rate on long-term government debt below a ceiling of 2.5 percent.22 Governments also used means other than explicit interest rate targets to intervene in capital markets. This was the case of the UK government in World War I, which controlled new capital issues from other sources and limited rates on deposit accounts in an attempt to steer more money into government bonds.23 These were all forms of implicit taxation of capital owners.24
To summarize, a look at government spending and indebtedness does not alter the basic conclusions of this book. Compensatory claims associated with mass warfare led to heavy taxation of the rich, and a look at government actions in the area of spending and debt does not alter this conclusion. The final question we consider in this chapter is whether the design of wartime tax changes might be explained by pure fiscal necessity.
FISCAL NECESSITY—THE WILLIE SUTTON EFFECT
Willie Sutton said he robbed banks because that’s where the money was. A key question for us is whether governments during the two world wars taxed the rich for exactly the same reason. Perhaps high top rates of income and inheritance taxation didn’t have anything to do with fairness; they may have been a simple necessity for governments. This “Willie Sutton” argument has several different variants, and so we consider each here. According to the first variant, the rich were taxed because all other sources of finance had been used to their maximum, and therefore there was no alternative. According to the second variant, the rich were taxed because governments with finite fiscal capacity found it desirable to concentrate their energies where they would have the greatest yield in terms of revenue.
The first variant suggests that when other sources of revenue have been used up, then governments in desperation will be forced to tax the rich heavily. Perhaps this is what happened during the two world wars. More specifically, we might think that once a government had raised indirect taxes to a maximum that the economy could sustain, and once it had borrowed to the point where it no longer had access to credit, then taxation of the rich would remain the final option. In fact, in the case of World War I this is exactly the opposite of what happened. During the war itself the British and American governments went further than governments in Germany, France, Austria, or Russia in taxing the rich. It was also the case that in comparison with these other states, the British and American governments funded much more of their war effort out of current taxation, as opposed to borrowing. Under these circumstances it is difficult to argue that heavy taxes on the rich arose out of desperation. In fact, just the opposite seems to have been the case.
The other variant of the fiscal needs argument suggests that in a world where fiscal capacity is finite, it would make sense to focus attention on taxes on the rich because these would have the greatest yield. Once we consider this argument more closely we can see that there is little basis for believing that this is why governments taxed the rich heavily during the two world wars. If finite fiscal capacity were the problem, then governments would have chosen to tax the rich heavily centuries before the two world wars because bureaucratic capacity was certainly in shorter supply than in 1914. But this is not what they did.
There is also further evidence against this finite fiscal capacity idea. To see this, consider the structure of income tax schedules that were adopted during the First World War. In all countries with income taxes the vast majority of individuals or households were exempted. Now, it is plausible that this choice was influenced by considerations of administrative capacity—concentrated on the households that would yield substantial revenue. This might help us understand why only the top 10 percent of households were subject to the income tax instead of the top 50 percent. However, if this were all that was determining the choice of who to tax, the simplest strategy would be to raise taxes on the 10 percent by the same proportion by adopting a flat tax. After all, one of the key arguments for a flat tax is its administrative simplicity. As we already saw in chapter 3, this is precisely the opposite of what happened. During and after the First World War, governments raised income tax rates on the very richest households by considerably more than on those who were merely within the top 10 percent of households or even within the top 1 percent. This differential taxation within the top 10 percent also undermines the argument that wartime governments taxed the rich because of falling trade revenues. Chapter 3 also presented evidence that the effect of war on top rates of income taxation was greater in democracies than non-democracies. This difference again suggests that some factor other than finite fiscal capacity drove the war effect.
The history of the inheritance tax gives us a final reason to be skeptical about the finite state capacity argument. Recall from chapter 4 that historically, governments began taxing inheritance well before they began taxing income. They made this choice because it was easier to tax inheritance. This is a clear case where availability of administrative capacity influenced the choice of taxation. However, recall also another finding. Historically, governments have never relied on inheritance taxation as a principal source of revenue. Prior to the two world wars they also never levied inheritance taxes at very high rates. If governments had access to a tax that required little capacity to collect, yet they didn’t tax the rich heavily, then it seems that state capacity alone cannot be the explanation. Once again, some other factor must have been operating.
We have shown that neither the first variant of the fiscal needs effect (desperation) nor the second variant (finite capacity) can explain why the rich were taxed so heavily during the two world wars. There’s also a final fact to which we already referred in chapter 1. It is true that the two world wars were events that were more expensive than anything that had happened previously. However, it isn’t the case that they were more expensive than what would happen subsequently. It’s useful to remember that over the last half century many governments have steadily increased the amount of revenues they draw from their economies. They have done this to the point where, even in a normal year, governments are collecting as much revenue relative to the size of the economy as they were at the height of World War II. Has this resulted in a massive shift toward taxing the rich? In fact, exactly the opposite has taken place. Governments have reduced taxes on the rich. Therefore, just because a government needs a lot of revenue doesn’t mean that it needs to tax the rich.25
To sum up, there’s no doubt that when governments have taxed the rich, they have done so because they needed money. But that doesn’t mean that this was the only way to achieve this objective. We have shown that the pattern of taxation of the rich does not support the Willie Sutton argument. The governments that have taxed the rich most heavily were not those that had the fewest alternatives for raising finance. Likewise, there were many cases where governments lacked fiscal capacity, yet they chose not to tax the rich heavily. All of this reinforces the idea that the story of taxing the rich has more to do with politics.
In this chapter we have asked if the patterns we saw with income taxation in chapter 3 and with inheritance taxation in chapter 4 tell us something meaningful about trends in the taxation of the rich relative to the rest of the population. A look at the broader picture confirms our prior conclusions. We have also argued that this trend cannot be explained by simple necessity. In the next three chapters we show how changing perceptions about a fair tax system helped to lead first to the rise and subsequently to the demise of taxing the rich.