Introduction
Three decades after South Africa’s first democratic election, the top 10 per cent of the population owns more than 90 per cent of the total wealth (Davis Tax Committee, 2018: 4), and the country remains one of the most unequal societies on earth (Sulla et al., Reference Sulla, Zikhali and Cuevas2022). Unequal access to land, education, employment opportunities and the spatial design of cities and towns continue to reflect the legacies of apartheid. Prominent among the sources of continuing economic and political inequity has been the failure of post-apartheid land redistribution (Ngcukaitobi, Reference Ngcukaitobi2021). While most agree that inequality detrimentally shapes the life opportunities of the majority of South Africans, there is increasing evidence that it is also undermining the post-apartheid settlement – whether in the form of public protests, corruption or simply increasing disillusionment with the political and constitutional order. It is in this context that land has once again become a central focus of political and legal conflict (Klug, Reference Klug2018).
Since market-led reform policies have clearly failed to produce the necessary redistributive justice required to address apartheid’s legacies, it is time to explore more interventionist options. This raises an important question: might a transformational tax provide the basis for a new social contract that will further the promise of South Africa’s post-apartheid constitutional order? To address this question, I explore a comparative history of wealth taxes to reflect on the forms a proposed transformational tax may take. This comparative approach explains in part why recent debates about an annual wealth tax in South Africa failed to see the potential such a tax presents to address inequality in South Africa. The Davis Tax Committee, appointed in 2013 by then Finance Minister Pravin Gordhan to advise on tax policy, investigated the idea of a wealth tax, focusing its attention on an annual tax, which the Committee found would not offer a significant advance on the existing tax system and would be difficult to implement.
This was not the first proposal for a wealth tax in South Africa. That came from the African National Congress (ANC) in October 1991 in a report from a commission on land at a conference on affirmative action organised by the ANC Constitutional Committee in Gqeberha (then Port Elizabeth). The commission reported to the conference that to address the history of colonial dispossession and apartheid forced removals, as well as the exclusion of Africans from the land market since 1913, there would need to be a significant redistribution of land. In this context, it proposed a wealth tax to serve as a source of funding to compensate those whose land would be expropriated to enable restitution or redistribution. Compensation would ensure that individuals would not bear the brunt of a process designed to address historical legacies. At the time, the ancien regime and the establishment press vociferously rejected the idea of a wealth tax (Krige, Reference Krige1991). It is worth noting that while often rejected as either utopian or unworkable, the idea of a wealth tax in South Africa is not (now or then) an outlandish idea. In fact, in the immediate aftermath of the first democratic election, a small one-off ‘transition levy’ of 5 per cent ‘on individuals and companies with an income in excess of 50,000 rands’ a year was successfully used to cover the costs of the democratic transition (Carlin, Reference Carlin1994). Now, after over a quarter of a century in which the legacies of apartheid persist, and the country has experienced economic, political and pandemic disruptions, the need for a new social compact is being recognised. In this new context, compensation for necessary expropriations will be one among many needs a transformational tax might address. In fact, identifying specific needs, including rural and urban land reform, would be an important aspect of any new social compact. It is also clear that while a relatively moderate threshold exemption on wealth would exclude the vast majority of black South Africans from the tax, it would apply to all with wealth over the defined threshold, regardless of their earlier status among the oppressed. The burden of the tax could be moderated by imposing a sliding scale so that a higher rate applies to the very wealthiest 1 or 2 per cent of the population.
Before describing a proposed transformational tax, this chapter first presents a brief historical survey of different forms of wealth taxes in several countries. This comparative analysis demonstrates that capital levies have been a more effective means of ensuring redistribution compared to annual wealth taxes. The conclusion to be drawn from this is that any plans for a transformational tax to address the legacies of apartheid and gross inequality must consider the imposition of a capital levy. The final section of this chapter applies the analysis of former experiences with capital levies to imagine the outlines of a transformational tax for South Africa.
International Experience with Wealth Taxes
Over the twentieth century, there were distinct periods in which wealth taxes were proposed and implemented in various countries. The first period, around World War I, saw wealth taxes used as a means of reducing public debt. The second period occurred in the aftermath of World War II, when wealth taxes of different forms were introduced in many countries, including France, West Germany and Japan. Finland resorted to a capital levy twice in the 1940s, once to address the plight of Finish citizens who were expelled from the Karelia Peninsula, which the Soviet Union took in 1940, and then again in 1944. A third period followed the 2008 financial crisis, while the economic impacts of the COVID-19 pandemic have produced new proposals for wealth taxes.
Within this history, it is important to distinguish between annual net wealth taxes and one-off wealth taxes or capital levies, as each form has distinct goals and means of implementation, with significant consequences for the idea of a transformational tax in South Africa.
Annual net wealth taxes are what are most regularly considered when reference is made to a wealth tax. The prime examples include the Swedish wealth tax introduced in 1910 and taken up in various European countries in the 1970s and again after the global financial crisis in 2008. Many of these annual net wealth taxes were ended in the 1990s, and while some were reintroduced post-2008, others have faced constitutional and other challenges. In the case of Germany, where wealth taxes of various forms have been repeatedly used and are provided for in the Basic Law, the failure to regularly update real property values led in 1995 to a Constitutional Court challenge, which struck down the annual net wealth tax as unconstitutional for violating the Basic Law’s equality clause. Annual net wealth taxes, as well as the utopian idea of a global tax on capital suggested by Thomas Piketty in his 2014 book, Capital in the 21st Century, are quite distinct from the idea of a capital levy or one-off wealth tax like the German Lastenausgleich or equalisation of burdens tax that was adopted in the wake of World War II. A capital levy may be ‘defined simply as an extraordinary tax which is assessed on capital owned at a given date’ (Robson, Reference Robson1959: 23). If we focus on these one-off wealth taxes, or capital levies, which sought to achieve more than debt relief, we encounter a few very particular historical cases from the period after World War II. In the case of France, the levy served both to raise public finances and also to punish those who had profited by collaborating with the Nazi occupiers. The levy was 25 per cent on capital as of 1945, plus 100 per cent on additions to capital during the occupation from 1940 to 1945 (Carroll, Reference Carroll1946). Four countries – Germany, Japan, Finland and Korea – adopted versions of capital levies whose overall goals were reconstruction, equalisation and democratisation. The German case was initiated by the process of financial reform imposed by the occupying powers in 1949 and was incorporated into the sharing of burdens law or Lastenausgleich (Equalization Law of 1951) in 1952. In the case of Japan, the occupying forces imposed a 90 per cent capital levy on the top 2–3 per cent of the population, who were considered beneficiaries of Japanese militarisation and aggression. Finally, Finland and South Korea introduced programmes linked to land redistributions that effectively served as forms of one-off capital levies.
Sharing the Burdens of Reconstruction: The German Equalisation Tax
One of the more significant and ambitious capital levies in world history came out of West Germany immediately following the end of World War II. Most post-war levies were intended to combat inflation or supplement ordinary public spending (Robson, Reference Robson1959: 28–32). The German levy, however, was, from its start, intended to distribute the harms of war as equitably as possible (Robson, Reference Robson1959: 28–32). Hitler’s regime intentionally ran up German war debt during the war with the promise of compensating citizens out of the plunder of conquest (Hughes, Reference Hughes1999: 1). The defeat of the Nazis left the nation, like most of Europe, physically and economically destroyed. German cities suffered extensive destruction. Hamburg alone took more damage than all the bombed cities in Britain. In Western Germany, over 20 million people were homeless when the war ended (Botting, Reference Botting1985: 123–25).
The destruction was not, however, uniform across Germany. Where some were left completely destitute, with homes and businesses destroyed, others escaped largely uninjured (Hughes, Reference Hughes1999: 2–3). While all war-damaged countries implemented some level of post-war aid to citizens, Germany is largely unique in its attempt to distribute wealth so that pre-war levels of property ownership were restored (Hughes, Reference Hughes1999: 2–3). However, the money for this rebuilding could not come from everyone equally, as many had nothing to give (Berghahn & Poiger, Reference Berghahn and Poiger1945–1961: 7). The solution became known as a Lastenausgleich or ‘equalisation of burdens’. The burden of rebuilding the country would fall upon each German proportional to their own needs and surviving property (Berghahn & Poiger, Reference Berghahn and Poiger1945–1961: 2). Social justice would be the driving factor, with those least harmed by the war being levied to compensate those most harmed (Heller, Reference Heller1949: 227).
The Lastenausgleich represented not only a shift away from Nazism but also a break from the pre-war German republic. The programme sought both to balance out the harms of the war and assist the nation in becoming more prosperous for all. Article 20 of the newly adopted Basic Law (Constitution) mandated that German society maintain itself as a ‘democratic and social federal state’ (Berghahn & Poiger, Reference Berghahn and Poiger1945–1961: 7). Beyond the immediate social benefits of the programme were also geo-political concerns. The perceived threat of the Soviet Union in East Germany pressured the Western Allies to ensure that a quickly rebuilt Germany could play a part in its own defence (Berghahn & Poiger, Reference Berghahn and Poiger1945–1961: 9; Hughes, Reference Hughes1999: 168), especially in the emerging ideological struggle of the Cold War.
Taking the asset base of 1948, the Equalization Law set a 50 per cent tax rate on surviving post-war assets and spread the tax debt over the next thirty years, which saw the tax being collected quarterly until 1979 – raising, it is claimed, 42 billion Deutsche Mark (DM) over this period (Bach, Reference Bach2012: 6). Additional features of the Equalization Tax include the fact that it was mainly assessed on property and business assets (including state-owned enterprises), while financial assets were granted a relatively high exemption of 150,000 DM. In addition, a tax allowance of 5,000 DM was granted for natural persons with leviable assets up to 25,000 DM, with a gradual decline to zero exemption for those with assets over 35,000 DM. To place these numbers in context and demonstrate their nominal value, the average annual pensionable income in post-war Germany in 1952 was 3,850 DM. As Stefan Bach concludes, ‘[d]ue to high growth rates of national product and income, [the] … economic significance and burden gradually decreased in subsequent decades. At the same time, it was possible to mobilize significant resources for reconstruction and the integration of displaced persons and refugees. In this respect, burden sharing was a financial, economic, and sociopolitical success’ (Bach, Reference Bach2012: 6). In its implementation, the Länder (German states or provinces) were directed to ‘devote 85 percent of the income for Lastenausgleich purposes, such as housing construction for war damaged individuals’, and the Länder were required to transfer 15 per cent of the income to the central authorities for ‘supra-regional balancing out’ (Hughes, Reference Hughes1999: 74).
The Lastenausgleich proceeded in two distinct phases. Recognising that a comprehensive levy and distribution would take years, the German government first rushed out a smaller levy intended to provide more immediate aid to those facing imminent harm due to the destruction (Hughes, Reference Hughes1999: 73). Taking effect in 1949, this levy imposed a 2 per cent tax on the value of real property with an exemption of 3,000 DM, increasing to 3 per cent for property with a value of more than 15,000 DM. This levy also distinguished between ‘necessary’ and ‘excessive’ material assets, taxing the former at 4 per cent and the latter at 15 per cent. The proceeds of 2.75 billion DM were used to great effect as a welfare-like entitlement (Heller, Reference Heller1949: 229). Those who had been expelled from their homes, who had had homes destroyed, who had lost their money in the currency revaluation and who had been politically persecuted were eligible for payments even if they demonstrated only a relatively low threshold of loss. For example, a person expelled from their home could get monthly aid for showing a loss of 300 DM in assets. This levy also provided support for those who could not work due to disability or age as well as supplements for the worker’s dependants (Hughes, Reference Hughes1999: 77). Most importantly, this levy established as precedent the principle that future levies would be calculated using the value of a person’s property on 21 June 1948 (Hughes, Reference Hughes1999: 78).
The second phase saw a major levy of assets meant to assist in the rebuilding of German society and economy. At its core, the levy was a one-time tax on the value of an intact property. The Lastenausgleich law imposed the tax at a rate of 50 per cent on real property. The payments to discharge this levy were to be made over a period of thirty years (Robson, Reference Robson1959: 31). This number came from an analysis done in 1950, which concluded that the German economy could not afford to levy more than 1.5 billion DM a year (Hughes, Reference Hughes1999: 151). The government decided to apply the 50 per cent rate on the theory that it would demonstrate the equal nature of the levy. Amortising payments over thirty years would result in a yearly revenue of about 1.5 billion DM (Hughes, Reference Hughes1999: 151). Further exemptions for the first 5,000 DM of leviable assets ensured that lower and middle-class German citizens would not be overburdened. Exemptions on the first 150,000 DM were available to Nazi victims whose property had been restored after the allied victory. Complete exemptions were available for property given to successor organisations when the true heirs could not be found (Hughes, Reference Hughes1999: 153). This was clearly in the interest of not taxing the victims of war for the costs of those defeated.
The second capital levy raised a total of 42 billion DM, around twenty times the amount raised with the first and 60 per cent of the nation’s 1952 gross domestic product (GDP) (Bach, Reference Bach2012: 6). The money raised was distributed based on a number of criteria. First, those with recognised legal claims for things such as property loss and damage were given direct compensation (Hughes, Reference Hughes1999: 155–56). Others without legal claims were allowed to make use of generous loans to support economic reintegration (Hughes, Reference Hughes1999: 156). Those who had lost goods rather than real estate were also entitled to payments. Persons who had lost at least 50 per cent of their household goods were entitled to graduated yearly sums of at least 800 DM for twelve years based on the amount of income they had at the time of the payment (Hughes, Reference Hughes1999: 157). Importantly, the claims of those who had lost money in the currency reform were not recognised under the second levy; this was on the theory that the other forms of compensation would be available to them anyway (Hughes, Reference Hughes1999: 158). Furthermore, the final law placed no maximum on the amount of compensation a single person could get, though the amount they received was proportionally reduced the more their claims rose (Hughes, Reference Hughes1999: 163).
One of the most surprising aspects of the entire programme was how relatively few barriers to implementation it faced. The elites of Germany had stood firmly against similar attempts at reform following World War I. The disaster of World War II, however, seemed to leave a bad taste towards any kind of war or post-war profiteering. Simply being rich in post-war Germany might indicate a failure to make or at least appreciate the sacrifices made by the populace. The Lastenausgleich was seen to be a part of the general denazification of the state where the immoral profits of the past would be collected and used for the public good (Hughes, Reference Hughes1999: 113). The result was mass popular support by most sections of West German society and among the Western Allies (Hughes, Reference Hughes1999: 81, 113).
Building Democracy: Capital Levies in Post-War Japan
The Japanese case saw a one-off capital levy imposed in 1946–1947 as one component of a sweeping political and economic overhaul that included tax reform, land reform and constitutional reform. The levy’s first objective was to reduce the internal debt burden inherited from wartime. The second objective was to provide finance for the recovery programme, and the third was to reduce income inequality. The goal of this last objective was to reduce the wealth holdings of a small minority of exceptionally rich individuals – the Zaibatsu – owners of the great holding companies who were considered responsible for promoting the war and had profited well from it. The wealth tax was imposed on families whose property was worth at least 100,000 yen as of 3 March 1946. The rates of the tax rose from 10 per cent on the lowest bracket to 90 per cent on estates worth more than 15 million yen. As a result of the existing inequality, the levy was only imposed on 2–3 per cent of the richest families.
World War II left the country physically and economically destroyed. The Japanese government had insured nearly every private war enterprise and guaranteed numerous loans from private banks (Shavell, Reference Shavell1948a: 133). Indeed, some 80 per cent of the total expenditure for the war came from borrowing. By the end of the war, Japan had accrued over 100 billion yen in debt, more than twice the total capital reserves of all Japanese businesses combined (Kurihara, Reference Kurihara1946: 844). Many capital levies in the post-war world were intended to address these staggering levels of debt. Like other nations, Japan’s economy underwent extreme restructuring at the behest of the occupying United States (Bisson, Reference Bisson1954: 1).
A capital levy was but one part of this post-war reform. Simple economic improvement was not, however, the primary justification for the levy itself. Imperial Japan was a stratified society with massive wealth inequality and an ingrained aristocracy (Shavell, Reference Shavell1948b: 131; Bisson, Reference Bisson1954: 11–13). This old guard stood in the way of the American occupiers who sought to rebuild Japan into a peaceful and democratic partner in the Far East (Shavell, Reference Shavell1948b: 131). To accomplish this, the occupiers made it their primary objective to distribute the concentrated Japanese wealth widely among the population (Shavell, Reference Shavell1948a: 127). The primary design of Japan’s capital levy was, therefore, not primarily a means to pay for government expenses (though this was an element) but, rather, a targeted attack on the richest and most powerful of Japanese society (Shavell, Reference Shavell1948b: 130).
The Zaibatsu, literally ‘financial clique’, was the chief target of the occupation administration (Bisson, Reference Bisson1954: 1). The clique was an interrelated cartel of family businesses that represented just the top 3 per cent of Japanese society but controlled the majority of commercial and financial interests (Shavell, Reference Shavell1948a: 127). Made up primarily of four large organisations, the Zaibatsu exerted almost plutocratic power over Japan and were occasionally even delegated some government functions, such as tax collection and currency distribution (Bisson, Reference Bisson1954: 7). For example, the Mitsui corporation, one of the largest of the Zaibatsu organisations, employed nearly 3 million people within Japan and East Asia in 1945 (Bisson, Reference Bisson1954: 11). Naturally, this kind of power led to extreme concentrations of personal wealth among the families that controlled them. Nineteen families in 1930 had yearly incomes of at least 1 million yen compared to the 84 per cent of the population who made less than 800 yen per annum (Bisson, Reference Bisson1954: 19). Only the Imperial Household itself had personal wealth comparable to these families.
Even after the war, this distribution of economic resources had not changed, and indeed had worsened. By the time the valuation of leviable assets was completed, only 269 households had sufficient assets to be placed within the levy’s top two tax brackets, with combined taxable assets (6.9 billion yen) well above that of the 58,000 households in the lowest taxable bracket (Shavell, Reference Shavell1948b: table 5). The interrelated nature of this clique, representing the executives of practically every major company in the country, drew the attention of the American occupiers, who demanded its dissolution. Indeed, the firms were one of the main drivers of the overall Japanese economy. For example, in 1944 just four Zaibatsu banks lent out 6.7 billion yen or 74.9 per cent of all private money lending (Yamamura, Reference Yamamura1964: 540–41). Changing this system would be necessary if the Allies were to successfully rebuild Japan as a democratic nation.
To that end, the levy attacked only those with the highest levels of personal wealth in Japan. This strategy meant that the Zaibatsu would end up paying most of the levy. Real and intangible property starting at a value of 100,000 yen was subject to a graduated one-time tax. This increased from 10 per cent of the first 15,000 yen above the 100,000-yen exemption to a full 90 per cent of assets worth over 15 million yen (Shavell, Reference Shavell1948a: 132). For perspective, the average monthly household income in 1956, well after economic recovery began, was between 5,000 and 6,000 yen (Yamamura, Reference Yamamura1965: fn. 21). Household furnishings, clothing and other necessities were exempted from the levy, meaning that only genuinely wealthy landowners ended up contributing to the overall levy. Indeed, over half the total levy was eventually collected from the value of real estate. Critically, however, the final levy specifically excluded taxation of corporate assets on the grounds that this would result in unfair double taxation of those already subject to some of the highest levels of the levy (Shavell, Reference Shavell1948a: 132). Despite their exemption, the old corporate structures were faced with significant regulation by the occupation administration, which intended to break the power of the companies themselves (Bisson, Reference Bisson1954: 120–21).
The greatest problem faced by the levy was from post-war inflation of the yen, which occurred while the government was still attempting to establish the total property value to be taxed (Shavell, Reference Shavell1948b: 132). Between the surrender in August 1945 and May 1946, the average cost of living rose 850 per cent (Kurihara, Reference Kurihara1946). Inflation was not truly brought under control until 1949, by which time prices in Tokyo were over 200 times their 1934 level (Bisson, Reference Bisson1954: 94). The government originally intended that the levy be imposed in mid-1946. However, despite the massive inflation in prices, it was not until December of that year that collection actually began. In total, more than a year was allowed to pass between the time that taxable assets were valued and the time of actual collection. This delay resulted in a significant loss to the potential amount of revenue that could have been collected (Shavell, Reference Shavell1948b: 132). However, the levy was recognised as having an overall deflationary effect on the Japanese economy (Shavell, Reference Shavell1948b: 133; Kurihara, Reference Kurihara1946: 851–52), thus slowing inflation.
The levy was an overall success, as shown by the absence of significant attempts to dodge the tax, the total amount generated and the reshaping of the economic system. Those subject to the levy voluntarily declared 39 billion yen in total liability by the original deadline (Shavell, Reference Shavell1948b: 133). The finance ministry attributed the success of this portion of the levy to one particular method of enforcement: the government retained the option to mandate the sale of any piece of land at the value originally assessed if it determined that that valuation was inadequate (Shavell, Reference Shavell1948b: 132). The final amount raised was roughly equal to the target yield of 43.5 billion yen, or 120 per cent of total tax revenues for 1946–1947, and 9 per cent of Japan’s total private national wealth in March 1946 (Shavell, Reference Shavell1948b: 131). The Zaibatsu continued to exist and shared in the overall economic recovery, but the concentration of wealth in only a few companies was largely replaced with a much more open and competitive economy (Rotwein, Reference Rotwein1964: 263; Yamamura, Reference Yamamura1964: 552–53). The top family members saw their personal wealth greatly reduced and were largely excluded from the operational control of their companies (Bisson, Reference Bisson1954: 202). Indeed, some families saw their personal assets decrease by as much as 95 per cent (Bisson, Reference Bisson1954: 93). Most importantly, the control structure of the firms changed dramatically, with many shareholders controlling small portions of the firms where once one family might control an entire industry (Bisson, Reference Bisson1954: 201; Rotwein, Reference Rotwein1964: 266).
Funding Land Reform and Industrialisation: A Capital Levy in South Korea
Where most taxes on wealth are intended to raise money for debt relief or extraordinary spending, the South Korean Land Reform Bill of 1950 sought to change property ownership in Korea from its historical, semi-feudal, tenant economy to a more egalitarian system (Morrow & Sherper, Reference Morrow and Sherper1970: 25). At the end of the war, some 70 per cent of all farmers in South Korea were tenant farmers paying more than half of their overall crop to aristocratic landlords (Pak, Reference Pak1956: 2). The American occupiers and newly installed government, like their counterparts in Japan and Europe, feared the growing threat of the Soviet Union and its influence on the working classes. Ending widespread tenant farming was believed to be necessary to curb class conflict. By pursuing an aggressive policy of land redistribution, the US-allied South Korean government sought to retain the support of the tenant class (Morrow & Sherper, Reference Morrow and Sherper1970: 25). The new Korean Constitution thus mandated land reform to improve the condition of the farmers and increase overall agricultural productivity (Pak, Reference Pak1956: 2).
In addition to the overall political aims, the programme sought to benefit both the agricultural and industrial economies by the transfer of and compensation for land (Pak, Reference Pak1956: 2). A farmer who merely rented the land, it was argued, had little incentive to invest his savings in its improvement. More productive land would likely only be met with increased rent. Ideally, by giving the tenant direct ownership, clear incentives for land improvement would be created, resulting in an overall increase in agricultural output. By compensating former landlords for the loss of their land, the Korean government hoped that the new capital would be invested in the industrial sphere (Pak, Reference Pak1956: 26). In this way, the level of agriculture would be maintained while emerging Korean industry would be funded.
The Final Bill was passed in March 1950 and contained three main features. First, owners of agricultural land were required to cultivate the land themselves. Secondly, the maximum amount of land a single person could own was set at just under 3 hectares. Thirdly, tenancy and the renting of agricultural land were permanently prohibited. The land reform itself was relatively simple. After completing a nationwide survey of agricultural land in June 1949, land was purchased from the landlords with redeemable bonds and sold back to the cultivating tenants for payments in kind, usually unprocessed rice (Jeon & Kim, Reference Jeon and Kim2000: 3). The ‘survey’ was completed in less than a year as the size and value of most pieces of land were taken from the records of the Japanese colonial government. The final price of the land was determined by averaging annual crop yields, discounted by 40 per cent to account for decreases in productivity since the Japanese occupation (Morrow & Sherper, Reference Morrow and Sherper1970: 28).
Those chosen to receive land under the programme were selected by a priority list (Shin, Reference Shin1976: 9). The first to receive land were those who had actually been cultivating it at the time the law was enacted. They were followed by freeholders of small land plots and citizens with agricultural experience. In practice, most of the land ended up simply being given to those who were currently working it. In just the first two years, a total of 331,766 hectares of farmland was redistributed to 918,548 households (Morrow & Sherper, Reference Morrow and Sherper1970: 30). Redistribution of the land was completed by the 1960s with most compensation for landlords being completed by 1962 (Jeon & Kim, Reference Jeon and Kim2000: 3). The final bond payment took place in 1969, about twenty years after the land reform process began (Morrow & Sherper, Reference Morrow and Sherper1970: 30).
Perhaps the largest difference between the Korean experience and other countries was the immediate influence of the Cold War. The nationwide survey of landholdings for redistribution began in mid-1949, with the official budget being passed on 27 April 1950 (Morrow & Sherper, Reference Morrow and Sherper1970: 27). Less than two months later, the Korean War began. However, while the loss of the capital city of Seoul forced a postponement of the programme until its recapture in September 1950, the programme was implemented during the conflict and likely had a major effect on the outcome. Buyers of the redistributed land were required to pay the government back in kind, rice being the primary staple of Korean military provisions (Morrow & Sherper, Reference Morrow and Sherper1970: 28). Repayment from the new landowners amounted to 1,158,780 metric tons of rice by 1952, a time when the new Korean government was fighting for its survival (Morrow & Sherper, Reference Morrow and Sherper1970: 29). The land reform programme was thus an immediate success in terms of its political objectives. Land redistribution resulted in a total of 577,000 hectares, or one-third of all Korean arable land, being taken from landlords and sold to the tenants (Morrow & Sherper, Reference Morrow and Sherper1970: 30). The number of freeholding farmers increased to 1,812,000 in 1950 from 349,000 in 1949 (Jeon & Kim, Reference Jeon and Kim2000: 3), with farm tenancy becoming virtually non-existent.
The Comparative Advantage of Capital Levies over Annual Wealth Taxes
For most of the twentieth century, the central goal of the wealth tax in Europe was to repay state debts (most commonly war debts), alternatively to address either economic inequality more broadly or a specific economic, political or social crisis, such as the needs of displaced war refugees and those who lost their property due to war – the war-damaged. In Asia, wealth taxes in Japan and Korea served quite different purposes, although they were also imposed in conflict or post-war contexts. In the case of Japan, the decision to impose a high capital levy on the Zaibatsu was justified both on economic grounds and, perhaps more significantly, as a means of securing democracy. In Korea and Taiwan, the land-to-the-tiller land reforms of the 1950s served to redistribute wealth (granting opportunities to tenant farmers to own land) and to direct capital investment into industrialisation. Despite these diverse histories, capital levies have shared a common set of goals – debt relief, sharing the burden of significant economic and social crises, constraining inequality and securing democracy.
Annual wealth taxes seem, by comparison, to be mostly geared towards raising revenue and reducing inequality. Implicit is an assumption that the expenditure of this revenue will be for the benefit of the less fortunate through the funding of social welfare programmes. While this general assumption may have justified annual wealth taxes in European social democracies, the diffuse nature of the benefit has meant that unless left-leaning political parties were in power and defended the programme, governments found it relatively easy to abandon annual wealth taxes, especially if the revenue stream was rather modest.
An alternative approach, more common in the case of capital levies, was to tie the income stream to specific expenditures or spending goals. Thus, the Finnish capital levy was directly tied to compensation for refugees, while the German Lastenausgleich both provided aid to the war-damaged and created a significant fund for reconstruction, particularly for housing. Thus, when considering the objectives of wealth taxes, it is important to distinguish between the different revenue goals as well as plans for the expenditure of the revenue raised by the tax.
While justification for many of the capital levies imposed in the early twentieth century was to address public debt, the imposition of annual wealth taxes was often justified in terms of constraining inequality and raising revenue. Significantly, however, the comparative history demonstrates that annual net wealth taxes do not manage to collect large amounts of revenue as compared, by percentage, to other taxes collected in the jurisdictions studied. Furthermore, annual wealth taxes do not seem to have any significant impact on the distribution of wealth (Wijtvliet, Reference Wijtvliet2014), although if continued over decades, there is some evidence that the degree of inequality may be moderated. In comparison, the imposition of capital levies does seem to have addressed some of the articulated goals justifying the use of wealth taxes as opposed to other fiscal mechanisms.
To secure their goals, the legal frameworks for different wealth tax programmes address a similar range of administrative and legal issues. Among the most ubiquitous issues facing the implementation of wealth taxes are defining the tax base, the valuation of wealth and the relationship to other forms of taxation. There are also concerns about the cost of administration and the likelihood of evasion or tax avoidance. Finally, there is a question, especially in the case of capital levies, whether the revenue should be earmarked for specific purposes or simply be used to pay down the public debt. By exploring the comparative historical experience, we can identify the issues and modalities that need to be considered in constructing and adopting a proposed transformational tax for South Africa.
A Transformational Tax for South Africa?
Instead of focusing on an annual net wealth tax – which has been shown internationally not to produce much income, or reduce inequality, and possibly increases capital flight and tax avoidance – this proposal is to adopt a one-off post-apartheid capital levy or transformational tax to address the continuing legacies of colonialism and apartheid. Furthermore, when considering the adoption of a transformational tax in South Africa today, we need to be very clear about both its purpose and normative basis. There are four main justifications for adopting a transformational tax or capital levy in South Africa. First, there is agreement that South Africa remains a highly unequal society, particularly when it comes to wealth. While the top 10 per cent of earners may now include 40 per cent black Africans and 48 per cent whites (using the standard South African government categories), when it comes to wealth, the distribution is even more skewed, with the richest 4 per cent earning over R750,000 per annum in 2014 and the top 1 per cent controlling 95 per cent of personal financial assets (Makgetla, Reference Makgetla and Khadiagala2018). Second, despite the Truth and Reconciliation Commission concluding that apartheid was a crime against humanity, the question of reparations for that crime has never been addressed. Third, the notion that the market for land, and hence market value, is neutral belies the fact that since at least 1913, this market was reserved for less than a fifth of the population. While the Constitution of the Republic of South Africa, 1996 (Constitution) provides for land restitution for those who were dispossessed, it is the constitutional duty to engage in land redistribution that must address this broader process of economic exclusion from the land. Finally, since ‘a tax is always more than just a tax: it is also a way of defining norms and categories and imposing a legal framework on economic activity’ (Piketty, Reference Piketty2014: 520), the effect of a significant surcharge on income (which would result from the imposition of the tax on wealth, since payment of the tax will come primarily from income), should produce a change in lifestyle choices that will reduce the conspicuous consumption that only highlights inequalities in the society.
With these explicit premises, it is now possible to imagine a one-off transformational tax to build a legitimate post-apartheid economic foundation, one that addresses two significant questions: who should be compensated, and who should pay? While there has been increasing discussion of the need for a new social compact, there is unlikely to be willing agreement on the imposition of a wealth tax. Instead, we need to understand the imposition of a transformational tax three decades after the dawn of democracy as a ‘democratically imposed social compact’ designed to address the specific legacies of apartheid that are undermining the very legitimacy of the constitutional breakthroughs of 1994 and 1996. While overall inequality in access to income, education, employment and other social criteria need to be continually addressed using the regular budget, it is the failure to advance both land redistribution and urban reconstruction that this proposal targets. With the poorest South Africans still locked in the former ‘bantustans’ and the provision of Reconstruction and Development Programme housing seen to be exacerbating geographic apartheid in our towns and cities, there is a clear need for a dedicated process to fund and address these sources of inequality. Especially in urban areas, the need for investment in infrastructure must be tied to overcoming the legacies of geographic apartheid and the lack of affordable housing that continues to shape, undermine and erode the sustainability of our system of democratic and constitutional governance.
While the informal ANC proposal of 1991 for a wealth tax on existing landowners to cover compensation for land expropriation for the purposes of land redistribution is too constrained to serve present conditions, existing levels of inequality mean that the tax would still fall primarily on the beneficiaries of apartheid. Even if there is now a small group of black South Africans whose wealth would reach beyond the proposed threshold for the tax, adopting a strictly racially based tax would be inconsistent with the country’s constitutional vision. Given the very small number of black South Africans who have actually accumulated significant wealth and the growing concern that entrenched and increasing inequality will undermine the democratic and constitutional project, it seems only just that a transformational tax should be based solely on a criterion of net wealth. Given both the need to address the legacies of apartheid and to create a more equitable and sustainable society, it does seem possible that we might today, in the aftermath of the great recession, state capture, COVID-19 and the attempted 2021 insurrection, achieve greater agreement or at least acceptance of the need for a transformational tax.
Imagining a Transformational Tax
How may we use the comparative experience with wealth taxes over the last century to best design a transformational tax for South Africa that addresses both the problem of inequality and the concerns of those who, like the Davis Tax Committee, argue that wealth taxes are not really effective? Comparing the historical experience of annual net wealth taxes with those situations in which significant capital levies were imposed demonstrates that one-off capital levies are significantly more effective in raising revenue, breaking concentrations of wealth and promoting democratic goals. There is, however, an important caveat, and that is the fact that significant capital levies have only been imposed in circumstances in which the political opposition to such an intervention is cowed either by the extent of the crisis or by a foreign force, such as the occupation powers in Japan and West Germany, which were in support of the tax. Lacking such circumstances, the only means of securing a significant capital levy, even if there is real democratic support, will be for the wealthy to accept that solidarity in the face of social and economic catastrophe will be the best means of maintaining a social compact that will secure their futures as well as those of the community more broadly. COVID-19 and climate change, as well as the continuing challenge to the legitimacy of the post-apartheid constitutional and economic order, like the collapse of the Iceland economy in 2010 (Philip et al., Reference Philip, De Mooij, Matheson and Micheielse2011), may provide just such a circumstance.
If this is the case, what are the modalities of a transformational tax that will ensure an effective capital levy that can be used for the reconstruction of the physical and social infrastructure and economy that will address the legacies of apartheid? From a review of the historical comparative cases, there seem to be six crucial design elements. First, any transformational tax will need to define the tax base to include all forms of wealth measured globally in the same way the present US tax system includes all individual income from whatever source. Secondly, while a transformational tax should set a high exclusion amount, for example, over R5 or R10 million, it should not create categorical exclusions as to forms of wealth.
Thirdly, when it comes to valuation, the great benefit of the one-off capital levy is that there is no need to conduct continuing processes of evaluation since the law can designate a date – for example, 27 April 2019, the twenty-fifth anniversary of the 1994 election or any date prior to the adoption of the tax – and use the market value as of that date. To ensure honesty and prevent the hiding of wealth, there are two interesting legal mechanisms derived from past experiences. One is that any property not declared would be forfeited to the state if discovered. The other is that if the owner of property declares a value that is later discovered to be significantly below market value, the state would be free to either purchase the property at the declared value or place the property on the market at the declared value.
The tax should be imposed on a sliding scale on all wealth as recorded on the date selected. The record of wealth may be based on submissions from the taxpayer (a tax form that offers the opportunity to record all assets as of the relevant date) and checked against the existing government and private data, including property values contained in local government rates records, banking information on mortgages and accounts, insurance company records and prior tax returns. Since this data is already in the system, there is little room for either capital flight or the hiding of assets. The tax would apply to both family wealth and legal entities, thus avoiding the difficulty of capital being distributed through various legal entities such as trusts, shares or other forms of capital holdings. The potential resources from such a tax will not be insignificant since, for example, in 2015 the Annual Financial Statistics reported by Statistical Services South Africa indicated that total company assets in the formal sector amounted to R8 trillion with a GDP of R4 trillion per annum.
Fourthly, to ensure the two central goals of a transformational tax, a significant revenue stream and the liberating of democratic politics from gross inequality and the influence of wealth, the tax rate will also need to be high. In the case of the German Lastenausgleich, it was set at 50 per cent, while in Japan, the rate was set in relation to overall wealth and reached as high as 90 per cent for the top bracket. In Finland, where the tax was indeed an act of solidarity, it was set at 40 per cent. Under present conditions of extreme inequality, it seems that a graduated scale would be most effective since the top 1 per cent now holds extreme amounts of wealth and economic power.
Fifthly, another benefit of the one-off capital levy over the annual net-wealth tax is that there is little opportunity for either tax avoidance or evasion. Capital flight is less likely in a situation in which the amount owed has already been defined, and the only question is how it will be collected. Some economists have argued that the threat of repeated ‘one-off’ capital levies will mean that there is a decline in savings and thus a threat to future economic prosperity; however, there is little evidence of this in the historical record.
Finally, any design of a transformational tax will need to consider whether the revenue generated will simply flow into government coffers or whether it will be effectively earmarked for specific needs. As already indicated, among the continuing legacies of apartheid the obvious target for spending these funds will be, on the one hand, to promote agrarian reform and, on the other, to address urban reconstruction to transform the geographic and social order of our cities and towns. Exactly how these resources will be allocated and whether they should be used as no-interest loans or grants are choices to be considered. While treasury departments across the globe argue that earmarking limits government expenditure choices and is thus to some degree undemocratic, it is important to consider two aspects of this debate. On the one hand, a transformational tax will not be the only source of government funding since it will not replace regular forms of taxation that need to be progressive to prevent a recurrence of the gross inequalities the tax is designed, in part, to address. To this extent, regular government expenditures will remain subject to regular democratic and constitutional procedures. On the other hand, the legitimacy of a transformational tax and the renewed social compact it seeks to establish is that expenditures will address the social and economic conditions that justified the imposition of the tax in the first place.
Open access