How expensive is a GND?

In early 2019, the US House of Representatives articulated the goals of a GND as ensuring all people in the United States have access to health care, housing, economic security (through a family-supporting wage rate or income), food security, clean air and water, and access to nature. The goals are addressed through upgrades to the transportation system and energy grid and through the introduction of a universal healthcare system, programmes for green housing and food security, and guaranteed jobs. The suggested dates of implementation are over a 10-year period, at the time from 2020 to 2029 (US House of Representatives, 2019).

The American Action Forum (AAF) estimated the expenditures associated with these goals over a 10-year period as approximately US$ 93 trillion or about US$ 670,000 per household. The estimated cost of a low-carbon energy grid is US$ 5.4 trillion (US$ 39,000/household); a net zero emissions transportation system costs US$ 1.3–2.7 trillion (US$ 9000–20,000/household); guaranteed jobs programme costs US$ 6.8–44.6 trillion (US$ 49,000–332,000/household); universal health care costs US$ 36 trillion (US$ 260,000 per household); guaranteed green housing ranges from US$ 1.6–4.2 trillion (US$ 12,000–30,000/household) and food security is an estimated US$ 1.6 billion (US$ 10/household), (Reference Holz-Eakin, Bosch and GitisHolz-Eakin et al., 2019). The staggering figure of US$ 93 trillion, the highest so far, gives one pause about the feasibility of a GND.

Reference Nersisyan and WrayNersisyan and Wray (2019) revisited the figures to account for the effects of feedbacks between the components and savings obtained if the States adopted different policy stances. Their approach emphasises the mobilisation of resources for a GND, recognising that utilisation of capacity varies across industries and better allocation of resources, particularly underutilised resources, can yield savings. For instance, the projects involving upgrades to infrastructure and energy sources will entail the employment of workers who are part of the jobs guarantee programme. The two components cannot be evaluated on a stand-alone basis, as with the AAF estimation, which leads to double-counting costs. Based upon simulations from the Levy Economics Institute, the estimated resource cost of the job guarantee component is US$ 200 billion annually or about 1% of GDP (Reference Nersisyan and WrayNersisyan and Wray, 2019). The green initiatives were estimated to be 5.6% of GDP or US$ 1100 billion, determined as an average of estimates reviewed by the authors. Together, these two components amount to US$ 1300 billion per year or 6.6% of GDP. Healthcare – Medicare-for-All – enters as a negative figure of −3.7% of GDP, as savings can be obtained by reducing administrative overhang involved with healthcare plans and insurance coverage, fraud reduction, waste elimination and price reductions on pharmaceuticals. Other savings could be found by ending ‘forever wars’. Additional funds could be obtained by taxing the rich via income tax changes and implementing a payroll tax surcharge. In all, Reference Nersisyan and WrayNersisyan and Wray (2019) suggest that the resource costs of a GND could be reduced to 1.7% of GDP (without payroll tax surcharge) or even −0.7% (with tax surcharge). After accounting for feedbacks, the most expensive component, in their view, is the green projects (US$ 11.1 trillion over 10 years).

There are other proposals for a GND, and the cost or resource estimates for each will depend on the configuration and assumptions made. For space considerations, this article cannot provide an overview of the variations (see Reference Galvin and HealyGalvin and Healy (2020) for a history of GND proposals and discussion of recent cost estimates). Rather, this contribution will focus on the green projects’ component since the range of proposals have this at their core and this component will be the most expensive. It is also the area in which President Biden’s de-carbonisation policy will emerge as the starting point to address climate change. Recent cost estimates of green initiatives range from Biden’s suggested US$ 190 billion/year to US$ 1.0–1.1 trillion/year (Reference Holz-Eakin, Bosch and GitisHolz-Eakin et al. (2019), Reference Nersisyan and WrayNersisyan and Wray (2019), Reference Pollin, Garrett-Peltier and HeintzPollin et al. (2014) and Bernie Sanders (as elaborated by Reference Galvin and HealyGalvin and Healy (2020)). Again, if a wealth tax is feasible at the high-end of the cost spectrum, say, US$ 1.1 trillion/year, then it will have credibility for proposals at the low end, where the Biden proposal sits.

We begin with the trajectory of national debt without the incorporation of any element of a GND. According to the CBO, the projected national debt (held by the public), debt–GDP ratio, net interest outlays and interest rate projections are located in Table 1.

Table 1. Estimates of national debt, interest outlays and interest rates, United Sates, 2020–2030 (billions of USD).

a actual.

Source: CBO (2020).

Between 2021 and 2030, the national debt is expected to increase approximately 48.4%, from US$ 22.5 trillion to 33.3 trillion. The debt-to-GDP ratio is expected to rise to 105.0% from 102.3%. Net interest outlays will increase approximately 129.4% between 2021 and 2030, where the total estimated amount due over this period is US$ 4.057 trillion. From the current low rates of 0.16% on 3-month Treasury bills and 0.83% for 10-year treasury notes, both the short- and long-term (LT) rates are expected to rise to 1.96% for the short-term rate and 3.15% for the LT rate by 2030.

Notice that the net interest outlays vary over the projected 10-year range. The wealth tax needs to be such that it generates revenues which are consistently greater than those outlays. Anything more than net interest outlays can be used to slow the growth of debt and/or finance green initiatives. Slowing the growth of new borrowing, or reducing the reliance on new borrowing, could potentially stabilise, if not reduce, the debt/GDP ratio. Controlling the ratio, moreover, will help keep austerity measures at bay and contain financing costs on new borrowing.

A Kaleckian wealth tax

The post-Keynesian literature is noted for its lack of discussion and recommendations regarding a wealth tax. Rather, one often finds Keynes’ suggestion in The General Theory of Employment, Interest and Money (1936) of euthanising the rentier to strengthen the marginal propensity to consume and economic growth (Reference Zorn and PressmanZorn and Pressman, 2020). The late Michal Kalecki (1899–1970), a Polish economist whose work has a tremendous influence within the post-Keynesian community, argued years ago that financing public expenditure needed to occur without distorting the incentives to invest or consume. He thought it was possible to formulate a capital or wealth tax in such a way that national debt is sustainable and do so equitability where everyone pays the same rate on their assets. An income tax, in contrast, will likely fall heavily on the earnings of waged and salaried workers, and earnings of businesses. Their taxes will pay the interest on debt held by financial wealth holders. Their incentives to consume and invest are weakened, thus, reducing aggregate demand. We are interested to go beyond Kalecki’s insights by adding the financing needs of green initiatives to the national debt and the tax calculation and, suggest what to do when a country experiences negative interest rates.

The idea behind an equitable capital or wealth tax is to set a tax on privately held assets (the value of capital) in the same proportion as the interest rate on the national debt:

$\text{iD = tK or t = i}\left(\text{D/K}\right)$

where i is the rate of interest, D is the national debt, K is total national wealth (privately held) and t is tax. (Notation after Reference NellNell (1988)). Assets include plant and equipment, real estate, savings accounts, bonds, stocks – any form of wealth (privately held). Investment or capital allocation, and, thus, production and employment, are not distorted as the tax is applied to all forms of wealth. Moreover, the propensity of wealth holders to consume is also unaffected. If government borrowing increases, the interest income to wealth holders increases. The capital tax increases accordingly to cover the additional income, (Reference NellNell, 1988: 247). As the tax would apply to gross assets, and not net assets, debt cannot be used to lower one’s tax liability.

The following application is based upon the United States because of data availability. Balance sheet data for the sectors can be located at the Federal Reserve Economic Data (FRED). Data for households and non-profit organisations are in Table B101 (code TABSHNO), non-financial non-corporate firms are in Table B104 (code TABSNNB), non-financial corporate firms are located at Table B103 (code TABSNNCB) and domestic financial sectors are in L108 (FBTFASQ027 S). Annual data series were used (average for the year). Total private assets data were constructed for years 1990–2020, where each year’s observation is the sum of the corresponding observations of each sector (the data are available in Table A of the Online Supplementary File). The observations for total private assets in future years were compiled by assuming total assets grow at their average historical rate of 5.99% (1990–2020) and are noted where appropriate in subsequent tables.

At the end of 2020, the US national debt owed to the public is estimated to be approximately US$ 21 trillion. The data in this paragraph is employing material from table A. The size of both household and non-profits assets was US$ 137.801 trillion; the assets of non-financial, non-corporate firms was US$ 21.892 trillion; non-financial corporations’ assets were US$ 47.138 trillion and assets of the domestic financial sector was US$ 115.367 trillion. In all, privately held assets amounted to US$ 322.199 trillion. Under these conditions, what is the capital or wealth tax? Let us assume the interest rate is 2%. iD = tK implies 2% × US$ 21 trillion (debt) = t × US$ 322.199 trillion (assets), so that t is 0.001304% or 0.1304%. A less than 1/4 percentage point tax on assets will generate revenue of 0.001304 × US$ 322.199 trillion = US$ 420.147 billion, which is greater than the projected net interest outlay of US$ 345 billion. This is promising. However, interest rates fluctuate. To help ensure the tax generates an excess over interest outlays, the calculation could utilise the largest forecasted interest rate over the time-period. Using the top-end of the 10-year treasury note, 3.15%, means in our previous example the wealth tax becomes 0.2053% and generates revenue of US$ 661.5 billion. US$ 316.5 billion remains after subtracting the net interest outlay. If private assets do not grow, t starts at 0.002053 and gradually rises to 0.003256 in 2030 (see Table 2).

Table 2. Wealth tax calculation United States private assets (no growth), 2020–2030.

Source: Author’s calculations.

However, private assets grow and tend to grow faster than the national debt. The national debt held by the public is estimated to grow approximately 58.4% between 2020 and 2030, whereas private assets are estimated to grow nearly 78.9%. If we recognise this growth, the wealth tax declines over this period, 0.002053–0.001820 (see Table B in the Online Supplementary File). If debt and assets grew similarly the tax rate is stable, or changes with the interest rate. If the interest rate is stable, the wealth tax becomes less of a burden if assets grow faster than the growth of the national debt.

Because of the variability in the wealth tax, the suggestion here is to select the highest tax rate over the forecast period for use. In the above example, the highest observation of the tax is 0.21%. The annual revenues from this tax are consistently higher than the expected net interest outlays, which means there is a consistent surplus to allocate to, say, slow the growth of new borrowing and contribute to green initiatives of a GND. The tax can be reviewed every 3–4 years for adjustment based upon revised forecasts. A periodic revision will facilitate adjustments if unanticipated events influence revenues; the initial example with 2020 data suggests the US$ 420 billion of revenues would be approximately 10%–11% of the total federal tax take.

What happens when we add the resources cost for the green initiatives of the GND? As discussed in the previous section, the average annual increase in expenditures for this component is US$ 1.1 trillion per year. If we continue to assume the interest rate (LT) over this period is 3.15%, the capital tax ranges between 0.2177% in 2021 and 0.1880% in 2030. If we adopt the highest tax in this period, 0.22%, the revenues over the 10-year period are US$ 9.898 trillion, which is greater than the anticipated net interest outlays of US$ 4.057 trillion. (see Table 3).

Table 3. Wealth tax calculation with green programmes, United States, private assets (grow at 5.99%/year), 2020–2030.

Source: Author’s calculations.

It is difficult to say exactly how the debt-GDP ratio will behave. While the national debt will increase, the multiplier effect on public expenditure will positively influence GDP. A multiplier effect of 1.5, for instance, implies that GDP (as per CBO (2020) projections of GDP ^{Footnote 1} ) will increase by US$ 1.65 trillion annually (US$ 1.1 trillion x 1.5) from this influence. The debt-GDP ratio hovers between 99.0 and 103.1, improving upon the CBO’s estimated range of 100.9 – 105.0 (without a GND). US$ 5.841 trillion remains after accounting for net interest outlays. US$ 2.5 trillion (US$ 250 billion per year) could be directed to slow the growth of new borrowing, lowering the range for debt-GDP further, and keep austerity programmes at bay. The remaining US$ 3.341 trillion would be directed towards green initiatives – paying entirely for Biden’s de-carbonisation policy with US$ 1.441 trillion left. Note that a shortfall of approximately US$ 7.659 trillion remains for the green projects.

This shortfall is not daunting when one recognises that the Kaleckian tax is meant to be an equitable way to achieve sustainability, and not meant to reverse wealth inequality. At this point, Warren’s tax on ultra-wealth holders could be introduced to yield additional US$ 3.75 trillion and reverse inequality. The remaining funds for green initiatives could come from, say, the savings to be had from implementing a Medicare-for-all programme as it frees resources from the administrative costs of healthcare and insurance plans, and controlling pharmaceutical prices under the current system. Those savings are estimated by Reference Nersisyan and WrayNersisyan and Wray (2019) to be 3.7% of GDP or approximately US$ 11 trillion over the 10 next years. Remaining funds could come, instead, from a carbon tax, financial transactions tax and/or income tax. The point is that resources are available to initiate an ecological transition. The Kaleckian tax could be an equitable first step for locating funds to keep austerity programmes at bay and initial funds for Biden’s de-carbonisation plan.

At present, several economies are experiencing negative interest rates, and forecasts for the next 10 years are difficult to ascertain. A suggestion here is to set tax on private assets at a rate which at least covers the largest annual interest outlay in the forecast window with an increase to avoid shortfalls and garner funds for initiatives. Working with data from the United States, the net interest outlay divided by the private assets yields a range for the wealth tax of 0.000740–0.001386. If the tax was rounded to 0.15%, revenues are not only consistently greater than net interest outlays but also generate a revenue of US$ 6.749 trillion over the 2021–2030 period (see Table C in the Online Supplementary File). The Kaleckian formula is a good start and can help guide the tax calculation under more severe circumstances.

An advantage of a wealth tax is there is no distinction between wealth that has been inherited, attained from personal effort and savings, obtained by luck or simply increased asset values (OECD, 2018: 51). A capital income tax, on the other hand, focusses only on those assets that generate a rate of return. A wealth tax can be applied to immovable and moveable assets and financial investments. Assets that do not generate an annual rate of return, such as artwork, are liable for the wealth tax; note, while artwork does not generate an explicit stream of returns it appreciates in price over time. A wealth tax also has the advantage of being more stable in its ability to generate revenue over volatile economic conditions. Rates of return on assets can fall during a downturn, and so the tax return generated from a capital income tax also falls to zero. A wealth tax, however, can generate revenue so long as assets’ values remain positive (OECD, 2018: 49).

Relevance and feasibility of a Kaleckian wealth tax

Given Kalecki’s vision for a wealth tax was expressed some time ago, does it hold relevance for how the world has changed over the last 30 years? A key driver of wealth inequality has been financialisation, the increased reliance on financial activities for some form of income. With respect to firms, liberalisation of trade and capital flows enabled producers in the United States and elsewhere to outsource production overseas. The strategy allowed US non-financial corporations, for instance, to increase profits and profit share. However, this did not lead to strong growth in domestic reinvestment in plant and equipment. Rather, investment in financial assets increased, as has the payment of dividends, stock buybacks and, mergers and acquisitions (Reference AssaAssa, 2016; Reference Milberg and WinklerMilberg and Winkler, 2009; Reference PalleyPalley, 2013). Reference StockhammerStockhammer (2004), Reference AssaAssa (2012), Reference Hein, Dunhaupt, Alfageme, Arestis and SawyerHein et al. (2017) have noted similar developments within OECD countries.

Financialisation is typically proxied by the contribution of finance, insurance, and real estate activities to GDP (value-added basis). Table 4 illustrates the trends of financialisation for OECD countries using the OECD National Accounts database. Financialisation in the OECD countries has, in general, increased since 1995.

Table 4. Financial services as a % of GDP (value-added), 1995 to 2018, selected OECD countries. ^a

OECD: Organisation for Economic Co-operation and Development.

Source: OECD National Accounts database

(a)The most recent year of data is 2017.

(b)The most recent year of data is 2016.

(c)Starting year is 1997.

(d)Starting year is 1998.

a OECD data are not complete or available for all countries.

Wealth distribution reflects the influence of financialisation, at least in part. Table 5 illustrates the recent observations of the share of (net) wealth owned by the top 5% and by the bottom 60% of households. The difference between the two is striking. The United States’ wealth distribution is the most severe – where the top 5% own over 68% of wealth and the bottom 60% owns a mere 2.4%. Germany (46.3% vs 6.5%, respectively), Austria (43.5% vs 8%) and Ireland (37.7% and 7.2%) are not far behind.

Table 5. Distribution of wealth shares for top 5% versus bottom 60%, selected OECD countries.

OECD: Organisation for Economic Co-operation and Development.

Year of observations in parentheses.

Source: OECD Wealth and Distribution Database.

Reference PikettyPiketty (2014) notes the importance of the relationship between income and wealth inequality. Wealth inequality is exacerbated by income inequality as wealthy individuals derive income through the returns from the capital or wealth they own. Reference Zorn and PressmanZorn and Pressman (2020) provided a theoretical explanation, based on Kalecki’s work, for the phenomenon documented by Piketty. In a closed economy, if capitalists save all their income and invest it, economic growth equals the rate of return on capital. Any spending by capitalists out of saving implies economic growth becomes less than the rate of return. The more capitalists spend the more growth stagnates and inequality accelerates. It can be expressed in the following way (notation after Reference Zorn and PressmanZorn and Pressman (2020)):

Savings by capitalists (S) is equal to some rate of savings (s) from profits (P):

$\text{S = sP}$

Investment is made from capitalists’ savings:

$\text{I = sP }\left(\text{= S}\right)$

This implies that profit is equal to investment divided by the rate of saving:

$\text{I = sP }\left(\text{= S}\right)$

The rate of profit (r) is determined by dividing through by capital stock (K):

$\text{r = P/K = }\left(\text{I/K}\right)\left(\text{1/s}\right)\text{,}$

and I/K is the pace of capital accumulation (g):

$\text{r=g}\left(\text{1/s}\right)$

Piketty observed r>g. This can be attributed, as per Zorn and Pressman, to capitalists not saving all their income, that is, they spend some of it. r>g because of that expenditure. The less they save the more sluggish (and unequal) the economy becomes. Zorn and Pressman do not mention an explicit role for public investment, nor government expenditure on goods and services. This scenario is akin to a balanced budget so that profits (net of taxes) is equivalent to capitalist investment and consumption. If the budget is in deficit, say, to support the creation of jobs, the private sector receives more from the government than it pays in taxes. In this context profit is no longer negligible of taxes, and one should consider how the budget’s position may influence profits. The first effect is an increase in employment, enabling profit to increase beyond the level determined by private investment and capitalists’ consumption. (New public expenditure has taken place without a reduction in spending elsewhere.) As expected returns increase, the inducement to invest is not only not compromised but will be stronger than before the introduction of the wealth tax. The inducement to consume by workers is also strengthened because of the increase in employment (Reference KaleckiKalecki, 1971 [1937]). Of course, ‘the counterpart to the budget deficit is the increase in debtedness of the government towards the private sector’, (Reference KaleckiKalecki, 1968: 48–52).

If the economy grows in a context where all wages are spent and all profits are invested (capitalists save), the rate of profit (r) is equal to the pace of capital accumulation (g). Assume that the economy grows, but there is a chronic budget deficit. The deficit leads to an increase in debt, and debt accumulates at the same pace as capital. Competition will ensure that the interest rate on government bonds (i) equals the profit rate, and hence, growth of capital: i = g = r. The interest on debt is equal to the growth of debt: iD = gD. The deficit can be financed by borrowing, and the wealth tax pays the interest rate. If everything grows apace, the situation can continue indefinitely (Reference NellNell, 1988: 247). The incorporation of Zorn and Pressman’s insights on Piketty suggests if capitalists’ savings fell, exacerbating inequality, then tK < iD and t must rise to maintain the ability to generate revenue to cover the interest on debt. Financialisation acts akin to capitalists’ spending out of savings. Capital accumulation and growth slowed, and inequality increased. Kalecki’s ideas and design of a wealth tax are more relevant today than ever.

The distribution of assets across the sectors suggests, initially, the burden of tax falls on households. The share of assets owned by households and non-profits is 42.8% (down from 46.0% in 1990); the share for households only is 41.5%. The assets of non-financial, non-corporate firms is 6.8% (vs 8.0% in 1990), whereas non-financial corporations’ share of assets is 14.6% (down from 18.0% in 1990). The domestic financial sector grew to 35.8% from 28.0% in 1990. The shift in shares is attributed to whether each sector’s assets grew more or less quickly than the growth for all assets, which was 472.8% or an annual average of 5.99%, between 1990 and 2020. The assets of households and non-profits grew 432% (5.73% average, annual growth), the assets of non-financial, non-corporate firms grew 386.6% (5.41% average, annual growth); for non-financial corporations assets grew 366.7% (5.3% average, annual growth); and the assets of the domestic financial sector grew 632.6% (6.9% average, annual growth). At first glance, it seems a bit odd that the asset growth of households and non-profits grew faster than non-financial firms (non-corporate and corporate). However, one needs to consider the evidence on profit shifting and wealth accumulation overseas, as constructed by Gabriel Zucman and his colleagues. Reference Saez and ZucmanSaez and Zucman (2020) estimate that nearly 60% of profits made by US multinationals are parked in low-tax countries, mainly Ireland and Bermuda. Moreover, the overseas assets of households are likely to be underreported; Reference ZucmanZucman (2013) estimated that 8% of global financial wealth of households is not reported, a figure that does not include overseas real estate holdings. The asset growth rates in certain sectors are likely be sluggish because of the influence of underreporting.

With respect to households, the growth of assets for the bottom 50% of households, according to wealth deciles, was 302.9% (average, annual growth 4.80%). At the end of 2020, assets for this group amounted to US$ 7.59 trillion, for a share of 5.95% of total household assets (US$ 138.99 trillion). The assets of the next 40% wealth percentiles grew 345% (average, annual growth 5.10%) to US$ 41.93 trillion, or 30.1% of total assets. The assets of the top 90% – 99% grew 491.5% (average, annual growth 6.10%) to US$ 50.22 trillion, or 36.1% of total assets. Finally, the top 1% saw their assets grow 762.6% (average, annual growth 7.00%) to US$ 39.36 trillion, or 28.3% of total assets. In 2020, the top 1% of households controlled nearly 30% of total household assets; and, together, the top 10% of households controlled nearly 65%. (Source of data is the Assets Breakdown Dataset from the Federal Reserve, St Louis.) ^{Footnote 2}

A recent OECD study finds the top 20% of personal wealth holders have the highest concentration of financial assets as a portion of their holdings (OECD, 2018: 51, Figure 2.6). High wealth holders often hold multiple housing properties, mainly for investment purposes, and are likely to have self-employed businesses. With respect to non-financial assets, the primary residence is the key holding for the mid-range of households. Those with less wealth are more likely not to own a home. Their real assets will be comprised vehicles and household effects and valuables. These features influence the resilience of households in times of severe downturns. The poorest deciles of households of OECD countries can weather 2–12 months of unemployment (no income) with savings and other liquid assets, whereas the highest deciles can weather years (Bank for International Settlements (BIS), 2020).

The state of inequality suggests thresholds and exemptions need to be considered in the final design. For instance, if the wealth associated with home ownership were removed from the stock of assets, how would the tax be affected? Owner-occupied dwellings for the US were an estimated US$ 31.095 trillion in 2020. If this entire amount were deducted from total private assets for 2020, private assets in the tax calculation becomes US$ 291.104 trillion and the wealth tax increases to 0.2272% from 0.2053% (national debt = 21 trillion and the interest rate = 3.15%). As a result, the share of households (and non-profits) assets becomes smaller than the domestic financial sector, 36.6% vs 39.6%, so the latter now bears the burden of the tax.

Under today’s global economic climate, it would be advantageous to stagger the implementation of the tax so that the top tiers of wealth holders are addressed first before shifting towards those with less wealth. This would enable those with less wealth to recover more fully from the pandemic before being asked to contribute. A threshold could be implemented on the size of asset holdings to attain before the tax comes into effect. In addition to an exemption on primary residences below some threshold, exemptions could also be made for household effects, working tools, and transportation vehicles of the poor. As far as implementing the tax, it could be deducted from income prior to the determination of income tax.

The OECD recently recommended a wealth tax entail a low rate, be progressive, have limited exemptions (such as, exemptions for personal and household effects to a certain limit), use market-determined valuations (updated every few years), allow payments to be made in instalments, enhance transparency, develop third-party reporting, create tax treaties to avoid international double taxation and monitor the effectiveness of the tax (OECD, 2018: 94). The Kaleckian wealth tax exhibits a number of these features. It is low (less than 1/2% on gross assets). It is progressive by weighing more heavily on those with large wealth holdings. It can accommodate exemptions for primary residences and their contents. It can employ market-determined valuations. Information sharing, enhanced transparency, tax treaties, third-party reporting are feasible and headway in some of these areas is already underway. The construction of comparable national wealth figures across countries will need to be addressed, however. There would need to be a mechanism to periodically update asset values. The state of information technology and its rapid development could support the development of this mechanism.

The effectiveness of the tax depends on several factors. Effectiveness, for instance, would be reduced if the tax triggered tax avoidance and capital flight. This issue is not of much concern to working- and middle-class individuals as much of their income consists of wages, pension, and investment income. Information about these sources of income is regularly reported to the Internal Revenue Service (IRS) and severely limits tax dodging (Reference Saez and ZucmanSaez and Zucman, 2020: 61). The elite, however, dodge taxes with the assistance of the tax-planning industry and the use of tax shelters, particularly in overseas tax havens. Reference Saez and ZucmanSaez and Zucman (2020) suggest there are probably millions of tax shelters in existence, which creates a layer of financial opacity which challenges tax authorities, regulators, and investigators’ ability to penetrate. Google, for instance, shifted its intellectual property to Bermuda, a known tax haven, by selling it to a shell company called Google Holdings. Moreover, assets and services, such as ‘logos, trademarks and management services often do not have an observable market value, (Reference Saez and ZucmanSaez and Zucman, 2020: 74). Reference ZucmanZucman (2014) has suggested the creation of a World Financial Registry to promote transparency in reporting assets and income by individuals, and an establishment of global cooperation to pressure tax havens to relinquish illegal activities. The IRS has managed to require foreign banks transmission of data and set a template for international cooperation. Advances made on information sharing and transparency between countries – such as Automatic Exchange of Information, Exchange of Information on Requests and information exchange agreements – make it more difficult to avoid taxes through relocation of assets (OECD, 2018: 67). Accountancy firms, such as PricewaterhouseCoopers, Deloitte, Ernst and Young and KPMG record transactions, suggesting that data reported to the IRS need to be expanded to include these firms to enhance the effectiveness of a wealth tax. The creation of multilateral rather than bilateral tax treaties could also facilitate taxation of overseas assets, above a threshold, by the home country of the wealth holder. Forensic accountants can be employed by the government to monitor for shifts and accuracy of information.

A wealth tax’s effectiveness also depends on how the tax interacts with other types of taxes, for instance, capital income tax. This interaction is often referred to as ‘stacking’. A capital income tax is levied on the returns of assets. Examples of capital income taxes include corporate income taxes and personal capital incomes taxes (rents, dividends, interest and capital gains). Wealth and capital income taxes can be interdependent. Increasing a capital income tax, for instance, means the discounted stream of returns from an asset will decline. As values of assets declines, wealth declines as do revenues from the collection of a wealth tax. The logic also leads to the idea that imposing wealth tax, or raising one in existence, leads to a reduction in wealth. This concern can be allayed when one recognises, as earlier noted, that the growth rate of assets more than compensates for the wealth tax. The Kaleckian wealth tax simply reduces the growth of assets and does not deplete the asset base. Interaction of wealth taxes with other taxes are somewhat inconclusive, and need additional research (OECD, 2018: 62–63). Reasons for inconclusiveness are due to differences in tax systems across countries. For instance, countries could implement wealth taxes at the national or local level. Taxes on primary residences are sometimes conducted at the local level where the resulting revenues are used to support school systems.

Another concern regarding effectiveness is whether wealth taxes could stymie entrepreneurship and new investment. The breadth of the tax base would ensure that all wealth holders are treated fairly, and thus, not influence investment. However, a wealth tax could be perceived as inequitable because it is invariable to changes in rate of return. The following is an example used by the OECD to compare capital income taxes and wealth taxes. Suppose there are two investors with (net) wealth of EUR 10 million. One invests his wealth for a rate of return of 4% (EUR 400,000). If capital income tax is 30%, the tax liability for this investor is EUR 120,000. The other investor would have the same tax liability if subjected to a 1.2% wealth tax. His assets could well yield 4%, too. But the tax is imposed irrespective of the actual returns on assets. In other words, ‘a net wealth tax implicitly imposes a lower effective tax on the return of high-yield assets compared to low-yield assets’ (OECD, 2018: 49). If the rate of return on investment for the first investor increases to 5%, the tax liability for the first investor increases to EUR 150,000, while the tax liability for the second remains the same. What is not accounted for in this example is the tendency for assets to appreciate in price. As noted earlier, the financialisation process has reinforced the tendency for asset prices to increase. A wealth tax will capture that and narrow the difference in tax liabilities and could even reverse who bears the heavier liability.