Tax equity around the world: a discussion

In his paper ‘Globalisation, taxation and inequality’ in this issue, Gabriel Zucman argues that the current tax system is not appropriate for our times. In particular, the system heavily taxes labour and consumption and largely exempts capital and capital income from taxation. This tax structure is at odds with rising inequality and the high capital share of income for the rich.

Zucman acknowledges that the international community has made progress towards limiting international tax competition and evasion and avoidance, thereby facilitating the taxation of capital. This includes the automatic exchange of information about financial accounts and investments, and the OECD/G20-mediated ‘Pillar 2’ agreement on a global minimum tax on the profits of the largest multinational corporations. However, he argues that these recent global measures on tax avoidance and evasion do not go far enough.

First, Zucman advocates for an improvement and extension of the automatic exchange of information. The system should be expanded to other assets, most importantly real estate, and countries should more systematically deploy the information they receive through the automatic exchange of information in tax enforcement. In addition, the system should expand to cover more countries, especially developing countries, and have an inbuilt mechanism to ensure accurate reporting.

The case for expanding the exchange of information to real estate is clear: the ownership of foreign real estate is a form of wealth that is extremely concentrated at the very top and growing in importance since the automatic exchange of banking information. However, exchanging real-estate information internationally in a coherent manner could be hugely challenging. The exchange of banking information requires information from banks, which are typically thought of as well-organised and technically advanced institutions, that have long had to adhere to international norms, standards and IT systems. Records of real-estate ownership, in contrast, are maintained by local government agencies which typically have less capacity than central governments. Besides, in the context of decentralisation, the format of these local property registers and the manner in which they are maintained can vary widely across municipalities or states.

While it may seem worthwhile to start an exchange of information focused on high-income countries and/or large cities that are attractive targets for real-estate investment, investors are likely to respond by diverting some of their investment towards assets in locations not covered by the exchange of information. After all, Bomare and Le Guern Herry (2022) have shown investors respond to the exchange of banking information by shifting into real estate, so they are likely to also be able to shift across real assets in different locations. The exchange of information on real estate would also need to be combined with a ban on the use of shell companies, so that ownership information actually allows governments to identify the ultimate owners of properties. To be clear, we do not think the challenges we outline are good reasons to stick to the status quo of exchanging only banking information. If anything, the challenges involved in exchanging real-estate information mean that the international community should attempt to tackle them sooner rather than later. We are merely suggesting to be mindful of the length of the process required and cautious about the potential tax revenue gains.

In addition, Zucman highlights that it is important for countries to use the information they already have in their tax enforcement strategies. Several studies document evidence for increased tax compliance in response to the exchange of information.1 But how large the revenue gains can be once the information is fully exploited and investors’ responses have been realised, and how much lower-income countries in particular can gain, is still uncertain. The cost of using the information in enforcement is likely higher in lower-capacity administrations, hence limiting the potential net revenue gains for them.

In the second part of his paper, Zucman advocates for the design and adoption of modern wealth taxes. The predominant form of wealth taxation is currently the property tax, an ‘archaic and regressive wealth [tax]’. Zucman argues that countries should instead adopt a tax that covers all forms of wealth, at its market value, and that allows taxpayers to deduct debt. Conditional on having a wealth tax, it makes sense that the base for this tax includes all sources of wealth, at market value. However, it is worth pointing to the ongoing debate about the desirability of a wealth tax. The UK Wealth Tax Commission website2 and the Fiscal Studies symposium on the wealth tax3 provide a fantastic set of evidence papers on wealth tax design in theory and in practice, drawing on the experience of seven different countries. In particular, Adam and Miller (2021) discuss the economic arguments for and against a wealth tax. They conclude that the case for the one-off wealth tax is simple to make (because the opportunities for a one-off tax, implemented promptly, to distort economic activity are limited), but the case for an annual wealth tax is more difficult – though not impossible – to make based on economic theory (in part, because of the greater scope for economic distortions in response to a recurrent tax).

Zucman also argues that the (annual) wealth tax should have a progressive schedule with high exemption thresholds. He contrasts this progressive wealth tax with the ‘archaic’ property tax typically levied at a flat (or even regressive) rate. While the property tax may be archaic in many ways, it is worth keeping in mind that some jurisdictions have property taxes that – at least on paper, and for the tax base they cover – are mildly progressive. For instance, the property tax in Mexico City is progressive for high-end properties.4

While wealth taxes in the past had ceiling mechanisms, such that the wealth tax liability could not exceed a fraction of income, Zucman suggests that such mechanisms are not needed if the wealth tax is applied only above a very high exemption threshold. However, even very wealthy individuals may face constraints on their ability to pay an (increased) wealth tax if they have committed consumption and illiquid forms of wealth. If these constraints render them unable (or less willing) to pay, the success of the wealth tax and the political support for it could be undermined. It is hence important to think through the details of how to ensure liquidity to pay and hence the enforceability of a wealth tax.

Finally, Zucman argues that the use of pre-populated returns for the modern wealth tax could dramatically reduce the scope for tax evasion. While we think that pre-filling returns is a worthwhile policy to experiment with, we would caution that pre-filling does not necessarily increase tax payment.5 The effect of pre-populated returns may depend on how complete the government's information set is. If it is very incomplete (or more incomplete than taxpayers expect), pre-populated returns might create a reference point for taxpayer reporting that could be lower than what they would have otherwise voluntarily reported. In a way, a pre-filled return can reveal to the taxpayer how much (or how little) a government knows about the tax base. The automatic exchange of banking information, and hopefully of real-asset information in the future, will be crucial in helping to populate the returns.

A key tenet of Zucman's article is that ‘tax competition, tax evasion and tax avoidance are not laws of nature’, but instead ‘policy choices’, and that alternative choices are possible. This makes sense, but one might ask how technically and legislatively complicated it would be to implement these policy choices, and what political and economic costs they are associated with. For example, Zucman argues that financial assets are not necessarily highly elastic (internationally mobile) to tax rates, as ‘elasticities are not immutable parameters: they are affected by tax design’. While we agree with this view, which has also been documented in numerous studies,6 we think it is important to also take into account the trade-offs that arise when designing policy with the aim that ‘elasticities can be reduced, possibly to very low levels’. A welfare-maximising policymaker would adopt such a policy if the benefits from reduced tax evasion and avoidance, and hence increased public spending, outweigh potentially adverse real effects. It is possible that these real effects are small, but it may not necessarily always be optimal to drive elasticities to ‘very low levels’.

Before concluding, it is worth briefly touching upon Zucman's comment that the global minimum corporate income tax is a ‘first and important’ step towards the international regulation of globalisation and the setting of tax rates (as opposed to simply tax bases). We agree that the commitment to a global minimum tax is an exciting achievement. But it remains to be seen how significant the revenue gains are, and to what extent they benefit low- and middle-income countries.

As discussed in Perry (2023) in a recent Fiscal Studies symposium on the global minimum tax, many low- and middle-income countries (LMICs) have structured their economic development and investment promotion strategies around tax incentives and exemptions for multinational corporations and enterprises (MNEs). Honduras's development model is one example, as discussed in the recent World Bank conference and in a TaxDev blogpost.7 A concern with the early proposal for a global minimum tax (GMT) was that it would prevent competition on real economic location choices. In response to this, the current version of the global minimum corporate income tax allows ‘carve-outs’, i.e. reductions from profits based on payroll costs and tangible (capital) assets. The countries where MNEs within scope of the GMT are headquartered have the first right to impose a top-up tax on affiliates of these MNEs that have an effective tax rate below the global minimum. This means that LMICs that currently rely on tax incentives to attract investment may lose out if they take no action to reform their tax system: if they host the affiliates but not the headquarters of MNEs in scope, they would be less likely to attract investment and would not benefit from the top-up tax payments. Countries are hence working to adapt their development strategies and tax systems. Planned reforms in Honduras under its ‘Tax Justice Bill’, for example, partially replace profit-based incentives with double deductions for payroll costs for permanent workers and accelerated depreciation for capital investment to more closely align its tax incentives regime with the structure of the global minimum tax. Countries will also need to consider whether to implement a qualified domestic minimum top-up tax (QDMTT). The aim of a QDMTT is to ensure that countries that host MNE affiliates, rather than the countries where the parent MNE is located, receive the top-up tax payments.

Zucman's reference to the global minimum corporate income tax being a ‘first step’ also begs the question as to whether there is scope for international coordination on tax rates applied to other tax bases. In the case of wealth, our view is that the greater diversity of opinion on whether wealth should be taxed on an annual basis at all, and major differences in wealth distributions across countries, will preclude the forging of a consensus on a minimum tax rate.

None of the above is to question the importance of the issues and debates that are raised by Zucman. The recent World Bank conference on global tax equity, and the papers from participating policymakers in this symposium, illustrate the action a growing number of countries are taking, and the challenges and opportunities they are facing, as they seek to increase the contribution of MNEs and high-net-worth individuals to revenues, and tackle tax avoidance and evasion. An increased focus on international cooperation, both on high-level policy (such as the global minimum corporate income tax) and administration and enforcement activities (such as automatic exchange of information) may be the foundation for further change. However, with regard to some of the more far-reaching proposals in Zucman's paper, our view is that more research, and particularly more empirical evidence on quantifying the key trade-offs involved, is still needed in order to allow policy to be as informed as possible about the full potential consequences on behaviour and, in turn, revenues.