Are global taxes feasible?

Interestingly, Piketty’s contribution to Astarita (2015), like that of most other contributors to this EU workshop, focused entirely on the presumably more feasible task of strengthening national wealth taxes. Although many (e.g., Hubbard 2015) have discussed Piketty’s global wealth tax proposal in detail, the present paper focuses on the broader question of whether any ‘global’ tax is feasible.

For example: why have countries found it easier to reach agreements about international trade flows than about international investment flows? There is a World Trade Organization (WTO), but there is no corresponding World Tax Organization. For a recent look at some considerations bearing on such issues, see, e.g., Dreher and Lang (2016). As Bird and Gendron (2007) note, the trade liberalization required for WTO membership, which generally reduced tariff revenues, influenced domestic tax policy in some countries, leading them to adopt a value-added tax (VAT) in part as a substitute revenue source—though not always with full success (Baunsgaard and Keen 2010). But it is not as easy to draw a direct line between international agreements (e.g., the OECD treaty model) and corporate income tax systems in different countries (see Sect. 4).

For a useful compilation of wealth data for different countries, see Davies (2008); for more recent reviews of trends, see Alvaredo et al. (2017) and Lange et al. (2018). A useful review of the case for increased taxation of wealth and especially transfers of wealth may be found in Institute for Fiscal Studies (2010, 2011).

An alternative approach to global redistributive taxation suggested by Steinberg et al. (1978) was to finance development aid through a ‘shadow tax’ calculated by assessing countries in terms of their capacity (ability to pay) as measured by the estimated yield of a set of ‘standard’ taxes if levied on the relevant tax base at the average rates applied by all countries. Unlike most global redistributive schemes, this approach leaves countries free to collect what taxes they want to collect and involves no international intrusion on national fiscal sovereignty. Like all such proposals, however, it requires a degree of worldwide political commitment that does not exist.

Useful earlier reviews of global taxes from various perspectives include Steinberg et al. (1978) Cline (1979), Mendez (1992, 1997, 2001), Shome (1995), Frankman (1996), Baumert (1998), Paul and Wahlberg (2002), Wahlberg (2005), Herman (2013), and Carter (2015). Many of these reviews are by advocates of such taxes; for a notable exception see McMahon (2001).

See also UNDP (2012) on these (and other) ideas for ‘innovative’ ways to finance development aid.

See Frankman (1996). Several of these proposals are discussed later in this section; see also Bird (2015a).

For a good discussion of proposed fiscal approaches to this issue, see Brzoska (2004).

Extensive discussions of fiscal and other approaches to providing global public goods may be found in Kaul et al. (1999, 2002), Kaul and Conceicao (2006), Sandler (1997, 1998, 2002) and Barrett (2007) though not all the issues discussed in these sources are strictly global public goods, defined by Barrett (2007, p. 1) as outcomes making “people everywhere better off.”

See https://​phys.​org/​news/​2012-11-imf-chief-sms-tax-philippines.​html.

This argument has some merit: for an early example, see Roller and Waverman (2001) and for a recent thorough review, see Matheson and Petit (2017). However, many who make it appear to assume the internet is ‘free,’ which of course it is not and never has been. As Morozov (2013) notes, the common assumption that “the internet” has a clear and unarguable meaning inevitably leads to confusion.

Pogge (2008); see also Casal (2011), Steiner (2011) and Pogge (2011) for philosophical discussion of this idea.

In the 1980s over 160 countries ratified a UN convention on the Law of the Sea—although some, including the USA, specifically objected to provisions that may be interpreted to support such tax proposals.

As noted in Sect. 4, much the same can be said about the most visible outcome of the recent discussion of the international aspects of corporate income taxes—the acceptance of a uniform country-by-country reporting system.

The effects of these levies are only beginning to be understood: see, for instance, the analysis of the German bank levy by Buch et al. (2014) and the discussion of their incidence in Kogler (2015). To illustrate how complicated some of these taxes are, see the 252-page official manual issued at the time of the introduction of the UK bank levy (available at http://​www.​hmrc.​gov.​uk/​budget-updates/​autumn-tax/​bank-levy-manual.​pdf).

Unsurprisingly, given its role as a financial center, by far the strongest opposition to this approach came from the UK (Seely 2013).

Increased capital requirements were a central focus of the ‘Basel 3’ framework that emerged from extensive discussion of financial regulation under the auspices of the Bank for International Settlements (see www.​bis.​org/​bcbs/​basel3.​htm).

IMF (2010) preferred FAT to FTT largely because it would be easier to implement than (in effect) attempting to encompass the financial sector within the VAT and its effects would likely be less distorting than those of a cascading FTT; for a review of the extensive but inconclusive literature on including the financial sector within VAT, see James (2015). Vella (2012) also argued that there was a stronger case for a FAT than for an FTT on both equity and efficiency grounds; the opposite view is presented by Cannas et al (2014).

Bird (2015a) lists various FTTs in 38 countries in 2013. Additional countries—for example, Denmark, Ecuador, Singapore, Sweden—have had some variety of FTT in the past—some on securities, some on bank debits or transactions, some on currency transactions and some on automated payment systems. Many countries also levy special taxes on transfers of real property, which appear to be more politically acceptable (though economically less desirable) than imposing annual taxes on wealth held in this form (Bird and Slack 2004).

Before India decided to recall all large denomination currency notes in 2016, a similar idea was suggested in part to reduce corruption (ArthaKraniti 2013). As Coelho (2009) notes, however, even in Brazil, where his analysis suggests the tax at one time was perhaps the most productive in Latin America, it never yielded much more than 1.5% of GDP. For a similarly restrained view of the likely revenue potential of an FTT, see European Commission (2010).

The UK tax is a small (0.5%) levy paid by purchasers of UK shares; it is assessed and collected primarily through an electronic securities settlement system. It is considered to work well, although registered market makers and large banks are exempted. In contrast, a similar small tax imposed by Sweden in 1986 is usually considered to have been a failure because of its dramatic effect on trading volumes: most bond trading soon vanished and most (60%) of trading in Swedish shares moved abroad. Although this tax was abolished in 1991, a new ‘stability fee’ at a rate of 0.036% was imposed on the liabilities of banks and other credit institutions in 2009. This levy is like the bank levy imposed on bank balance sheets in 2011 by the UK—at a rate of 0.05%, increased in 2013 to 0.13% to ensure that despite reductions in the corporate tax rate, the taxes paid by banks would not be decreased (Seely 2013).

The 11 countries—all in the Eurozone—that have indicated they will sign on are Austria, Belgium, Estonia, France, Germany, Greece, Italy, Portugal, Slovak Republic, Slovenia, and Spain. Several other members of the Eurozone (e.g., Finland and Ireland) have at times indicated they too might sign on, but none of the non-Euro EU countries has publicly shown much interest in doing so.

See www.​eu.​eruopa.​eu/​taxation_​customs/​taxation-financial/​sector_​en#prop. See Heber and Steinberg (2014) for further discussion.

Unsurprisingly, perhaps, ECLAC (2011) suggested that revenues from both the FTT and the Tobin tax should be devoted to this purpose.

The (recently largely abolished) Brazilian tax—the IOF—is described in Coutinho (2012); for an analysis of the earlier Chilean experience, see Agosin and Ffrench-Davis (1996).

Consider, for example, the unusually well worked-out proposal in Taskforce (2010) for what is essentially a Tobin tax (the Global Solidarity Levy). This proposal assumes the tax would be administered by “an authority with formal oversight powers for licensed international settlement infrastructure and executive oversight of the proposed settlement institution’s tax raising functions in conformity with the legislation in the jurisdiction of residence or operation of the settlement institution” (Taskforce 2010, p. 28). This authority was to be developed by an inter-governmental tax commission, chaired by the IMF and convened by such worthies as the finance ministers and central banks of the G20 and the countries that host the larger international financial centers, the Board of Directors of the Bank of International Settlements, and representatives of the World Customs Organisation Council. The proceeds would then flow to a Global Solidarity Fund for distribution, as determined by ‘…its own decision-making board comprising a range of stakeholders including civil society and business sector [from developed and developing countries] together with the Consultative Forum [which would include NGOs and other stakeholders not on the board]’ (Taskforce 2010, p. 31). The vision may be inspiring, but its realization requires a much greater degree of convergence and agreement across countries than anyone has ever seen.

See http://​whqlibdoc.​who.​int/​publications/​2003/​9241591013.​pdf.

The UNITAID tax on air travel discussed later finances work on communicable diseases but has no direct effect in reducing their incidence (although it may have a minor effect by reducing carbon emissions).

For two very different analyses demonstrating the non-optimality of uniform ‘Pigovian’ taxes when marginal social costs, demand elasticities, or both, are heterogeneous, see Fleischer (2015) and Griffith et al. (2017). As Keen (1998), Cnossen (2006) and others have discussed, although some arguments may be made for an ad valorem tax, specific taxation of tobacco generally makes sense.

This number (which has sometimes been raised to 75%) appears to have been derived from an earlier standard of 67% suggested in World Bank (1999) which was itself apparently based largely on the argument that the best ‘target’ (yardstick) available was the tax level observed in countries considered to have the strongest tobacco control policies—that is, the highest taxes. Unsurprisingly, the benchmark suggested turned out to be close to that set in the European Union (EU), where Council Directive 2010/12/EW requires that by no later than the end of 2017 excise duties on cigarettes should be at least 60% of the weighted average retail price in all member states (and the actual average level in 2016 was 61.6%).

The impact of taxes on smuggling is, for example, discussed by Cnossen (2006) for the EU, by Irvine and Sims (2012) for Canada, and in more global terms by ICIJ (2009). For other reasons for varying rates, see the references cited in note 29 above.

In 2013, for example, France (where almost all the revenue is collected) charged 1 euro for a domestic economy seat, 10 euros for a higher-class seat, 6 euros for an international economy seat, and 40 euros for a higher-class international seat. In addition to France, Korea, and seven developing countries, most in Africa, now impose such a levy (and Norway allocates part of its tax on carbon emissions to UNITAID) (www.​unitaid.​eu/​en/​how/​innovative financing).

Kaufman et al. (2017) suggest that one way to reduce what they label the ‘democratic deficit’ between the aid people seem to want to support and the amount of actual aid might be for governments to ‘match’ charitable expenditure abroad in the same way as they match (usually through the tax system) charitable donations domestically. This is already done to a limited extent in some countries; in Canada, for example, foreign donations are given the same tax treatment as domestic donations when given to a charity to which the government has itself given funds. Another way to test the extent to which taxpayers were really willing to support foreign aid might be to permit them to earmark a small fraction of their income taxes for this purpose (Hirschman and Bird 1968). Even in the unusually generous Nordic countries, however, the linkage between how much (and what kind of) aid the public would support if they had the choice and what is done by their governments seems tenuous (Selbervik and Nygaaard 2006).

The EU included international air flights in its more general Emissions Trading System in January 2012 for flights to and from European airports, in effect introducing a ‘carbon tax’ like that discussed in Sect. 2.6 (though one that produced no revenue). Subsequently, however, the tax was first suspended for flights to and from non-member states and then, in October 2013, revised to apply only to the portion of such flights occurring within EU territory (with flights to and from many developing countries being exempted).

On fuel subsidies, see also OECD (2013b), World Bank (2016), and Coady et al. (2015).

The Kyoto Protocol committed countries to meet international binding targets for reducing GHG emissions, with heavier burdens being placed on developed countries as being primarily responsible for the increased GHG found in the atmosphere: see http://​unfccc.​int/​kyoto_​protocol/​items/​2830.​php. As usual with such international agreements, the targets set are mandatory only to the extent that countries voluntarily live up to them since there is no enforcement mechanism (Gillenwater 2013).

Some jurisdictions (e.g., California) auction allowances rather than dispensing them for free. Other policies not discussed here, such as energy efficiency standards, fuel taxes and subsidies, and support for renewable energy, may affect the ‘price’ of carbon emissions.

For descriptions of existing market-based mechanisms for reducing carbon emissions in the form of both emission trading regimes and taxes, see OECD (2013b), World Bank (2016), and Coady et al. (2015).

Analysis of the political distortions that arise from both the regulatory and the taxation approaches suggests that both may be distorting (Masciandro and Passarelli 2012). Determining which is best in any political setting requires knowing much more than anyone can know. Life is difficult to model.

The arguments against uniform Pigovian taxes mentioned in note 29 above are much weaker in this case. Indeed, as Fleischer (2015) notes, taxing carbon emissions is the ideal ‘textbook’ case for a uniform tax.

Credible estimates of the ‘correct’ level at which to set a carbon tax are difficult, but a 2010 estimate for the USA was $35 per ton of CO₂ (IAWG 2013), considerably higher than most existing carbon taxes. In addition, many countries have substantial taxes on motor fuels, which also reduce emissions (Miller and Vella 2013).

For example, Britain adopted the Clean Air Act of 1956, which ended the choking yellow smog that had for decades blanketed London only after the great smog of 1952 killed at least 4000 people and made at least 100,000 ill. (The figures cited come from an official source (http://​www.​metoffice.​gov.​uk/​education/​teens/​case-studies/​great-smog; other sources give higher estimates.).

See www.​upu.​int/​finances.​hmtl and ITU (2014).

See Mehrotra (2013) for a detailed account of the move from one model to the other in the US tax system.

See Mendez (1997) as well as sources cited in http://​www.​globalpolicy.​org/​un-finance.​html. See also Barrett (2007) for an interesting view of UN finance.

The other 20% is kept by the collecting country. The EU also collects a progressive personal income tax on its own employees, but this is of course simply a transfer within its budget.

Owing to the convoluted way the so-called “UK correction” is calculated as well as a few adjustments made for several other several countries “to mitigate perceived imbalances” as EU (2014, p. 193) puts it, the EU’s fiscal framework is more complex than this capsule description indicates.

There are some differences between VAT bases in EU countries and even to some extent in how GNP (GNI) is estimated, but both must correspond closely to agreed standards.

For examples, see the essays in Martin et al. (2009), Yun-Casalilla et al. (2012), Cardoso and Lains (2010), and Soifer (2015), as well as such detailed country studies as Daunton (2001) on the UK, Mehrotra (2013) on the USA, and Heaman (2017) on Canada. Building a ‘fiscal state’ has proven an especially delicate and difficult task in ‘fragmented’ countries—those with geographically concentrated ethnic–linguistic differences—as discussed in Bird and Ebel (2007) and Vaillancourt and Bird (2016).

Though what is considered ‘fair’ by many may more often resemble what Sheffrin (2013) calls ‘folk justice’ than the welfare-maximizing optimum of the economist or the egalitarian utopia envisaged by some.

This assertion may perhaps be acceptable in the present context but of course, like many of the complex issues touched on in this paper, it is not easily documented and has long been discussed from different perspectives by philosophers (Wellman 2000; Coons 2001), psychologists (Ashmore et al. 2001), and economists and political scientists (Gradstein and Konrad 2006; United Nations 2006), among others.

For a telling example, see Bartik (2017) on the continued waste of billions in business subsidies in the US today.

The sad story of death taxes in the USA (Graetz and Shapiro 2005) makes the point nicely; a similar, less dramatically told, story happened earlier in Canada (Bird 1978).

An extended version of some of the argument in this section may be found in Bird (2016).

For inevitably approximate attempts to estimate the possible revenue losses, see Crivelli et al. (2015) and Cobham and Jansky (2017). The considerable country-to-country variation noted by Cobham and Jansky (2017) likely reflects not only different economic circumstances but also, as Genschel and Seelkopf (2016) note, differences in size and in quality of governance.

http://​www.​oecd.​org/​tax/​beps/​country-by-country-reporting.​htm. The Extractive Industries Transparency Initiative (Sect. 2.3 above) is an example of a recent sectoral attempt along these lines. As Forstater (2017) discusses, it is not clear that the information to be collected is either necessary or sufficient, but it is clear that much more effort is needed to put it to good use.

Over 100 countries have agreed to adopt a common reporting standard, including the automated exchange of tax information, as well as to modify their bilateral tax treaties to implement some of the BEPS reforms (see http://​www.​oecd.​org/​tax/​treaties/​multilateral-convention-to-implement-tax-treaty-related-measures-to-prevent-beps.​htm). For an optimistic view of the prospects that future international tax information exchange and cooperation—driven in part by USA’s independent but related FATCA (Foreign Account Tax Compliance Act)—will improve outcomes in this respect, see Grinberg (2013). For more restrained appraisals from different perspectives, see Eccleston (2012), Shaviro (2014), Avi-Yonath and Xu (2017), and Forstater (2017), Cockfield (2017).

For an elaboration of this approach at the sub-national level, see Bird and Slack (2014).

This concept comes close to Krasner’s classic definition of an international regime (1982, p. 186) as ‘implicit or explicit principles, norms, rules, and decision-making procedures around which actors’ expectations converge in a given area of international relations.’ For discussions of the global governance literature as it applies to international tax issues, see Rixen (2008), Eccleston (2012), Genschel and Rixen (2015), and Dreher and Lang (2016).

For interesting perspectives on how some outside the process view the outcome of the BEPS discussions, see Durst (2014, 2015) and Myszkowski (2016); Civelli et al. 2015) provide a rare empirical estimate of the impact on developing countries; see also the more recent extended appraisal by Cobham and Jansky (2017). The OECD, long seen as a rich country club, may not be the best forum in which to pursue these issues, but this topic is beyond the present discussion.

For example, the Forum of Tax Administrators (FTA), a panel of national tax administrators established in 2002 by the OECD’s Committee on Fiscal Affairs to promote dialogue between administrations; the Leeds Castle Group, a group of tax administrators from a number of major countries, including some non-OECD countries like China and India, who meet regularly to discuss mutual compliance problems; and the Joint International Tax Shelter Center established by the USA, UK, Canada, and Australia to develop and share information on abusive tax avoidance.

Harmonized does not mean uniform: as with the tobacco tax proposals discussed earlier, given the different circumstances facing different countries as well as their different regulatory regimes, if the aim is to bring about similar results in dissimilar contexts often different measures (e.g., different tax rates) are required.